POWER ELECTRONIC CONVERTER TOPOLOGIES ...

9 downloads 0 Views 3MB Size Report
including the electric cars made by TESLA. General Motors Company has an induction motor design for Chevy Spark. The Hyundai Sonata uses surface mount.

POWER ELECTRONIC CONVERTER TOPOLOGIES USED IN ELECTRIC VEHICLES By Eng. Shady Mamdouh Sadek

A Report Submitted to the Faculty of Engineering at Ain Shams University In Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy IN ELECTRICAL POWER AND MACHINES ENGINEERING

FACULTY OF ENGINEERING, AIN SHAMS UNIVERSITY CAIRO, EGYPT September, 2016

Table of Contents Abstract Introduction History of Electric Vehicles Types of Electric Vehicles EV Drives Versus Industrial Drives EV Technology EV Propulsion System 6.1. EV Motors 6.1.1. DC Motors 6.1.2. Induction Motors 6.1.3. Permanent Magnet Synchronous Motors 6.1.4. Brushless DC Motors 6.1.5. Switched Reluctance Motors 6.1.6. Evaluation of EV Motors 6.2. Electronic Controllers 6.3. Energy Storage Systems 7. Power Electronics in EVs 7.1. Architectures of PE in EVs 7.2. Hybrid Energy System 7.3. DC/DC Converters 7.3.1. DC/DC Converters used for supplying LV loads 7.3.1.1. PWM DC/DC Converter 7.3.1.2. Quasi-Resonant Converter 7.3.1.3. Zero Voltage Switching Multi Resonant Converter 7.3.1.4. Const. Freq. Quasi Resonant & Multi Resonant Converter 7.3.1.5. Nonlinear Switch Converter 7.3.1.6. Resonant Converters 7.3.1.7. Soft Switching PWM Converter 7.3.1.8. ZVS/ZCS PWM Converters 7.3.1.9. Selection of Converters 7.3.2. DC/DC Converters used for boosting the Battery Voltage 7.3.2.1. Non isolated Bidirectional H-bridge Converter 7.3.2.2. Isolated Bidirectional Full Bridge Converter 7.3.2.3. Isolated Bidirectional Dual H-Bridge Converter 7.3.2.4. Interleaved DC/DC Converters 7.3.3. DC/DC Converters used for DC Motor Drives 7.3.3.1. Two Quadrant ZVS MR Converter 7.3.3.2. Two Quadrant ZVT Converter 7.3.3.3. Two Quadrant ZCT Converter 7.4. DC/AC Converters 7.4.1. H-bridge Technology with Single DC Link 7.4.2. Soft Switched Inverters 7.4.2.1. Resonant DC-Link Inverter 7.4.2.2. Auxiliary Resonant Commutated Pole Inverter 7.4.2.3. Auxiliary Resonant Snubber Inverter 7.4.2.4. Zero Current Transition Inverter 7.4.3. Multi-Level Inverters 7.4.3.1. Cascaded H-bridge Multi Level Inverter 1. 2. 3. 4. 5. 6.

i

1 2 2 3 4 5 7 8 8 8 9 9 10 11 12 12 14 15 21 22 23 23 24 25 26 26 27 28 29 30 31 32 34 36 39 42 43 43 44 45 46 47 47 49 49 51 52 52

7.4.3.2. Back to Back Clamped Multi Level Inverter 7.4.3.3. T-Type Five Level Inverter 7.4.4. Z-Source Inverter 7.5. Battery Chargers 7.5.1. Power Factor Correction Stage 7.5.1.1. Decreasing Impact on Grid 7.5.1.2. Decreasing Impact on Switches 7.5.2. Integrated Chargers 7.5.3. Inductive Chargers 7.5.4. Wireless Chargers 7.6. Challenges & Problems facing EV Power Electronics 7.6.1. Power Density & Specific Power 7.6.2. Electromagnetic Interference 7.6.3. Applications of SiC & GaN Devices 7.6.4. Thermal Management 8. Conclusions 9. References

ii

54 55 58 60 62 63 63 64 65 65 67 67 68 68 70 71 75

Power Electronic Converter Topologies used in Electric Vehicles

Abstract: With the increasing interests on energy efficiency, energy cost, and environmental protection, the development of electric vehicles (EVs) technology has been obvious nowadays. Based on the air pollution regulations in the USA and Europe as well as a lot of countries in the world, the fossil-fueled vehicles have been targeted as the major source of emissions that create air pollution leading to the global warming crisis. The oil resources in the earth are limited and the new discoveries of it are at a slower pace than the increase in demand especially with the increase in the world population so that the need for alternatives is becoming crucial. The technologies involved in EVs are diversified and include electrical and electronics engineering, mechanical engineering, automotive engineering, and chemical engineering. EVs depend on which called electric propulsion, in which an electric motor is used to drive the vehicle instead of the internal combustion engine (ICE) and the energy sources are batteries, fuel cells, or capacitors instead of gasoline or diesel fuel in the conventional ICE vehicles. Thus, power electronics technology plays an important role in the electrical propulsion system in order to efficiently drive the electric motor of the vehicle and control the power converters and the associated electronic circuits. An overview of the current status of power electronic drives for EVs and recent research trends in EV motors, power converters, and energy storage systems will be discussed. Challenges associated with designing, controlling, and operating the power electronic converters used in EVs will be also discussed showing the latest achievements in this field. Challenges and problems associated with power electronic converters will be discussed also.

-1-

1. Introduction: The electric and hybrid vehicle market has been growing over the years and continues to grow today. Influenced by both government regulations and consumer demand, auto manufacturers have continued to pursue technologies to improve efficiency and fuel economy. As the market progresses, continued research and development is needed to enable large-scale market penetration of electric and hybrid vehicles in the future. In particular, there are three main areas in which current research aims to improve the vehicle: electric motors, power electronics, and energy storage. This report describes the evolution of these technologies and the road map for future development and implementation in electric and hybrid vehicles. The electric vehicle can trace its inception as far back as the early 20th century; however, was quickly overmatched by the internal combustion engine. The advantage that the ICE had was its energy storage capacity, i.e. it was capable of providing longer ranges at a lower fuel cost. However, by the end of the 20th century, technology improvements in electric machines, power electronics, and energy storage sparked increased efforts in electric vehicle development. Today, led by major car companies around the world, electric and hybrid vehicles have a share of the marketplace. New technologies will continue to revolutionize the industry and lead to large-scale adaptation of these cars. The objective of this report is to review the state-of-the-art technologies in electric motor, power electronics, and energy storage for automotive applications. The current trends in the technology are presented, as well as any other performance requirements set for the future. The metrics and benchmarking of the technologies are used to describe the current and potential future states of the technologies.

2. History of Electric Vehicles: In 1801, the steam-powered carriage was built, opening the era of horseless transportation. After thirty years of noise and dirtiness due to steam engines, the first battery-powered EV was built in 1834. Over fifty years later, the first gasolinepowered ICE vehicle was built in 1885. So, EVs are not new and already over 170 years old. With the drastic improvement in combustion engine technology, ICE vehicles showed much better performance and EVs were out of use from the 1930s to the 1950s. Interests in EVs started at the outbreak of energy crisis and oil shortage in the 1970s. The actual revival of EVs is due to the ever increasing concerns on energy conservation and environmental protection throughout the world as summarized below:

-2-

•EVs offer high overall energy efficiency over the ICE vehicles. Moreover, EVs can perform efficient braking by converting the kinetic energy back to electricity, virtually boosting up the energy efficiency by at most 25%. Moreover, while the maximum efficiency of the ICE vehicle can be 30-35 %, electric propulsion system drawing the power from battery can operate with a peak efficiency of around 90% [1]. •EVs allow energy diversification. Electricity can be generated not only from thermal power using fossil fuels; coal, natural gas and oil, but also from hydropower, wind power, geothermal power, nuclear power, tidal power, wave power, solar power, chemical power and biomass power. •EVs enable load equalization of power system. By recharging EVs at night, the power generation facilities can be effectively utilized, contributing to energy saving and stabilization of power cost. •EVs show zero exhaust emissions. •EVs operate quietly and almost vibration-free, whereas ICE vehicles are inherently noisy and with sensible vibration. Thus, EVs are welcomed by drivers and appreciated by local residents. Nowadays, many governments actively promote the use of EVs by providing facilities such as financial subsidies and tax reduction, as well as enforcing regulations such as zero-emission zones and ultralow-emission vehicles. Apart from numerous advantages of EVs, they have some disadvantages like, large battery charging time, lower flexibility and limited dynamic performance. An important limitation is its limited operating range per cycle of battery charge which acts as bottleneck of the technology. To improve the dynamic performance and all electric range, advanced electric vehicles such as hybrid electric vehicles (HEV), plug in HEVs and fuel cell vehicles have been proposed. These advanced vehicles are not only capable to compete against the convention ICEVs in performance but are also able to give higher fuel economy and low emissions [1].

3. Types of Electric Vehicles: The conventional ICE vehicle employs a combustion engine for propulsion. Its energy source is gasoline or diesel fuel. In contrast, the EV employs an electric motor and the corresponding energy sources are batteries, fuel cells, capacitors and/or flywheels. However the presently achievable specific energy of capacitors and flywheels precludes them from being the sole energy sources for EVs. The key difference between the ICEV and EV is the device for propulsion (combustion engine versus electric motor). Currently, there are three categories of EV systems [2]:

-3-

•The battery EV (BEV) utilizes batteries as the sole energy source, and electric motors as the propulsion device. This BEV has been commercially available though not yet under mass production, and is mainly designed for commuter operation with the driving range of about 100 km per charge. Battery-powered electric vehicles were one of the solutions proposed to tackle the energy crisis and global warming. However, the high initial cost, short driving range, long charging (refueling) time, and reduced passenger and cargo space have proved the limitation of battery-powered EVs. •The hybrid EV (HEV) incorporates both of the combustion engine and electric motor as the propulsion device. It adopts gasoline or diesel fuel as the main energy source, and utilizes batteries as the auxiliary energy source. The HEV can offer the same driving range as the ICEV (over 500 km per refuel), while produces much lower emissions than those of the ICEV. This HEV has been commercially available and under mass production. •The fuel cell EV (FCEV) adopts fuel cells as the main energy source, and the electric motor as the propulsion device. Since fuel cells cannot accept regenerative energy, batteries are generally adopted as the auxiliary energy source. Being fuelled by hydrogen or methanol, the FCEV can provide a driving range comparable with the ICEV. Because of its high initial cost, this FCEV is not yet commercially available [2]. A fuel cell uses hydrogen and oxygen to produce electricity through a chemical reaction. Fuel cell offers low emission and higher efficiency comparing to ICE. Today's fuel cells have achieved the power density suitable for vehicle applications. The major challenges for fuel cell vehicles are cost and fueling infrastructure. Various attempts are being made to reduce the manufacturing cost of a fuel cell. To address infrastructure issues, significant R&D effort is made in in-vehicle reformers which generates hydrogen-rich gas by reforming other fuels such as gasoline, methanol, or natural gas [3]. •The plug-in hybrid ( PHEVs) refer to vehicles that can use, independently or not, fuel and electricity, both of them rechargeable from external sources. PHEVs can be seen as an intermediate technology between BEVs and HEVs. It can be considered as either a BEV supplemented with an internal combustion engine to increase the driving range or as a conventional HEV where the all-electric range is extended as a result of larger battery packs that can be recharged from the grid.

-4-

4. Electric Vehicle Drives versus Industrial Drives: EV drives may be considered similar to industrial drives according to some engineers and researchers. However, EV drives usually require frequent start/stop, high rate of acceleration/deceleration, high-torque low-speed hill climbing, low-torque high-speed cruising and very wide-speed range of operation, whereas industrial drives are generally optimized at rated conditions. Thus, EV drives are so unique that they form an individual class. Their major differences in load requirement, performance specification and operating environment are summarized as follows [2]: •EV drives need to offer four to five times the maximum torque for temporary acceleration and hill-climbing, while industrial drives generally offer twice the maximum torque for overload operation. •EV drives need to achieve four to five times the base speed for highway cruising, while industrial drives generally achieve up to twice the base speed for constantpower operation. •EV drives should be designed according to the vehicle driving profiles and drivers’ habits, while industrial drives are usually based on a particular working mode. •EV drives demand both high power density and good efficiency map (high efficiency over wide speed and torque ranges) for the reduction of total vehicle weight and the extension of driving range, while industrial drives generally need a compromise between power density, efficiency and cost with the efficiency optimized at a particular operating point. •EV drives desire high controllability, high steady-state accuracy and good dynamic performance for multiple-motor coordination, while industrial drives seldom desire such coordination. •EV drives need to be installed in mobile vehicles with harsh operating conditions such as high temperature, bad weather, and frequent vibration, while industrial drives are generally located in fixed places. These temperature conditions exert great thermal stress on components such as power semiconductor modules and electrolytic capacitors.

5. Electric Vehicle Technology: The technologies involved in EVs are diversified and include electrical, electronic engineering, mechanical, automotive engineering, and chemical engineering. Specialists in these disciplines of engineering must work together in the main areas that must be integrated: body design, batteries, electric propulsion, and intelligent energy management. Concerns in this work will be directed towards electric part in the vehicles and its associated issues. -5-

Fig. 1 shows the general electrical configuration of EVs, including the BEV, HEV and FCEV. It consists of three major subsystems — electric propulsion, energy source, and auxiliary. The electric propulsion subsystem comprises the electronic controller, power converter, electric motor, mechanical transmission, and driving wheels. The energy source subsystem involves the energy source, energy management unit, and energy refueling unit. The auxiliary subsystem consists of the power steering unit, temperature control unit, and auxiliary power supply. Based on the control inputs from the brake and accelerator pedals, the electronic controller provides proper control signals to switch on or off the power devices of the power converter which functions to regulate power flow between the electric motor and energy source. The backward power flow is due to regenerative braking of the EV and this regenerative energy can be stored. The most available EV batteries as well as capacitors and flywheels accept regenerative energy. The energy management unit cooperates with the electronic controller to control regenerative braking. It also works with the energy refueling unit to control refueling and to monitor usability of the energy source. The auxiliary power supply provides the necessary power with different voltage levels for all EV auxiliaries, especially the temperature control and power steering units [2].

Fig. 1 "General Electrical Configuration of EVs"

-6-

6. Electric Propulsion System: Electric propulsion is to interface electric supply with vehicle wheels, transferring energy in either direction as required, with high efficiency, under control of the driver at all times. From the functional point of view, an electric propulsion system can be divided into two parts—electrical and mechanical. The electrical part includes the motor, power converter, and electronic controller. On the other hand, the mechanical part consists of the transmission device and wheels. Sometimes, the transmission device is optional. The boundary between electrical and mechanical parts is the airgap of the motor, where electromechanical energy conversion is taking place. Electric propulsion, a major power electronics area, plays a very important role in EVs. Sometimes, it is described as the heart of EVs [4]. Fig. 2 illustrates the functional block diagram of a typical EV propulsion system where the arrow-headed thick and thin lines represent the power and signal flows, respectively. Due to the availability of regenerative braking, the power flow is reversible. Depending on the motor control strategy, driver’s command, and data obtained from the EMS, the electronic controller provides proper control signals to the power converter. These signals are amplified via a driving circuitry to switch proper power devices. Thus, the power converter regulates power flow between batteries and the motor during motoring and regenerative braking. Finally, the motor interfaces with wheels via the transmission device.

Fig.2 "Functional block diagram of a typical electric propulsion system"

-7-

6.1.Electric Vehicle Motors: To keep up with the more advanced design requirements and fast-changing motor topologies, the design of EV propulsion motors turns to computer-aided design (CAD). The finite element method (FEM) outranks other numerical methods because of its applicability in electromagnetic, force, and thermal analyses. The development of motors is followed by the technologies of high-energy permanent magnets (PMs), sophisticated motor topologies, and powerful CAD techniques. The only drawback is the initial cost which is reflected by the price of motors. Apart from ferrites, alnico, and samarium–cobalt (Sm–Co), neodymium–iron–boron (Nd–Fe–B) PMs have been introduced. Because of their highest remanence and coercivity as well as reasonable low cost. Using these magnets, a number of new motor topologies with high power density and high efficiency have recently been developed [2]. 6.1.1. DC Motors: Traditional DC motors, have been used in EV propulsion. Their control principle is simple. The torque-speed characteristics of the DC motor drive also suit the traction requirement for being used in EVs. By replacing the field winding with high-energy PMs, PM DC motors permit a considerable reduction in stator diameter. Owing to the low permeability of PMs, armature reaction is usually reduced and commutation is improved. However, the principal problem of DC motors arises from their commutators and brushes which make them less reliable and unsuitable for maintenance-free operation. Nevertheless, because of mature technology and simple control, DC motors have ever been prominent in electric propulsion. As shown in Table1, all types of DC motors, including series, shunt, separately excited, and PM excited, had ever been adopted by EVs [2]. 6.1.2. Induction Motors: Recent technological developments have pushed AC motors to a new era, leading to definite advantages over DC motors: higher efficiency, higher power density, lower cost, more reliable, and almost maintenance free. As high reliability and maintenancefree operation are prime considerations in EV propulsion, AC induction motors are becoming attractive. However, conventional control of induction motors such as variable-voltage variable-frequency (VVVF) cannot provide the desired performance of EVs. One major reason is due to the nonlinearity of its dynamic model with coupling between direct and quadrature axes. With the advent of the microcomputer era, the principle of field-oriented control (FOC) of induction motors is becoming accepted [4].

-8-

6.1.3. Permanent Magnet Synchronous Motors By replacing the field winding with high-energy PMs, PM synchronous motors can eliminate conventional brushes, slip rings, and field copper losses. As these motors are essentially traditional AC synchronous motors with sinusoidal-distributed windings, they can run from a sinusoidal or PWM supply without electronic commutation. When PMs are mounted on the rotor surface, they behave as non-salient synchronous motors because the permeability of PMs is similar to that of air. On the other hand, by burying PMs inside the magnetic circuit of the rotor, the saliency causes an additional reluctance torque. These motors are generally simple and cheap, but with relatively low output power. Similar to induction motors, those PMS motors usually employ FOC for high-performance applications. Because of their inherent high power density and high efficiency, they have been accepted to have great potential to compete with induction motors for EV propulsion. In recent years, the corresponding efficiency has been further increased by applying self-tuning control to achieve optimal efficiency [2], [4].

6.1.4. Brushless DC Motors: By inverting the stator and rotor of PM DC motors, rectangular-fed ac motors, socalled PM brushless DC motors, are generated. The most obvious advantage of these motors is the removal of brushes, leading to the elimination of many problems associated with brushes. Another advantage is the ability to produce a larger torque at the same peak current and voltage. Moreover, the brushless configuration allows more cross-sectional area available for the armature winding, thus facilitating the conduction of heat through the frame and, hence, increasing the electric loading and power density. Although their configurations are very similar to those of PM synchronous motors, there is a distinct difference in that PM brushless DC motors are fed by rectangular AC wave, while PM synchronous motors are fed by sinusoidal or PWM AC wave. Different from PMS motors, these motors generally operate with shaft position sensors. Because of their inherent high power density and high efficiency, these motors have promising applications for EV propulsion. In recent years, the corresponding developments have been very active. The constant-power operating range can be extended greatly. By adopting self-tuning control, this motor can further achieve optimal-efficiency constant-power operation [2], [4].

-9-

6.1.5. Switched Reluctance Motors: Switched reluctance (SR) motors, though the principle of which has been known for over a century, have seen some interest in recent years. Similar to PM brushless DC motors, they usually require shaft position sensors. However, switched reluctance motors cannot attain the efficiency or power density of PM AC motors. They have the definite advantages of simple construction and low manufacturing cost. Although they possess the simplicity in construction, it does not imply any simplicity of their design and control. Because of the heavy saturation of pole tips and the fringe effect of poles and slots, their design and control are difficult. Also, they usually have acoustic noise problems. In recent years, research activities on SR motors have been quite limited. Nevertheless, an optimum design approach to SR motors has been developed, which employs finite element analysis to minimize the total motor losses while taking into account the constraints of pole arc, height, and maximum flux density. Also, fuzzy sliding mode control has been developed for those SR motors to handle the motor nonlinearities [2]. A typical classification of EV propulsion motors is illustrated in Fig. 3 where the motor types encircled by round-corner boxes have ever been used in EVs motor types have been accepted for modern EVs. As given in Table 1, both the GM Impact 4 and Nissan FEV employ the induction motor, while both the BMW E1/E2 and U2001 use the PM brushless DC motor. On the other hand, the other motor types are also employed in EVs such as the PM synchronous motor in the Ford/GE ETX-II, the switched reluctance motor in the Chloride Lucas, the DC series motor in the Daihatsu Hijet, the DC shunt motor in the Mazda Bongo, the DC separately excited motor in the Fiat 900E/E2, and the PM DC motor in the Suzuki Senior Tricycle [2], [4]. Among all passenger EVs/HEVs, a very small number of models use induction motor, including the electric cars made by TESLA. General Motors Company has an induction motor design for Chevy Spark. The Hyundai Sonata uses surface mount permanent magnet (SPM) machine. Almost all the other major car companies use interior permanent magnet (IPM) machines for EVs and HEVs. Other machine types may have been studied in research, but have not been used in production. In some other applications, such as electric bike or off-highway vehicle, machine topologies such as switched reluctance machine are also used [5].

Fig.3 "Classification of electric motor drives for EV and HEV applications" - 10 -

Table 1 "Applications of EV motors"

6.1.6. Evaluation of Electric Vehicle Motors: Comparison between various motor drives on certain parameters has been made. The parameters are the power density, efficiency, controllability, reliability of the drive, maturity of the drive technology and also comparison with respect to the cost factor. In terms of efficiency, the most efficient motor drives are the permanent magnet brushless motor. Next come the induction and the switched reluctance motor drives which have almost identical efficiency and amongst all the motors being used in the electric vehicle drive system the least efficient are the DC motors. In terms of the maturity of the technology for being used in propulsion system, induction motor and DC motor drives score the highest and these two technologies are slightly more mature than that of permanent magnet brushless and switched reluctance motors. One major reason being that a lot of research has already been done for these two as they are quiet old technologies [6]. Another factor that we consider while comparing different motor drives is the reliability of the drive i.e. maintenance and breakdowns should be at a minimum. In terms of reliability, the most reliable are the induction motor drives and switched reluctance drives, followed by permanent magnet brushless motor drives. The least reliable amongst all the different drives are the DC motor drives. When it comes to the power density, then permanent magnet brushless motors come out at the top followed by both induction and switched reluctance motors. Here once again the DC motor drives have the lowest power density. One of the most important characteristic that we consider when choosing a motor drive for electric vehicle propulsion is the cost factor. This is most important because to make anything commercially viable, it's cost has to be kept at a minimum. In terms of cost factor, the best to be used are the induction motors followed by the DC and the switched reluctance motors. Surprisingly permanent magnet brushless motors score the least in cost factors when compared with all the others [6].

- 11 -

6.2.Electronic Controller: Conventional linear control such as PID can no longer satisfy the stringent requirements placed on high-performance EVs. In recent years, many modern control strategies such as model-referencing adaptive control (MRAC), self-tuning control (STC), variable structure control (VSC), fuzzy control, and neural network control (NNC) have been proposed. Both MRAC and STC have been successfully applied to EV propulsion. Using sliding mode, VSC has also been applied to motor drives. By employing emerging technologies of fuzzy logic and neural networks to realize the concept of intelligent controllers, fuzzy control and NNC have promising applications to EV propulsion. In order to implement the aforementioned modern control strategies, powerful microelectronic devices are necessary. Modern microelectronic devices include microprocessors, microcontrollers, digital signal processors (DSPs), and transputers. Microprocessors are usually used to recognize the milestone of the development of microelectronics such as the 8086, 80186, 80286, 80386, 80486, and Pentium. Unlike microprocessors, which are the CPU of microcomputer systems, microcontrollers include all resources to serve as standalone single-chip controllers. Thus, microcontroller-based EV propulsion systems possess definite advantages of minimum hardware. The state-of-the-art microcontrollers are the 8096, 80196, and 80960. DSPs such as the TMS32030, TMS32040, and i860 possess the capability of high-speed floating-point computation which is very useful to implement sophisticated control algorithms for highperformance EV propulsion systems. Transputers such as the T400, T800, and T9000 are particularly designed for parallel processing applications. By employing multiple chips of transputers, any sophisticated control algorithms can be implemented [4]. 6.3.Energy Storage Systems: Perhaps the most important and currently most expensive component relative to the feasibility of hybrid electric and especially electric-only vehicles is the energy storage system namely the battery. In EVs, the battery is the most expensive component and it is the fundamental piece that determines most user-interested functions of the vehicle, including range, acceleration, and cost. Battery technology can still be considered in its infancy, so it is expected that it will continue to mature and reduce in price and size and increase in capacity. All of these effects are crucial to the continued growth of the electric and hybrid vehicle industry [4]. The two factors that constrain the number of Whrs in the battery of an EV are the cost per unit of energy and the energy density. Continued reduction in the price per unit of energy is necessary for the penetration of EVs in the marketplace. As mentioned, the energy density of the battery is equally important for the growth of EVs. A metric of importance is the specific energy, or Whr/kg, of the battery. It is critical because the weight and volume in a vehicle are invaluable. A reduction in weight is desirable because less energy would be required to propel the vehicle, thus extending range. - 12 -

Volume is equally important because there is a very finite amount of space within a vehicle for the battery, so being able to put more energy in the same amount of space valued. Currently many battery types exist and various types are used in vehicles to optimize the total cost and range. The two main types of batteries that have been implemented in EVs and HEVs are of the Nickel-Metal Hydride (NiMH) family or of the Lithium Ion (Li-Ion) family. These vary both in chemical and electrical properties, and some of their most important qualities are presented in Table 2 along with the specifications of lead acid batteries [5]. Table 2"Comparison of Battery Cell Types"

As can be seen from this data, Li-Ion batteries are better than NiMH batteries in terms of energy density, specific energy and specific power. Specific power (W/kg) is important in terms of vehicle performance, such as acceleration, and also is a factor in the amount of energy that can be captured from regeneration braking. Table 3 shows a summary of the battery specifications of recent vehicles. Other details important to manufacturers include the battery cooling mechanism, voltage level, and range. Manufacturers must optimize their battery to not only meet the consumer demands (range, lifetime, charging time, cost, etc.) but also to work most efficiently with the entire vehicle system, such as proper electric machine development and control of solid state power converters. These advancements have led to a diverse implementation of batteries in EV/HEV systems. Continued reductions in the cost of batteries along with the maturation of different battery chemistries and technologies that can improve performance are crucial to the growth of the industry. As Li-Ion improvements saturated, the field will turn to other solutions, perhaps Li-Air or Li-Metal, to continue improvements. Improvements in energy storage are not only important for EVs, but will also enable other green technologies, contributing to a more sustainable environment [5].

- 13 -

Table 3 "Comparison of Vehicle Battery Types"

7. Power Electronics in Electric Vehicles: In the past decades, power device technology has made a tremendous progress. These power devices have grown in power rating and performance by an evolutionary process. Among existing power devices, including the thyristor, gate turn-off thyristor (GTO), power bipolar-junction transistor (BJT), power metal-oxide field-effect transistor (MOSFET), insulated-gate bipolar transistor (IGBT), static-induction transistor (SIT), static-induction thyristor (SITH), MOS-controlled thyristor (MCT), and MOS turn-off thyristor (MTO), the IGBT is almost exclusively used for modern EVs. Nevertheless, the power MOSFET has also been accepted for those low-voltage low-power EVs. The evolution of power converter topologies normally follows that of power devices, aiming to achieve high power density, high efficiency, high controllability, and high reliability. The power converter topologies depend on the motors to be driven. The selection criteria of motor drives, including the motors and their power converters, for EVs can be divided into the mandatory requirements and the preferable requirements. The mandatory ones are that the motor drive can offer the torque-speed requirements of the EV driving profile without involving variable gearing or gearbox, and the motor drive can provide the capability of bidirectional power flow to recover the regenerative braking energy. In general, the DC, AC or SR motor drives can offer the desired torque-speed requirements under proper motor design. The preferable requirements of EV motor drives are low cost, high efficiency, high power density, high controllability and maintenance-free operation [2].

- 14 -

The demand of electrical power in vehicular application is growing rapidly as the mechanical components are being replaced by the electrical and electronic components . It is expected that the power demand in electric vehicular system (EVS) could reach 2 to 3 times of the current demand. In order to accomplish the growing power demands of EVs in desired manner, integration of power electronic components with electrical and mechanical loads of vehicle becomes crucial. The integration of power electronic converter (PEC) not only improves the overall performance and fuel economy but also reduces the emission as well as weight and size of the vehicle. For electric vehicular system, there is a need of highly reliable, flexible and fault tolerant electrical power processing system on board to deliver high quality of power based on vehicle demands. At present, this responsibility have been taken care by the available PEC that includes DC/DC converters, rectifiers (DC/AC), inverters (AC/DC) and battery charger composed of AC/DC and or DC/DC converters . Individually or combination of these converters can be taken to serve the purpose; however, operation of each PEC is entirely different from the other. PECs perform some of the critical tasks like ON/OFF switching of various loads; power conditioning and voltage/current modulation to create compatibility among the energy source system (ESS), traction motors and auxiliary loads. PECs not only serve the purpose of converting electrical power from one form to another (DC/DC, DC/AC and, AC/DC) but also help to step up or step down the system voltage level. In the last decade, significant advancement in converter topologies dealing with battery charger, voltage source inverter for motor drives and DC/DC conversion has been achieved. Regardless of this topological advancement, still conventional PEC topologies are being used in modern electrified vehicles [1]. 7.1.Architectures of Power Electronics in EVs: To emphasize the role of PECs, the architecture of different electrified vehicles are reviewed and compared. Contribution of PECs at different parts of vehicle operation has been well studied. The current status and future challenges of PECs and their components are briefly discussed. Amongst EVs/HEVs/PHEVs/FCVs, each member has its own capability and limitation in terms of emission rate, performance, fuel economy, size, weight, cost, safety and comfort. Among these electrified vehicle systems, hybrid electric vehicle (HEV) is the only option that has potential to compete against the ICE vehicles in terms of performance for all driving profiles and offers advantages like extended electrical range of operation, good fuel economy, higher efficiency, sufficient onboard power and better dynamic response.

- 15 -

However, integration of automobile technology with electrical technology adds complexity in controls, and makes the HEV vehicle system comparatively bulky and costly. At the present state, HEVs are seen to be promising technology which is growing rapidly and capturing significant market space at fast rate. Plug-in HEVs are in the stage of commercialization and needs good energy policies and infrastructure for charging stations in near future. Fuel cell vehicles are considered as futuristic transportation system which will be powered by hydrogen. FCV needs to overcome the certain technical hurdle for their successful commercialization in the long run. The detailed architecture of above mentioned vehicles are illustrated in Fig. 4(a-h). The comparative analyses and corresponding issues of ICEV/EV/HEV/PHEV/FCV are presented in Table 4.

Fig. 4

"Architecture of different vehicle technologies" - 16 -

Table 4 "Comparison of Different Vehicle Technologies"

In electric propulsion system, there are two popular configurations to interface the ESS with inverter-motor drive as shown in Fig.5: a) A high voltage ESS is directly connected to inverter-motor drive; b) A DC/DC converter is placed between low voltage ESS and inverter-motor drive. In first configuration, battery voltage level should match with inverter motor drive’s voltage level. This imposes certain constraints over design and optimization of battery, inverter and motor. In second configuration, addition of DC/DC converter increases overall component count but offers several advantages over first configuration. It boosts voltage level of ESS to match the rated voltage of inverter-motor drive. It provides bidirectional power flow between ESS and motor drive which assists rapid acceleration and recovers energy to charge the battery during deceleration and regenerative braking. It not only offers significant reduction in weight, size and cost of the ESS but also give the space for inverter control and motor design [1]. - 17 -

The internal impedance of the battery is often not negligible. When an electric motor and inverter are directly connected to the battery without a bidirectional DC/DC converter, as the current goes up, the battery terminal voltage starts to drop because of the voltage drop on the battery internal impedance. This voltage drop will significantly affect the performance of the powertrain motors. In addition, due to the available voltage at the input, the motor constant torque region is also affected (shortened). Another factor is that battery voltage is related to battery state of charge (SOC). As the battery SOC drops, the battery voltage will also drop. Therefore, the available voltage at a motor/inverter terminal is also changed, which will make it difficult to maintain the constant torque range. When a DC/DC converter is inserted between the battery and inverter/motor, the DC bus voltage before the inverter can be maintained as a constant. Therefore, the constant torque range will not be affected by the battery SOC or large power drawn by the inverter/motor [7]. Due to the switching functions of the inverter used in the powertrain system, there are high-frequency current harmonics on the DC side. The amount of current ripples that go into/out of the battery depends on the switching method, switching frequency, and the capacitance on the DC bus. When there is no DC/DC converter in place, the amount of ripple current of the battery is determined by the DC bus capacitance C and the ratio of capacitor impedance to battery impedance. Without the capacitance, the battery current will be directly determined by the switching status of the DC/AC inverter, that is, the combination of the three-phase current of the motor, as shown in Fig.6. When there is a DC bus capacitor in parallel with the battery, the amount of current ripple flowing into/out of the battery is determined by the capacitance and parasitic impedance of the DC bus capacitor. For example, if C=10 mF, the capacitive impedance of the capacitor at switching frequency is only 2.65 mΩ, which is far less than the internal impedance of the battery. Ideally the high-frequency ripple will flow through the capacitor and the battery current is supposed to be constant. However, the parasitic resistance of the capacitor is also not negligible. A high-quality 10 mF capacitor has 26 mΩ internal resistance and the second-class capacitor has 100 mΩ. The quality of the capacitor affects the current ripple of the battery. The lower the capacitor impedance, the lower the battery ripples, as shown in Fig.6. High-frequency ripple current is harmful to battery life. When a DC/DC converter is added, the battery current can be maintained with a relatively small ripple, as shown in Fig. 7. The regenerative braking of the two topologies, that is, one with and without a DC/DC converter, will also be different. In the topology where there is no DC/DC converter, the DC bus voltage will fluctuate during transition from motoring to braking. When braking occurs, a rise in the DC bus voltage will occur making motor control, such as vector control, very difficult. On the other hand, in a system that contains a DC/DC converter between the inverter/motor DC bus and the battery, the DC bus voltage can be maintained relatively constant. Hence, the transition between motoring and braking is easier to handle [7]. - 18 -

Fig. 5 "Architecture of basic configuration of electric propulsion system" (a) Interfacing of high voltage ESS; (b) interfacing of low voltage ESS with bidirectional DC/DC converter.

- 19 -

Fig.6 "Battery current without a DC/DC converter" (a) DC bus current for no DC-bus capacitance. (b) DC bus current for C=10mF, Rc=10 mΩ. (c) DC bus current for C=10 mF, Rc=100 mΩ.

Fig.7 "Battery current when a DC/DC converter is inserted between the inverter and the battery" - 20 -

7.2.Hybrid Energy System The ESS designed with battery, ultra-capacitor or fuel cell individually might not be sufficient to satisfy the peak power demand and transient load variations of the electrified vehicles. Furthermore, utilizing these energy sources alone results in large weight and volume because batteries need large weight and space to meet the power requirements. However, when these energy systems are combined, an energy system with high power and high energy density can be obtained. This energy system is termed as hybrid energy system (HES). The advantages of HES are increased life span of battery, cost reduction of ESS, efficiency, reliability and durability. The more going for hybridization; the dependency on power electronic converter is increases. In HESs, a DC/DC converter is the key element which provides compatible interfacing and integration of energy sources. Apart from output voltage regulation, DC/DC converter can also control the power flow between different energy sources and the load. Therefore, selection and design considerations of DC/DC converters are important factors. Therefore, the basic configuration shown in Fig.5 can be modified according to the selection of DC/DC converter topology and ESS configuration. In advanced architecture of vehicle’s electrical power system, it is expected to have a single DC voltage bus with different voltage level distribution and intelligent power and load management. The modern Architecture of electric power system for vehicular application is shown in Fig. 8. In this architecture, different energy sources and vehicle loads having distinct V-I characteristics and dynamic response are interfaced with common DC bus through PECs. Since PECs are controlling, managing and optimizing the power flow among energy sources and vehicle loads, therefore it is considered as the heart of electric propulsion system [1].

Fig.8 "Role of PEC in modern architecture of electrical power system for vehicular application" - 21 -

7.3.DC/DC Converters A DC/DC converter in its basic form converts unregulated DC input voltage at a certain level to a regulated DC output voltage at a different level with very high conversion efficiency (>90%). Modern day DC/DC converters are operated at high frequencies (10 kHz - 1 MHz). The size of the components such as inductors, transformers, and capacitors are reduced at high frequency operation. Transistors such as MOSFETs and IGBTs are used as switches. The former is preferred in high frequency, low and medium power applications, whereas the latter is preferred in low frequency high power applications. The switches are turned ON and OFF by pulsewidth modulation (PWM) technique. In addition, the role of DC/DC converter is very significant especially in terms of better utilization of energy sources, power management, dynamic performance, flexibility, system optimization and reduction of weight and cost. There has been a constant rise in vehicular power requirement. Performance loads, such as electric steering, that were traditionally driven by mechanical, pneumatic, and hydraulic systems, are now increasingly being replaced by the electrically driven systems, in order to increase the performance and efficiency of operation. Furthermore, luxury loads have also increased over time, achieving a higher demand of electrical power. It must be pointed out here that the rate of increase of automotive loads is about 4% each year. Therfore, load demands have resulted in the need to scale up the onboard vehicular power level. Several decades ago, the voltage was raised from its earlier 6-V level to the present day 12-V level and, now with an ever-increasing demand forecasted into the future, there is a need to switch over to much higher voltage levels of 42 V, 300 V, or higher. Due to the high voltage levels being produced in HEVs, it becomes essential to have DC/DC converters to supply all the auxiliary loads the vehicle. Although the DC/DC converter technology is well developed for low-power applications at lower cost, much work needs to be done for high-power applications. It is a challenge to meet all the vehicle standards for electromagnetic interference (EMI) and electromagnetic compatibility (EMC) as well as specifications of reliability and packaging [8]. From the architecture of the HES, the DC/ DC converters used in EVs may be used for: 1. Providing low voltage (12V or 14V) for low voltage loads. 2. Boosting the voltage level of the battery such that it matches the level of the DC bus voltage for driving the AC motor inverter if the traction motor is AC motor. 3. Motor propulsion if the traction motor is a DC motor.

- 22 -

7.3.1. DC/DC Converters used for supplying Low Voltage Loads (12-14V): Nominal voltages of the battery packs of EV and HEV typically range from 200 volts to 340 volts. The use of high-voltage battery packs requires some kind of power conditioning to provide 12 volts to 12-volt electrical loads. There would always be a need for high-voltage-to-low-voltage power conversion because some loads, such as headlights and microprocessors, require low voltages. Switched-mode converters are used to process high-voltage DC power available from the battery packs of EV and HEV into low-voltage DC power for low-voltage loads. Switched mode technology allows converter operation with high efficiencies and high power densities. The components and systems for automotive applications must provide high performance and high reliability at low cost. Not all types of converters, components, materials, and packaging technologies are suitable for the power converters of EV and HEV. It is important to know what different types of converters, control schemes, materials, and packaging technologies exist to be able to select the converters, control schemes, materials, and packaging technologies [9]. Several types of converters are available with characteristics which make different converters suitable for different applications. These converters include the pulsewidth-modulated (PWM), resonant, quasi-resonant, multi-resonant, and non-linear resonant switch converters. Recently, soft-switched PWM converters have been gaining popularity in the power supply industry. This section discusses various types of converters with the objective to understand their advantages and disadvantages to determine the suitability of these converters for EV and HEV applications. 7.3.1.1.PWM DC/DC Converters: The principle of operation of a PWM converter is discussed with the example of the buck converter. The converter's circuit is shown in Fig. 9. The converter operates by repetitively switching between two topologies. When the active switch is on, the converter elements are connected in one topology, and when and diode is on, the converter elements are connected in a different topology. This switching action converts the input DC voltage into a pulsating voltage as shown in Fig. 9. This pulsating voltage is filtered through a low-pass filter to get DC voltage at the output; the output voltage can be controlled by controlling the duty-ratio. The semiconductor devices should be switched at a high frequency. A high switching frequency leads to a smaller, lighter, and less expensive converter. However, an increase in the switching frequency leads to an increase in the switching losses of devices. The power density of the converter increases with an increase in the switching frequency up to a certain frequency. Any increase in the switching frequency beyond this upper limit leads to a reduction in the power density. The upper limit of the switching frequency of a converter depends on the application for which the converter is designed [9].

- 23 -

Fig. 9 "The circuit and waveforms of buck converter" The semiconductor devices of a PWM converter are switched at non-zero voltages and currents. The switching losses can be minimized by switching the devices when either the voltages or currents of these devices are zero. This is achieved by employing the principle of resonance. Various types of converters have been developed which employ the principle of resonance to minimize the switching losses. 7.3.1.2.Quasi-Resonant Converters: Quasi-resonant converters are derived from PWM converters by employing resonant circuits to modify the switching behavior of the semiconductor devices. There are two types of quasi-resonant converters, namely, the zero-current-switched quasi-resonant converters (ZCS QRC) and zero-voltage-switched quasi- resonant converters (ZVS QRC). The output voltages of both converters are controlled by controlling the switching frequency. As an example, the circuit of a half-wave zero-current-switched quasi-resonant buck convener is shown in Fig. 10(a). This converter is derived from the PWM buck converter and the active switch of this converter is switched at zero current. The resonant inductors and resonant capacitors of the converters are shown inside dashed boxes. The parasitic output capacitance of the switch in the converter of Fig. 10(a) is charged to the input voltage when the switch is off. The energy stored in the switch as a result of the parasitic output capacitance is dissipated in the switch each time the switch is turned on. The turn-on loss of a switch due to the energy stored in the parasitic output capacitance can be minimized by using this capacitance as a part of the resonant circuit and turning the switch on at zero voltage. This principle is also employed in a zero-voltage-switched quasi-resonant converter. As an example, the half-wave zero-voltage-switched quasi-resonant buck converter is shown in Fig. 10(b). This converter is also derived from the PWM buck converter. The reduction of switching losses due to zero-voltage switching allows the operation of a zero-voltage-switched quasi- resonant converter at switching frequencies higher than those in the corresponding zero-current-switched quasi-resonant converter.

- 24 -

Both zero-current-switched and zero-voltage-switched quasi-resonant converters have low switching losses. However, their conduction losses are higher than those of the corresponding PWM converters due to higher voltage and current stresses [9].

Fig.10 "Quasi-resonant converters" a) Half-wave zero-current-switched quasi-resonant buck converter. b) Half-wave zero-voltage-switched quasi-resonant buck converter.

7.3.1.3.Zero-Voltage-Switched Multi-Resonant Converters: Zero-current switching produces favorable switching conditions for the rectifying diode, but not for the active switch. Zero-voltage switching, on the other hand, produces favorable switching conditions for the active switch, but not for the rectifying diode. A zero-voltage-switched multi-resonant converter (ZVS-MRC) absorbs both the parasitic output capacitance of the active switch and the parasitic junction capacitance of the rectifying diode, thus providing favorable switching conditions for both devices. Such a converter is derived from a PWM converter by introducing an inductor in the circuit such that it forms resonant circuits with the parasitic output capacitance of the active switch and parasitic junction capacitance of the rectifying diode. As in the case of zero-current-switched and zero-voltage-switched quasi-resonant converters, the output voltage of a zero-voltage-switched multi-resonant converter is controlled by controlling the switching frequency. The circuit of a zero-voltage-switched multiresonant buck converter is shown in Fig. 11. Zero-voltage switching allows a zero-voltage-switched multi-resonant converter to operate with low switching losses. However, higher voltage and current stresses of the zero-voltage- switched multi-resonant converter lead to higher conduction losses than those of the corresponding PWM converter [9]. - 25 -

Fig. 11 "zero-voltage-switched multi-resonant buck converter" 7.3.1.4.Constant-Frequency Quasi-Resonant and Multi-Resonant Converters: The switching frequency is employed as the control variable in the conventional quasi-resonant and multi-resonant converters. Operation with variable switching frequency is undesirable, especially if the input voltage and load vary over a wide range. Variation of switching frequency over a wide range does not allow optimization of the designs of inductors, transformers, and input and output filters. Constant-frequency operation of quasi-resonant and multi-resonant converters can be achieved by replacing the diodes with active switches. The switching transitions and the voltage and current stresses of constant-frequency quasi-resonant and multiresonant converters are similar to those of the conventional quasi- resonant and multiresonant Converters [9].

7.3.1.5.Non-Linear Resonant Switch Converters Zero-current- and zero-voltage-switched quasi-resonant converters and zero-voltageswitched multi-resonant converters lead to significant reductions in switching losses and allow converter operation at switching frequencies higher than those in the corresponding PWM converters. However, the conduction losses of these converters are higher than those of the corresponding PWM converters. High conduction losses in these converters put an upper limit on the achievable power densities of Converters. The achievement of high power density in a converter requires both switching and conduction losses to be low. Non-linear resonant switch Converters are also derived from the PWM converters. The circuit of a non-linear resonant switch converter is similar to that of the corresponding zero-current- switched quasi-resonant converter except that the resonant inductor used in the non-linear resonant switch converter is non-linear. Thus, the circuit of Fig. 10(a) also represents the non-linear resonant switch buck converter.

- 26 -

When the current flowing through the switch of the non-linear resonant switch buck converter is below a set value, the resonant inductor is saturated so that the inductance in series with the switch is low. Low inductance causes the switch current to increase rapidly. When the current flowing through the switch exceeds the set value, the resonant inductor comes out of saturation and the inductance in series with the switch increases. Higher inductance decreases the rate of rise of current through the switch. When the switch current begins to decrease and reaches the set value, the resonant inductor becomes saturated again. Less inductance allows the current to decrease at a fast rate [9]. The result of this type of operation is that the switch current approaches that in a corresponding PWM converter and the peak value of switch current is significantly reduced. The peak voltage stresses of a non-linear resonant switch converter are equal to those of the corresponding PWM converter and the peak current stresses are typically 20 % greater than those of the corresponding PWM converter. These voltage and current stresses are significantly less than those in the conventional quasiresonant and multi-resonant converters. Thus, the use of a non-linear resonant inductor in a non-linear resonant switch converter leads to reductions in both switching and conduction losses, allowing the converter to operate with higher power densities than those in the conventional quasi-resonant and multi-resonant converters. 7.3.1.6.Resonant Converters: Resonant converters employ resonant circuits to achieve soft switching of devices. In the case of resonant DC-to-DC converters, the voltages or currents of the resonant circuits are rectified. The circuits of two of the basic resonant converters, namely, the series and parallel resonant converters are shown in Figs. 12(a) and (b), respectively. In the series resonant Converter, the resonant current is rectified, whereas in the parallel resonant converter, the resonant voltage is rectified. Constant-frequency operation of resonant converters which employ full-bridge arrangement of switches is possible. The switches in constant-frequency resonant converters operate at 50 % duty-ratio. However, the phase between the drive signals of diagonally opposite switches of these converters can be controlled to control the voltage applied to the resonant circuit. This allows the operation of resonant converters at constant frequency. Resonant converters operate with low switching losses. However, the conduction losses of these converters are high due to high current and voltage stresses [9].

- 27 -

(a)

(b)

Fig. 12 (a) Series resonant converters (b)Parallel resonant converters 7.3.1.7.Soft-Switched PWM Converters: PWM converters have low voltage and current stresses, and hence, low conduction losses. The principle of resonance can be employed to achieve zero-current or zerovoltage switching to reduce the switching losses of converters. Thus, the switching losses of a converter can be significantly reduced if the converter operates as a PWM converter for most of the time during a switching period and employs resonance only during transitions between the on and off states. This concept has served as the motivation for the development of zero-voltage-switched PWM converters. A zerovoltage-switched PWM converter of particular interest is the zero-voltage-switched full-bridge (ZVS FB) PWM converter shown in Fig. 13. This converter employs the leakage inductance of the transformer and the parasitic output capacitances of the devices to achieve zero-voltage switching of devices. The inductor inside the dashed box in Fig. 13 represents the leakage inductance of the transformer [9].

- 28 -

Fig. 13 "The circuit of the zero-voltage-switched full-bridge PWM converter"

7.3.1.8.Zero-Voltage-Switched/Zero-Current-Switched PWM Converters: An attempt to use IGBTs in converters has led to another family of soft-switched PWM converters, namely, the zero-voltage-switched/zero-current-switched (ZVS/ZCS) PWM converters. The circuit of a zero-voltage-switched zero-currentswitched PWM converter is shown in Fig. 14. The inductor inside the dashed box represents the leakage inductance of the transformer. The IGBT bridge in this converter does not employ anti-parallel diodes and is controlled such that two of the four devices of the bridge switch at zero voltage and the remaining two switch at zero current [9].

Fig. 14 "The circuit of a zero-voltage-switched/zero current switched PWM converter" - 29 -

7.3.1.9.Selection of Converters: The quasi-resonant, multi-resonant, and resonant conveners operate with high voltage and current stresses which lead to high conduction losses. With the present state of technology, these converters do not appear to be suitable for electric and hybrid vehicles which employ converters with power ratings of 1 kilowatt or more. These converters also require more parts which is not desirable, particularly for the converters of relatively high power ratings. The non-linear resonant switch converters have voltage and current stresses comparable to those in the corresponding PWM converters, but they use more parts. The quasi-resonant, multi- resonant, and non-linear resonant switch converters appear to be suitable for vehicles which employ converters with power ratings in the 50- to 100-watt range. The converters of electric and hybrid vehicles have high input voltages and low output voltages. This implies that the input currents are low and output currents are high. The phase-shift control scheme employed in the zero-voltage-switched fullbridge PWM converter leads to zero-voltage switching of semiconductor devices on the primary side of the transformer. This scheme does not lead to reduction of losses in the output diodes. Since the input currents are low and the output currents are high, the losses in the output stage of the converter form a significant part of the total converter losses. The benefits of zero-voltage switching employed in the zero-voltage-switched fullbridge PWM converter are realized more effectively if the output voltage is greater than 12 volts, for example, 48 volts or 72 volts [9]. A full-bridge PWM converter employs four transistors on the primary side and requires four drive circuits to control these transistors. If the required power level is such that the use of a full-bridge PWM converter is necessary, then it is desirable to employ a zero-voltage-switched full-bridge PWM converter. Otherwise, the use of a PWM converter which employs fewer parts than a full-bridge PWM converter turns out to be a good choice in terms of the simplicity of control, packaging, reliability, and cost. The comments which apply in the case of zero-voltage-switched full-bridge PWM converter also apply in the case of zero-voltage-switched/zero- current-switched PWM converters. Experience has shown that operation at switching frequencies from 100 KHz to 110 kHz with efficiencies from 88 % to 90 % is achievable in PWM converters designed for EV and HEV with power levels form 1 kilowatt to 1.25 kilowatts [9].

- 30 -

7.3.2. DC/DC Converters used for boosting the battery voltage: Conventional ICE vehicles mostly employ 12-volt electrical power systems. Some conventional vehicles, such as buses, employ 12/24 volt systems. The 12- and 24-volt systems are unsuitable for the propulsion of electric vehicles. High power requirements for the propulsion of electric vehicles require high system voltages for efficient operation. Also, high power DC/DC converters are needed for EVs since the vehicular power requirements are continuously increasing due to which the present day 12-V/14-V electrical system will be replaced by 42-V/300-V architecture. DC/DC converters are well developed for low and medium power applications, whereas development of highly efficient and cost effective high power DC/DC converters for vehicular applications is in continuous progress. This is partly due to the stringent EMI standards and also due to temperature related issues [10]. AC motors in EVs and HEVs are fed by inverters which in-turn is fed by a high voltage DC/DC converter. As an example, in Toyota Prius the low voltage bus is at 201.3 V Nickel-Metal Hydride (Ni-MH) battery output, which is converted to 500 V by a simple boost converter to feed a 50KW (1200-1540 rpm) PMSM. This arrangement is shown in Fig.15.

Fig.15 "Role of DC/DC converter in voltage boosting" DC/DC converters in an EV may be classified into unidirectional and bidirectional converters. Unidirectional DC/DC converters cater to various onboard loads such as sensors, controls, entertainment, utility, and safety equipment. They are also used in DC motor drives electric traction. Bidirectional DC/DC converters find applications in places where battery charging, regenerative braking, and backup power are required. The power flow in a bidirectional converter is usually from a low voltage end such as battery or a super capacitor to a high voltage side and is referred to as boost operation. During regenerative braking, the power flows back to the low voltage bus to recharge the battery (buck mode). As a backup power system, the bidirectional DC/DC converter facilitates the safe operation of the vehicle when ICEs or electric drives fail to drive the motor. Due to the aforementioned reasons, high power bidirectional DC/DC converters have gained a lot of importance in the recent past [10].

- 31 -

Both unidirectional and bidirectional DC/DC converters are preferred to be isolated to provide safety for the loading devices. Most of the DC/DC converters incorporate a high frequency transformer but the volume, weight and cost are increased. Since power in full-bridge converters can flow in both directions, development of bidirectional full-bridge based converters are in demand. To reduce the number of components and still maintain the benefits of full-bridge versions, many half-bridge based topologies are also developed. In addition to these, to overcome the high voltage/current stresses due to energy stored in the transformer leakage inductance, passive snubbers, active-clamping, active commutation, soft commutation, and softswitching solutions have been developed.

7.3.2.1. Non-isolated Bidirectional Half Bridge DC/DC Converter: Basic DC/DC converters such as buck and boost converters (and their derivatives) do not have bidirectional power flow capability. This limitation is due to the presence of diodes in their structure which prevents reverse current flow. In general, a unidirectional DC/DC converter can be turned into a bidirectional converter by replacing the diodes with a controllable switch in its structure. As an example, Fig.16 shows the structure of elementary buck and boost converters and how they can be transformed into bidirectional converters by replacing the diodes in their structure. The result is called the Non-isolated Bidirectional Half Bridge DC/DC Converter. The operating principle is shown in Fig. 17 [11].

Fig. 16 "(a) Elementary unidirectional buck converter, (b) elementary unidirectional boost converter and (c) transformation to bidirectional converter by substituting diodes with a controllable switch" - 32 -

Fig. 17 "Operation of the bidirectional boost converter: (a) circuit topology; (b) inductor voltage and current waveform during buck operation; and (c) inductor voltage and current waveform during boost operation"

- 33 -

Some of the major limitations associated with this converter are: • It can only operate in buck mode in one direction and boost in the other. • When the voltage ratio becomes large, this structure becomes impractical. • The lack of galvanic isolation between two sides.

7.3.2.2.Isolated Bidirectional Full Bridge DC/DC Converter (IBDC): Galvanic isolation between multi-source systems is required by many standards. Personnel safety, noise reduction and correct operation of protection systems are the main reasons behind galvanic isolation.

Fig.18 "IBDC Full Bridge topology" Fig. 18 shows a common IBDC topology which is sometimes called dual active (full) bridge (DAB). In this configuration, full-bridge voltage-fed converters are used at both sides of the isolation transformer and the control is performed based on softswitched phase-shift strategy. In its basic form, the diagonal switching pairs in each converter are turned on simultaneously with 50% duty cycle (ignoring the small dead time) and with 1800 phase shift between two legs to provide a nearly square wave AC voltage across transformer terminals. The phase shift between two AC voltages, denoted by φ, is an important parameter which determines the direction and amount of power transfer between DC buses. By adjusting this phase shift, a fixed frequency operation with full control over the power transfer is possible. Fig. 19 shows the ideal waveforms of A-to-B and B-to-A power transfer modes. To transfer power from Side A to Side B (A-to-B mode), vac,A should lead vac,B and φ is considered as positive. In B-to-A mode, vac,A should lag vac,B and φ is negative. This leading or lagging phase shift is simply implemented by proper timing control of converter switches [11].

- 34 -

Fig. 19 "Operating waveforms of IBDC (a) A-to-B mode and (b) B-to-A mode"

- 35 -

Some of the advantages of this converter can be listed as below. i. In this topology, each converter provides an AC waveform with a peak value close to the DC voltage at its terminal, therefore the voltage stress across each switch is limited to the bus voltage level. ii. The current stresses of all switches on each side are almost equal. iii. There is no need for additional active or passive elements for having soft switching. iv. Transformer has a simple structure that simplifies the designing and manufacturing tasks. v. Another important feature is the fast dynamic behavior due to lack of additional passive components. Some of the disadvantages are as follows. i. The currents flowing in DC buses contain high ripple content; therefore appropriate filtering circuits are necessary. ii. Proper control is required to prevent DC saturation on both sides as there is no inherent DC current blocking capability for transformer windings. iii. Similar to many other topologies, the converter may lose soft switching in light load conditions. iv. The control is highly sensitive to slight variations of φ, especially when bus voltages are high. Thus if a digital controller is considered, very high resolution phase shift timers are required. v. Another disadvantage is relatively high component count that leads to larger driver size, higher gate losses and increased cost compared to low switch count topologies

7.3.2.3.Isolated Bidirectional Dual Half Bridge DC/DC Converter: Fig. 20 illustrates another IBDC converter that is called dual half bridge (DHB). This topology consists of one voltage-fed half bridge converter in Side B (usually higher voltage side) and a modified current-fed half bridge converter (also called boost-halfbridge) in Side A. The current-fed side is the lower voltage side because it usually consists of battery or ultra-capacitor DC energy sources in which low ripple current is desirable. In practice the voltage amplitude is a few tens of volts for the low voltage side (battery or ultra-capacitor) and a few hundreds of volts for the high voltage side. Similar to DAB discussed in Sec. 7.3.2.2, the power regulation is achieved by controlling the phase shift between the voltages applied to two sides of transformer, or equivalently to the leakage inductance of the transformer. The leakage inductance (plus any series inductance) is the energy transfer element like in DAB [11].

- 36 -

Fig.20 "Dual half bridge IBDC" Basic waveforms of the converter with 50% duty cycle in both power flow modes are shown in Fig 21.

- 37 -

Fig. 21 "Operating waveforms of DHB (a) A-to-B mode and (b) B-to-A mode"

The main advantages of this configuration can be listed as follows. i. Low switch count compared to other topologies that normally use full bridge converters. ii. The total device rating of active elements is same as a DAB with the same power. iii. This converter adopts a dual half-bridge topology which can offer ZVS with neither a voltage-clamping circuit nor additional switching devices and resonant components, leading to a reduced number of devices and hence compact packaging. iv. Relatively simple control of the converter based on well-known phase shift modulation. v. Low ripple current at the current fed side that is desirable for batteries and ultra-capacitors. The main drawbacks of the converter are: i. Large ripple current in the splitting capacitors especially in LV side. ii. Unbalanced current stress between two switches in the LV side. It can be noted that this converter can be used as a DC motor drive for EV applications when the traction motor is DC motor. - 38 -

7.3.2.4.Interleaved DC/DC Converter: It consists of "n" numbers of DC/DC boost converters connected in parallel. The main goal of this converter is to reduce the size of the passive components by "n" times compared to the conventional boost converter (BC). Therefore, system reliability and converter power rating are increased by using the parallel phases. In [12], a four phase interleaved DC/DC converter (FP-IBC) was proposed. It consists of 4 DC/DC boost converters connected in parallel and it is shown in Fig.22. Fig.23 shows the switching devices gate signals at duty cycle D=0.25, the gate signals are phase shifted by TS / (n*m), where: TS: switching period, n: number of phases, m: number of devices in parallel per phase. For FP-IBC, m=1 and n=4. The current is shared equally between each phase and has ripple content of period T S/4. The frequency of the output voltage and current is "n" times higher than the switching frequency. As a result, the size of passive components reduced by "n" times compared to the conventional BC. The equally sharing current between each phase will provide tight size of power semiconductors, distribution of losses between modules and size optimization of the converter. These advantages opposed to other converter topologies such as conventional BC, multi device BC, two phase IBC, and multi device IBC shown in Figs. 24, 25, 26, and 27 respectively.

Fig. 22 "The structure conventional FP-IBC"

- 39 -

Fig.23 "The switching pattern of FP-IBC"

Fig.24 "Conventional Boost Converter (BC)"

Fig.25 "Multi device Boost Converter (MDBC)"

- 40 -

Fig.26 "Two-phase IBC"

Fig.27 "Multi device interleaved BC (MDIBC) (m=2and n=2)"

Also in [12], the power losses of the above mentioned DC/DC converters were calculated and efficiencies were compared as shown in Fig.28.

Fig.28 "DC/DC Converters efficiency versus switching frequency" - 41 -

Fig. 29 shows the distribution of all losses for each converter at full-load operation and 20 kHz. This case is chosen to investigate the worst case scenario in terms of power losses. As it is illustrated in Fig. 29, the proposed converter is able to reduce the total losses and especially the passive elements losses. This comes from the reduction of its passive components size by four times compared to conventional BC and two times compared to TP-IBC and MDBC. It is noticed that the proposed converter (FP-IBC) efficiency characteristics make it a good candidate for EVs, particularly in high-power applications.

Fig. 29 "Distributed power losses in (W) for each DC/DC converters" 7.3.3. DC/DC Converters used for DC Motor Drives: DC motor drives take the advantages of mature technology and simple control. However, their commutators and brushes make them less reliable and unsuitable for maintenance-free operation. Thus, the DC motor drives are mainly applied to low-cost EVs, such as motorcycles and mini EVs. The corresponding DC-link voltage and power ratings are 24-48 V and 1-3 kW, respectively. The power converters for DC motor drives are traditionally based on hard-switching which results in pulsating currents and voltages, thus imposing high voltage and current stresses on power devices and contributing to electromagnetic interference (EMI). Although there have been many soft-switching DC/DC converters developed for switched-mode power supplies, these converters cannot be directly applied to DC motor drives for EV propulsion. Apart from suffering excessive voltage and current stresses, they cannot handle backward power flow during regenerative braking. - 42 -

It should be noted that the capability of regenerative braking is very essential for EVs as it can extend the vehicle driving range by up to 25% [13]. 7.3.3.1.Two-Quadrant Zero-Voltage-Switching Multi-Resonant (2Q-ZVS-MR) Converter: In recent years, some soft-switching DC/DC converters have been specially developed for EV propulsion, which offers the capability of bidirectional power flow for motoring and regenerative braking. As shown in Fig. 30, a two-quadrant zerovoltage-switching multi-resonant (2Q-ZVS-MR) converter has been applied to DC motor drives. The 2Q–ZV–MR converter is created by adding a resonant inductor and two resonant capacitors to a conventional 2Q–PWM DC drive. The major advantages of this converter are ZVS operation of both power switches, full ranges of voltage conversion-ratio and load, constant-frequency operation, capability of short-circuit operation, and absorption of all major parasitics. However, the high circulating energy and hence the conduction losses are significantly increased, resulting the power devices and other circuit components to be rated for higher VA ratings, as compared with their PWM counterpart [13].

Fig. 30 "2Q-ZVS-MR converter for DC motor drives" 7.3.3.2.Two-Quadrant Zero-Voltage-Transition (2Q-ZVT) Converter: Consequently, as shown in Fig. 31, a two-quadrant zero-voltage-transition (2Q-ZVT) converter has been specially developed for DC motor drives. The 2Q-ZVT converter is created by adding a resonant inductor, a resonant capacitor and two auxiliary switches to a conventional 2Q-PWM DC drive. It possesses the advantages that both the main switches and diodes can switch with ZVS and unity device stresses during both the motoring and regenerating modes of operation. It also offers a simple circuit topology and low cost, leading to achieve high switching frequency, high power density, and high efficiency. Other key features are the use of the same resonant tank for both forward and backward power flows and the full utilization of all built-in diodes of the power switches, thus minimizing the overall hardware count and cost. This 2Q-ZVT converter is particularly useful for those power MOSFET-based DC motor drive applications, which generally suffer from severe capacitive turn-on switching losses [2], [13]. - 43 -

Fig.31 "2Q-ZVT converter for DC motor drives" 7.3.3.3.Two-Quadrant Zero-Current-Transition (2Q-ZCT) Converter: As an extension from the 2Q-ZVT converter, a two-quadrant zero-current-transition (2Q-ZCT) converter has also been proposed for DC motor drives. As shown in Fig. 32, the 2Q-ZCT converter is created by adding a resonant inductor, a resonant capacitor and two auxiliary switches to a conventional 2Q-PWM DC drive. This 2QZCT converter possesses the advantages that both the main and auxiliary switches can operate with zero-current switching (ZCS) during both the motoring and regenerating modes. It takes the role to be particularly useful for those IGBT-based DC motor drive applications, which generally suffer from severe inductive turn-off switching losses. At present, most commercially available electric motorcycles and mini EVs utilize DC motor drives for propulsion, and all of them adopt hard switching DC/DC converters. Since motorcycles and mini vehicles are widely accepted in populated cities such as Tokyo, Beijing and Taipei, the development of soft-switching DC/DC converters for electric motorcycles and mini EVs has a definite market value. The major challenges are to boost the power level of those soft-switching converters up to 3 kW, and to utilize at least 2Q but preferably four-quadrant (4Q) operation [2], [13].

Fig.32 "2Q-ZCT converter for DC motor drives" - 44 -

7.4.DC/AC Converters: A suitable DC/AC inverter draws DC power from the batteries to drive the electric traction motor, which in turn provides power to the wheels. The DC/AC inverter also performs the function of recharging the batteries during regenerative braking in HEVs. Also, Inverters are used for supplying utility loads such as Air conditioning and power outlets. Inverters are generally classified into voltage-fed and current-fed types. Because of the need of a large series inductance to emulate a current source, current-fed inverters are seldom used for EV propulsion. In fact, voltage-fed inverters are almost exclusively used because they are very simple and can have power flow in either direction. A typical three-phase full-bridge voltage-fed inverter is shown in Fig.33. The output waveform of an inverter may be rectangular, six–step or pulse width modulation (PWM), depending on the switching strategy for different applications. For example, a rectangular output waveform is produced for a PM brushless DC motor, while a six–step or PWM output waveform is for an induction motor. It should be noted that the six–step output is becoming obsolete because its amplitude cannot be directly controlled and rich in harmonics. On the other hand, the PWM waveform is harmonically optimal and its fundamental magnitude and frequency can be smoothly varied for speed control. Starting from the last decade, numerous PWM switching schemes have been developed for voltage-fed inverters, focusing on the harmonic suppression, better utilization of DC voltage, tolerance of DC voltage fluctuation as well as suitability for real-time and microcontroller–based implementation [13].

Fig. 33 "Three-phase full-bridge voltage-fed inverter (VSI)"

- 45 -

The development of inverters for AC motor drives has become one of the key research areas in electric propulsion. The wish list of those inverters include efficiency over 95%, power density over 3.5 W/cm3, switching frequency over 10 kHz, dv/dt below 1000 V/µs, zero EMI, zero failure before the end of the vehicle life, and redundancy with limp-home mode. In order to achieve this wish list, there are two major research directions, namely the soft-switching inverters and the multilevel inverters [2].

7.4.1. H-Bridge Topology with Single DC link:

Fig. 34 "H-bridge with topology with single DC link"

Combining two "three-phase full-bridge voltage-fed inverter'' circuits, one arrives at the H-bridge topology shown in Fig. 34, here fed from a single DC link, which operates an open-winding machine. This topology allows to achieve a higher phase voltage for a given DC link voltage VDC for pulse-width modulation (PWM) and block mode operation. For the conventional VSI inverter, space vector modulated PWM (SVPWM) and 1800 block mode is considered. The H-bridge with single DC link behaves similar to a VSI at 2VDC with sine-triangle modulation in PWM mode and an on-time of 1200 in block mode. In both modes, the phase voltage increases by 73% which is a factor of √ using an H-bridge. Since the switch count doubles while the phase voltage does not, the switch utilization ratio (SUR), is reduced by 13.4% which is a factor of √ /2. This leads to a corresponding reduction of the relative efficiency. In terms of voltage increase and SUR, the H-bridge with two electrically isolated DC links can be considered as the combination of two independent VSIs, thus doubling the voltage while keeping the SUR constant. Here, switching losses can be reduced by always operating one converter in block mode. Aspects in favor of an H-bridge topology may in certain cases outweigh the additional effort and losses (in case of a single DC link). - 46 -

Due to the higher phase voltage of an H-bridge inverter, eddy current losses in the windings can potentially be reduced. Without adjusting the machine, the high-speed power output can be increased. In an electric vehicle with more than one traction motor, limited propulsion (“limp home” capability) can be provided by the remaining motors (given that the faulty motor can be brought into a save state and does not disable the remaining system). In vehicles with only one traction motor, loss of this motor means total loss of propulsion. For motors with VSI inverter, this is the case if one of the switches fails. Under the assumption of an IGBT with an open-circuit fault and an induction machine as traction motor, the remaining system can be seen as a single phase induction machine [14].

7.4.2. Soft-Switching Inverters: The soft-switching inverters can be categorized as the resonant DC-link types and the resonant AC pole types. The milestone of those resonant DC-link types is the wellknown resonant DC-link inverter. Subsequently, many improved soft-switching topologies have been proposed, such as the quasi-resonant DC-link inverter, series resonant DC-link inverter, parallel resonant DC-link inverter, and synchronized resonant DC-link inverter. On the other hand, the resonant AC-pole types include the resonant pole inverter, auxiliary resonant commutated pole inverter, auxiliary resonant snubber (ARS) inverter, zero voltage transition (ZVT) inverter, and zero current transition (ZCT) inverter. Among them, the ARS inverter has been actively developed for electric propulsion. On the other hand, the ZCT inverter has also been actively developed for electric propulsion [15].

7.4.2.1.Resonant DC link Inverter: Fig. 35 shows a basic resonant DC link (RDCL) inverter. In this circuit, the inverter input voltage is pulsating by adding a parallel resonant network between the DC voltage source and the inverter bridge, therefore the link voltage has zero crossings which create the desirable ZVS conditions for inverter switches. The peak resonating voltage is twice the DC source voltage under no load condition and more than three times the DC source voltage under the transition from motoring mode to regeneration mode. By adding an auxiliary switch and a stored-voltage clamp capacitor in the conventional, as shown in Fig.36, a better voltage clamping level of 1.3-1.5 times the DC source voltage can be achieved, but the additional components result in increased cost and reliability penalty.

- 47 -

Major advantages of RDCL are as follows [13], [15]: – Minimum number of power devices. – Elimination of snubbers. – ZVS for main switches. – Low dv/dt at motor terminals. – High resonant and switching frequencies. – Low sensitivity to parasitic impedance. There are several improved versions based on the basic RDCL inverter. However, some technical problems remain unsolved, such as: – High voltage stress of 1.3-1.5 times the DC source voltage (even with clamping). – High switching losses for the auxiliary switches and diodes. – Pre-charging problem of the voltage clamp capacitor. – Rich in sub-harmonics.

Fig. 35 "Three-phase voltage-fed RDCL inverter"

Fig. 36 "Active clamped RDCL inverter" - 48 -

7.4.2.2. Auxiliary Resonant Commutated Pole Inverter: The auxiliary resonant commutated pole inverter (ARCP) or quasi-resonant inverter shifts the resonant inductor away from the main power flow and connected to the split capacitors for bidirectional commutation, as shown in Fig.37. A bidirectional switch is series connected to the resonant inductor to control the direction of resonant energy transfer. These auxiliary switches are operated with ZCS and required to withstand only half of the DC source voltage. Since the auxiliary switches are not associated with any load energy transfer, their power ratings are much smaller than the main power devices. Main features of ARCP inverters are as follows [13], [15]: – Conventional PWM or space vector modulation (SVM) can be applied for controlling the ARCP. – Unity voltage/current stresses on main switches – Equivalent spectral performance to hard-switching converter. – Auxiliary switches are required to withstand half of the DC source voltage only. – ZVS for main switches while ZCS for auxiliary switches. However, some technical problems remain unsolved, such as: – System performance varies with load current. – Additional bulky energy storage capacitors are required. – Several split capacitors are required. – Long resonant period since only half DC source voltage is applied for resonance.

Fig. 37 "ARCP inverter" 7.4.2.3.Auxiliary Resonant Snubber Inverter: An improvement of ARCP inverter, the resonant circuit of ARS inverter is placed between phase outputs, instead of using a center tapped DC link for commutation. The principle of ARS inverter is to utilize the resonant capacitor across the device to achieve zero turn-off loss and the resonant inductor along with the auxiliary switches to achieve zero-voltage turn on. The auxiliary branch is connected between two phase legs.

- 49 -

Circuit operation of ARS inverter is similar to ARCP inverter except its commutation relies on the interaction between at least two phase-legs. By using auxiliary switches and resonant inductors along with resonant snubber capacitors to achieve the soft switching condition, the three-phase topology of the ARS inverter is shown in Fig.38. Although this ARS inverter has promising applications to EV propulsion, it still needs continual improvement before practically applying to EVs. Particularly, the corresponding control complexity should be solved, while the corresponding PWM switching scheme needs to be modified to enable variable speed control of induction motor drives [13], [15].

Major features of ARS inverter are summarized as follows: – Minimum auxiliary components when compared with ARCP inverter and low cost. – Simple resonant inductor current control. – ZVS for main switches and ZCS for auxiliary switches. – Parasitic inductance and stray capacitance are utilized as part of the resonant components. – Comparatively less over-voltage or over-current penalty in main switches. – Modification of conventional PWM or SVM control strategies is required.

Fig. 38 "Delta-connected ARS inverter"

- 50 -

7.4.2.4.Zero Current Transition Inverter: It consists of a conventional PWM converter and an auxiliary bridge with series resonant tank Lo and Co. each switch of the auxiliary circuit provides ZCT operation of the active switch on the corresponding phase leg. The control is more complex compared to the hard-switched version as the current zero crossing has to be ensured. Fig. 39 shows the 6-switch ZCT inverter in which ZCS is achieved by using a bidirectional resonant current to divert the currents from the main switches to the auxiliary circuit. It takes the definite advantage that each main switch has its own auxiliary switch for ZCS, hence offering flexible PWM. Minimizing the voltage and current ratings on active switches and diodes and reducing the conduction losses in the auxiliary circuit but diodes are not subjected to soft switching. In order to save the number of auxiliary switches, the 3-switch ZCT inverter has been proposed as shown in Fig. 40 in which one auxiliary switch serves for one phase, rather than one main switch. For more details about soft switching inverters, [15] is appreciated.

Fig.39 "6-switch ZCT inverter for AC motor drives"

Fig. 40 "3-switch ZCT inverter for AC motor drives" - 51 -

7.4.3. Multilevel Inverters: Multilevel inverters can generate near sinusoidal voltages with only fundamental frequency switching, have almost no EMI, very small THD, and no common mode voltage. Traditional 2-level high frequency PWM inverters have problems associated with their high voltage change rates dv/dt which produces a common mode voltage across the motor windings. High frequency switching can exceed the problem because of the numerous times this common mode voltage is impressed each cycle. PWM controlled inverters also require a greater amount of heat removal because of the additional switching losses. Multilevel inverters solve these problems because their individual devices have a much lower dv/dt per switching, and they operate at higher efficiencies because they can switch at a lower frequency than PWM controlled inverters. Multilevel inverters are suitable for heavy duty trucks and military combat vehicles because of the high VA ratings available with these inverters [16]. The general function of multilevel inverter is to synthesize a desired voltage from several levels of DC-voltages. Therefore, they can easily provide the high power required for a large EV drive.

7.4.3.1.Cascaded H-Bridge Multilevel Inverter:

It consists of a series of H-bridge (single phase full bridge) inverter units in each of its three phases. Fig.41 shows an 11-level phase neutral cascade inverter connected in star configuration. Each H-bridge unit has its own DC source (battery unit in EVs). Fig.42 shows one H-bridge with its associated gating signals and output voltage. The output voltage of the inverter is almost sinusoidal, it is shown in Fig.43 and it has less than 5% THD.

- 52 -

Fig.41 "11-level phase neutral cascade inverter connected in star configuration"

Fig.42 "H-bridge with its associated gating signals and output voltage"

- 53 -

Fig.43 "The output voltage of the Cascaded H-Bridge Multilevel Inverter"

7.4.3.2.Back-To-Back Diode Clamped Multilevel Inverter: For heavy duty trucks and military vehicles, fuel efficiency, lower emissions, better performance are crucial issues. Back-to-back diode clamped multilevel inverter can meet the high power and/or voltage for the traction motor. Each of the three phases of the inverter shares a common DC bus that is divided into smaller equal levels of DC voltage by either batteries or capacitors, as shown in Fig.44. The output AC voltage waveform is constructed by connecting the three lines to different levels of DC voltage such that a staircase waveform is generated. As with the cascaded inverter, almost sinusoidal output voltage can be achieved. Fig.45 shows the Vab output voltage and the switching states for phase A [16].

- 54 -

Fig.44 "Back-To-Back Diode Clamped Multilevel Inverter"

Fig.45 "the Vab output voltage and the switching states for phase A"

7.4.3.3.T-Type Five Level Converter: The five-level T-type converter (which called T5 converter), shown in Fig.46, has a reduced switching elements configuration compared to conventional 5-level diode clamped converter (DCC). The advantages of it are elimination of clamping diodes from the power circuit and reduction of isolated DC power supplies for the driving circuits. The output voltage has low harmonics. This type of converter has shown better characteristics compared to the 2-level inverter as well as 5-level DCC in terms of losses and harmonics.

- 55 -

The phase displacement PWM technique is used to modulate this converter. This technique is much simpler than SVM. It is based on comparison between four carrier waveforms and the modulation waveform. In [12], comparisons between the conventional 2-level inverter, 5-level DCC, and Ttype 5-level inverter were made to calculate the power losses, efficiencies, and THD at different conditions. The T5 converter loss is lower than 5-level DCC. This comes from the reduced switching elements of the converter power circuit. Efficiencies of these three converters are shown in Fig. 47 versus the switching frequency. It is well known that the main power electronics losses are conduction and switching losses. In Fig. 48, a pi-chart shows the values and percentages of switching and conduction losses for two-level, 5-level DCC, and T5 converters at 20 kHz in case of motor full-load. The results clarified that T5 converter has the lowest total loss. Moreover, T5 topology switching loss to converter total loss ratio is the lowest compared to DCC and two-level converter. This reflects the low switching stress on the converter elements. For the harmonic analysis of the DC/AC converter, the total harmonic distortion (THD) factor is considered a common factor for evaluating the converter output. The THD factor is described by (1) as follows:

(1)

While Vh, V1 are the root-mean square (RMS) values for the harmonic-order and fundamental voltage respectively Fig. 49 shows the THD factor for a wide range of switching frequency operation at full-load case study. The THD factor for T5 converter appears as frequency independent. This reflects the availability of operating the converter at low frequency range. However the THD factor for two-level converter is high at the low frequency range. The THD factor machine current is damped because of the machine impedance. The THD factor for T5 converter is much lower compared to two-level one.

- 56 -

Fig.46 "Circuit diagram for one-leg of T5-type converter"

Fig. 47 "Converters efficiency versus switching frequency"

- 57 -

Fig. 48 "Distributed power losses in (W) for each DC/AC Converters"

Fig. 49 "THD factor for two-level and T5 converter topologies" 7.4.4. Impedance Source Inverter (ZSI):

Fig.50 "Impedance Source Inverter (ZSI)" - 58 -

The Z-source inverter employs a unique LC network to couple the inverter main circuit to the diode front end. The Z-source can produce any desired output AC voltage, even greater than the line voltage. For the traditional VSI, obtainable output voltage is quite limited below the input line voltage. The V-source inverter is a buck (step-down) inverter. In Fig.50, the three-phase Z-source inverter bridge has nine permissible switching states (vectors) unlike the traditional three-phase V-source inverter that has eight. The traditional three-phase V-source inverter has six active vectors when the DC voltage is impressed across the load and two zero vectors when the load terminals are shorted through either the lower or upper three devices, respectively. However, the three-phase Z-source inverter bridge has one extra zero state (or vector) when the load terminals are shorted through both the upper and lower devices of any one phase leg (i.e., both devices are gated on), any two phase legs, or all three phase legs. This shoot-through zero state (or vector) is forbidden in the traditional V-source inverter, because it would cause a shoot-through. This third zero state (vector) is called the shoot-through zero state (or vector), which can be generated by seven different ways: shoot-through via any one phase leg, combinations of any two phase legs, and all three phase legs. The Z-source network makes the shoot-through zero state possible. This shoot-through zero state provides the unique buck-boost feature to the inverter [17]. The advantages of Z-source inverter are listed as follows, 

  

The source can be either a voltage source or a current source. The DC source of a ZSI can either be a battery, a diode rectifier or a thyristor converter, a fuel cell stack or a combination of these. The main circuit of a ZSI can either be the traditional VSI or the traditional CSI. Works as a buck-boost inverter. The load of a ZSC can either be inductive or capacitive or another Z-Source network. Applications: 1. Renewable energy sources. 2. Electric vehicles. 3. Motor drives.

- 59 -

7.5.Battery Chargers: EV battery chargers can be classified as on-board and off board with unidirectional or bidirectional power flow. Unidirectional charging limits hardware requirements, simplifies interconnection issues, and tends to reduce battery degradation. A bidirectional charging system supports charge from the grid, battery energy injection back to the grid, and power stabilization with adequate power conversion. Typical onboard chargers limit high power because of weight, space, and cost constraints. They can be integrated with the electric drive to avoid these problems. On-board charger systems can be conductive or inductive. Conductive charging systems use direct contact between the connector and the charge inlet. An inductive charger transfers power magnetically. Fig. 51 Shows on/off board charging system and power levels for EVs. Charging may be classified to three types. Level (1) is the slowest method. It uses a standard 120V/15A single phase grounded outlet. It is used for home or business sites and low off-peak rates are likely to be available at night. It usually takes from 4-11 hours. Level (2) charging requires 240V outlet, providing ample power and it can be implemented in most environments. It takes from 4-6 hours and called semi fast charging. Level (3) charging (fast charging) is intended for commercial and public applications. It can be installed in high way rest areas and refueling points like gas stations. It typically operates with 480 V three phase supply and requires an off-board charger to provide regulated AC/DC conversion. Stations for public use usually use level (2) or (3) installed in parking lots, shopping centers, hotels, rest stops, theaters, and restaurants [18].

Fig. 51 "On/off board charging system and power levels for EVs" - 60 -

Two types of power flow are possible between EVs and the electric grid, as shown in Fig. 52. EVs with unidirectional chargers can charge but not inject energy into the power grid. These chargers use a diode bridge in conjunction with a filter and DC/DC converters. Today, these converters are implemented in a single stage to limit cost, weight, volume, and losses. High-frequency isolation transformers can be employed when desired [18].

Fig. 52 "General unidirectional and bidirectional topology" Fig. 53 shows a unidirectional full-bridge series resonant converter for a Level 1 charging system. Simplicity in the control of unidirectional chargers makes it relatively easy for a utility to manage heavily loaded feeders due to multiple EVs.

Fig. 53 "On-board unidirectional full-bridge series resonant charger for Level 1 system (3.3 kW)" A typical bidirectional charger has two stages: an active grid connected bidirectional AC/DC converter that enforces power factor and a bidirectional DC/DC converter to regulate battery current. These chargers can use non-isolated or isolated circuit configurations. When operating in charge mode, they should draw a sinusoidal current with a defined phase angle to control power and reactive power. In discharge mode, the charger should return current in a similar sinusoidal form. A bidirectional charger supports charge from the grid, battery energy injection back to the grid, referred to as vehicle-to-grid (V2G) operation mode, and power stabilization [18].

- 61 -

The topology shown in Fig. 54(a) is a non-isolated bidirectional two-quadrant charger. This circuit has two switches, which greatly simplifies the control circuitry. However, there are two high-current inductors that tend to be bulky and expensive, and it can only buck in one direction and boost in the other. The topology in Fig. 54(b) is an isolated bidirectional dual-active bridge charger. While this circuit provides high power density and fast control, the large number of components can add to cost. While most studies have focused on bidirectional power flow, there are serious challenges for adoption. Bidirectional power flow must overcome battery degradation due to frequent cycling, the premium cost of a charger with bidirectional power flow capability, metering issues, and necessary distribution system upgrades [18].

Fig. 54 "(a) Nonisolated bidirectional two-quadrant charger. (b) Isolated bidirectional dual active bridge charger" 7.5.1. Power Factor Correction (PFC) Stage: A PFC stage is usually placed between the rectifier and the DC/DC stage to avoid harmonics to the grid as well as to stabilize the DC link voltage. A typical PFC circuit, shown in Fig.55, consists of an inductor L, an active switch S, and a freewheeling diode D. Bridgeless PFC circuits have also been used in battery chargers [7].

Fig.55 "Power factor correction stage in a PHEV charger"

- 62 -

7.5.1.1.Decreasing Impact on the Grid: The AC grid side current with and without a PFC stage is shown in Fig.56, where the grid voltage is 110 V AC and the output power of the charger is 5 kW. It can be seen from Fig.56 that with a PFC, the input current is much better. The current is close to sinusoidal and the current peak is significantly decreased. The impact of harmonic currents on the grid is mitigated. Also the primary diode rectifier will undergo smaller current stress. The power factor is close to 1 [7].

Fig.56 "AC grid side current with and without PFC: (a) without PFC; (b) with a PFC"

7.5.1.2.Decreasing the Impact on the Switches: The PFC circuit also helps boost the DC link voltage to a higher level and stabilize the DC link voltage. Therefore, at the same output power, the current through the switches will be decreased to enhance safety and output capability. Fig.57a shows the switch current at different Vin with the same output power. When Vin is increased to 400 V DC, the switch current is significantly decreased. As long as the voltage across the switches does not exceed the breakdown value, higher DC bus voltage will lead to higher power capability for the same devices used. Fig.57b shows the maximum charging current that the system can deliver. Increasing the DC link voltage will benefit the output capability [7].

- 63 -

Fig.57 "Comparison of the switch current without PFC: (a) comparison of switch current; (b) Maximum charging current under different Vin" 7.5.2. Integrated Chargers: To minimize weight, volume, and cost integrating the charging function into the electric drive system has been proposed. The function can be integrated if charging and traction are not simultaneous. In an integrated charger, motor windings are used for filter inductors or an isolated transformer and the motor drive inverter serves as a bidirectional AC/DC converter. The most important advantage is that low-cost high power (Levels 2 and 3) bidirectional fast charge can be supported with unity power factor. Control complexity and extra hardware are challenges to implementation in commercial products. A combined motor drive and battery recharge system based on an induction motor is currently used by the Ford Motor Company. A non-isolated integrated charger based on a split-winding AC motor will be used in the automotive industry. There are some applications for electric scooters and two-wheeled vehicles. A typical integrated charger system is shown in Fig. 58 [18].

Fig. 58 "Typical structure of integrated PEV charger"

- 64 -

7.5.3. Inductive Charging: Conductive chargers use metal-to-metal contact as in most appliances and electronic devices. Inductive charging of EVs is based on magnetic contactless power transfer. An inductive charger transfers power magnetically. This type of charger has been explored for Levels 1 and 2 devices. The clear advantage of contactless charging is its convenience for the user. Instead of deep-cycling the battery, the vehicle battery can be topped off frequently while parked at home or at work, when shopping and even at traffic lights. Cables and cords are eliminated. Advantages include convenience and galvanic isolation. It is also possible to build charging strips into highways which enables charging while driving. Therefore, inductive charging could strongly reduce the need for a fast-charging infrastructure. Disadvantages include relatively low efficiency and power density, manufacturing complexity, size, and cost. Given that energy savings is an important motivator for EVs, the extra power loss is an important consideration. Basic principles of inductive power transfer (IPT) are similar to transformers, although most versions have poor magnetic coupling and high leakage flux. The secondary side may be stationary or moving (roadbed charging). Typical stationary IPT charging system is represented in Fig.59 [18].

Fig.59 "Typical inductively coupling EV battery charging system"

7.5.4. Wireless Charging: Wireless charging involves the use of power and energy transfer at a much longer distance. It is different from inductive charging which involves a transformer with closely placed primary and secondary windings. Although inductive charging can eliminate the direct electric contact, it still needs a plug, cable, and physical connection of the inductive coupler. Wear and tear of the plug and cable could cause danger as well. Wireless charging could eliminate the cable and plug altogether. In this scenario, a driver can pull the car over to a specially designed parking lot and the car battery is automatically charged without the pulling of any cable or plug as shown in Fig.60.

- 65 -

It provides the safest approach for EV battery charging. There have been a few different experiments carried out for wireless energy transfer. The most promising technology is using electromagnetic resonance. In this setup, there is a pair of antennas with one placed in the parking structure as the transmitter and one inside the car as the receiver. The two antennas are designed to resonate at the controlled frequency. The limitations are the level of power transfer, and efficiency due to the large air gap between the two antennas [7].

Fig. 60 "Wireless charging of a PHEV/EV on a parking floor"

The successful deployment of EVs over the next decade is dependent on the following: 1) Charger reliability, durability, and safety considerations will contribute to consumer acceptance of EVs. 2) Charger efficiency and reducing charger costs. 3) Suitability for V2G-bidirectional power flow, communication, and metering. 4) Charging systems that can accommodate high-power charging will provide more flexibility and choices to the consumer. 5) Charging strategies and setting limits for charging time and access rules. 6) The ease of use of the charger and connector, and how user friendly it is perceived to be by the consumer contributes to the development of a wider market for EVs and acceptance of the technology.

- 66 -

7.6. Challenges and Problems facing EV Power Electronics: 7.6.1. Power Density and Specific Power: One critical requirement for power electronic converters in automotive applications is to minimize their size and weight. Carrying too large or too heavy power converters on board will result in space reduction of passenger compartment and decrease in fuel economy. Hence, two important metrics, i.e. power density and specific power are considered. Typically, the size and weight of power converters are related to their power rating. Hence, power density is defined as ratio of converter volume to converter power, and specific power is the ratio of weight over power. The specific power and power density of each type of converters and power control unit (PCU) modules are listed in Table 5 for four HEV models, i.e. 2010 Prius, 2004 Prius, 2007 Camry, and 2008 LS600h [5]. The PCU modules from four HEV models are also compared side by side in Fig. 61. As can be seen, the 2008 Lexus LS 600h has the most compact design among the models compared, with an 8.03 kW/L power density and 6.15 kW/kg specific power. Its sub-modules such as DC/DC converters and motor inverters also have the highest power density and specific power. This can be partially due to the relative large power rating of LS 600h PCU (110 kW), which is also the largest among all. As an example, Fig. 62 shows the compartments of the 2010 Prius PCU. As can be seen, in order to achieve high power density and specific power, an optimized packaging layout solution is the key point on top of others [5].

Table5 "Power Density and Specific Power of Converters"

Fig. 61 "Comparison of PCU module from four HEV model" - 67 -

Fig. 62 "Compartments of 2010 Prius PCU" 7.6.2. Electromagnetic Interference (EMI): Electromagnetic interference has been recognized as a difficult issue, especially for IGBT inverters due to its extremely fast switching characteristics. Fast transient in voltage and current causes the conducted and radiated noise which in turn may cause malfunction of near-by electronic devices. Fast switching is also reported to cause failures in motor windings and or bearings. Due to the complexity of EMI problem and the fact that automotive power electronics is just in its early development stage, a great deal of research and development is needed. In order to minimize EMI, components must be carefully placed so that EMI is not contained by shielding and will have minimal effect on the rest of the system. All paths must be kept as close as possible so that the generated fields will nullify one another. To minimize parasitics and aid in the EMI issue, the lengths of wires need to be kept as short as possible. The control circuit needs to provide protection for overcurrent, short circuit, overvoltage, and under-voltage. The capability to detect any fault signal and turn off the gate drive signals to the primary switches is a critical part of power electronic circuit design. Fast fuses need to be used in the circuit to protect the converter from being damaged by any other faults and used for safety [7]. 7.6.3. Application of SiC and GaN Devices: The present silicon (Si) technology is reaching the material’s theoretical limits and cannot meet all the requirements of hybrid vehicle applications in terms of compactness, light weight, high power density, high efficiency, and high reliability under harsh conditions. New semiconductor materials, such as wide bandgap (WBG) semiconductor material "Silicon Carbide (SiC) and Gallium Nitride (GaN)", for power devices have the potential to eventually overtake Si power devices in hybrid vehicle powertrain applications. SiC power devices potentially have much smaller switching and conduction losses and can operate at much higher temperature than comparable Si power devices. - 68 -

Hence, a SiC-based power converter will have a much higher efficiency than that of converters based on Si power devices if the same switching frequency is used. Alternatively, a higher switching frequency can be used to reduce the size of the magnetic components in a SiC-based power converter. In addition, because SiC power devices can be operated at much higher temperatures without much change in their electrical properties, ease of thermal management and high reliability can be achieved. On the other hand, by increasing the switching frequency of WBG converters significantly, the considerable reduction in filter size will be resulted. Fig. 63 and Fig.64 shows the DC/DC converter loss and efficiency as a function of switching frequencies using three different device combinations. Corresponding converter size and weight as a function of switching frequency is also shown is Fig. 64. Fig. 65 shows the whole PCU size comparison between Si technology and SiC technology. It is also claimed that SiC PCU can achieve 90% loss reduction compared to Si PCU [5].

Fig. 63 "Power loss on converters of various switching frequencies under full load condition"

Fig. 64 "Weight and volume of the converters as a function of switching frequency" - 69 -

Fig. 65 "Size comparison of Si PCU (left) and SiC PCU (right)"

7.6.4. Thermal Management: Compared with industrial applications, automotive power electronic systems have to work in a more stringent environment. The ambient temperature in an engine compartment can vary from -400C to 1050C. The temperature between cold engine and hot engine conditions can be as high as 80°C. These temperature conditions exert great thermal stress on components such as power semiconductor modules and electrolytic capacitors. In addition, severe vibration in a road vehicle demands close attention to the design of the connectors and fasteners; salt and water spray requires sealed package for power electronic systems. At power levels of 100 kW, even with an efficiency of 96–98%, the power losses of each power electronic unit is 2–4 kW. With two or three powertrain motors and associated power electronics circuits, together with the high-power bidirectional DC/DC converter, the heat generated in the hybrid vehicle system could be significant. Significant advances in the thermal management of both power electronics and motors for HEV propulsion system must be achieved to meet the automotive industry’s goals of reduced weight, volume, and cost. Through the optimization of existing technologies and the expansion of new cooling methods, higher power densities, smaller volumes, and increased reliabilities can be realized in the hybrid powertrain components. The main areas of concern in thermal management of power electronics are: operating temperature of IGBTs (it should be less than 1250C); contact resistance between various layers of a power module; low-thermal-conductivity thermal paste; heat flux limitations (ideally, faster IGBTs would have to reject heat at a rate of 250 W/cm2); limitations on the inlet cooling fluid temperature (it is desirable to use the engine coolant at 1050C); the cost of the cooling system; weight and volume [7].

- 70 -

8. Conclusions: Automotive power electronics is a fast growing area both as a technology and as a business. It holds the key to the future cleaner and more energy efficient transportation system. The cost is the most critical challenge. Combining forces of auto industry and other power electronics industries can be a very effective way to overcome this challenge. Such collaboration will not only benefit the auto industry, but also the existing power electronics industries. In the area of propulsion motor and other motor control technologies, EV motors are compared and evaluation is made to judge their suitability for use in EVs and HEVs. Development of low cost and high temperature magnets would lead to wide spread use of permanent magnet motors. PM motors have higher efficiency and need lower current to obtain the same torque than other machines. This would reduce the cost of power devices also. Methods to eliminate the speed/position sensors, inverter current sensors, etc. have been under investigation for several years. These technologies have not yet proven to be practical for automotive applications. Electronic controllers are discussed and recent developments are outlined. The controllers need to be developed for robust operation. The energy storage systems are also discussed with emphasis on the crucial factors affecting their cost such as energy density is explained. Comparison between various batteries that are used in the recent EVs is given. Continued reductions in the cost of batteries along with the maturation of different battery chemistries and technologies that can improve performance are crucial to the growth of the industry. As Li-Ion improvements saturated, the field will turn to other solutions, perhaps Li-Air or Li-Metal, to continue improvements. Improvements in energy storage are not only important for EVs, but will also enable other green technologies, contributing to a more sustainable environment. The demand of electrical power in vehicular application is growing rapidly as the mechanical components are being replaced by the electrical and electronic components. Important issues related to power electronic converters as heart of electric propulsion system are presented. To study the responsibility of power electronics converters, architecture of different electrified vehicles has been reviewed. The present status and future challenges of power electronic converters as well as electric propulsion system are discussed briefly. The characteristics and uniqueness of EV drives has been identified, and an overview on the development of power electronic drives for EV propulsion, including DC/DC converters, DC/AC converters, and battery chargers has been given. Recent research on EV motors and EV power converters have also been discussed, with emphasis on the soft-switching converters for DC motor drives, soft-switching inverters and multilevel inverters for AC motor drives.

- 71 -

When a DC/DC converter is added to the electric propulsion system between battery and inverter, the voltage of the DC-bus can be maintained constant and the battery current can be maintained with relatively small ripples. DC/DC converters can be found in many places in the EVs. First, some types are placed after the DC-bus to provide the suitable low voltage for light loads such as lighting, microprocessor, and luxury. Second types are placed between the energy storage system and the DC-bus to boost the voltage level to be suitable for motor propulsion. If the traction motor is a DC motor, other DC/DC converters can be found. The different types of these DC/DC converters are discussed showing their characteristics, advantages, and disadvantages in order to select the most suitable ones for EVs and HEVs. These converters include the pulse-width-modulated (PWM), resonant, quasi-resonant, multi-resonant, and non-linear resonant switch converters. Recently, soft-switched PWM converters have been gaining popularity. The DC/DC converter is a part of the complete system of an electric or hybrid vehicle. As such, its design affects and is affected by the designs of other components. The effects of various components on each other in terms of the electrical performance packaging, and thermal management should be taken into account to optimize the vehicle system for high performance, high reliability, safety, and low cost. Isolated-bidirectional DC/DC topologies provide safety and reduce noise because of the galvanic isolation. Full bridge DC/DC topologies have the advantages of reduced voltage stress across switches, equal current stress of all switches, no need for additional elements to have soft switching, and fast dynamic behavior, but the current has high ripple content, the control is very sensitive, and they suffer from high component count. Half bridge DC/DC topologies have the advantages of low switch count, simple control, and low ripple content in current, but they suffer from unbalanced current stress across switches. Interleaved DC/DC converters have a lot of merits over the other types of converters. The equally sharing current between each phase provides tight size of power semiconductors, distribution of losses between modules and size optimization of the converter, thus the efficiency is improved making it suitable for EV and high power applications. The development of inverters for AC motor drives has become one of the key research areas in electric propulsion. The wish list of those inverters include efficiency over 95%, power density over 3.5 W/cm3, switching frequency over 10 kHz, dv/dt below 1000 V/µs, zero EMI, and redundancy with limp-home mode. To achieve this wish list, there are two major research directions, namely the softswitching inverters and the multilevel inverters.

- 72 -

Although soft switching converters have the advantage of lower switching losses and low EMI, they need more components, higher operating voltage devices (depending on the topology), and complicated control compared to hard-switched converters. Hence the soft-switched converter application is limited to very special types of needs. There is a need to develop a converter topology to achieve the performance of the soft-switched converter with less components and simplified control. The general function of multilevel inverter is to synthesize a desired voltage from several levels of DC-voltages. Therefore, they can easily provide the high power required for a large EV drive. Multilevel inverters can generate near sinusoidal voltages with only fundamental frequency switching, have almost no EMI, very small THD, and no common mode voltage. The T-type five-level converter has shown better characteristics compared to the 2-level inverter as well as 5-level DCC in terms of losses and harmonics. This comes from the reduced switching elements of the converter power circuit. The Z-source inverter employs a unique LC network to couple the inverter main circuit to the diode front end. The Z-source can produce any desired output AC voltage, even greater than the line voltage. Battery infrastructure and charging power levels are categorized into three types: Level 1, Level 2, and Level 3. Charger systems are categorized into off-board and onboard types with unidirectional and bidirectional power flow. Unidirectional charging limits hardware requirements, simplifies interconnection issues, and tends to reduce battery degradation. Bidirectional charging supports battery energy injection back to the grid. Typical on-board chargers restrict power to meet weight, space, and cost constraints. There is a possibility of avoiding these problems by using the electric drive system as an integrated charger. The most important advantage of integrated chargers is that low-cost high-power (Levels 2 and 3) bidirectional fast charging with unity power factor is supported. On-board charger systems can be conductive or inductive. Inductive charging has the long-term promise of supporting active roadbed systems. Various charger power levels and infrastructure configurations were presented and compared, based on the amount of power, charging time and location, cost, and suitability. Success of EVs depends on standardization of requirements and infrastructure decisions, efficient and smart chargers, and enhanced battery technologies. A PFC stage is usually placed between the rectifier and the DC/DC stage to avoid harmonics to the grid as well as to stabilize the DC link voltage.

- 73 -

One critical requirement for power electronic converters in automotive applications is to minimize their size and weight. Hence, two important metrics, i.e. power density and specific power are considered. Typically, the size and weight of power converters are related to their power rating. In order to achieve high power density and specific power, an optimized packaging layout solution is the key point. In order to minimize EMI, all paths must be kept as close as possible so that the generated fields will nullify one another. To minimize parasitics and aid in the EMI issue, the lengths of wires need to be kept as short as possible. Fast fuses need to be used in the circuit to protect the converter from being damaged by any other faults and used for safety. Integrated EMI filters for control of EMI generated due to switching of the devices needs to be part of the inverter /converter topology. The research on WBG devices needs to be accelerated to enable their application to high power switching devices at higher operating temperatures. The devices and the rest of the components need to withstand thermal stress and extreme vibrations. Fault tolerant topologies and control techniques need further investigation. The system needs to be fault tolerant and provide limp-home capability. The challenges in reliability and EMC can be addressed through technical innovation. Auto industry's experience in quality and reliability lends itself very well to the development of automotive power electronics products, which can benefit power electronics industry tremendously.

- 74 -

9. References: [1] L. Kumar, K. K. Gupta, and S. Jain, "Power Electronic Interface for Vehicular Electrification" Industrial Electronics (ISIE), 2013 IEEE International Symposium, 2013. [2] K. T. Chau and Z. Wang, "Overview of power electronic drives for electric vehicles", HAIT Journal of Science and Engineering B, vol. 2, no. 5, June 2005, pp. 737-761. [3] X. Xingyi, "Automotive power electronics pportunities and challenges , Proc. Int. Conf. Electr. Mach. and Drives, IEEE, 1999, pp. 260-262. [4] C.C. Chan and K.T. Chau, "An overview of power electronics in electric vehicles", IEEE Trans. on Industrial Electronics, vol. 44, no. 1, February 1997, pp. 313. [5] B. Sarlioglu, Casey T. Morris, Di Han, Silong Li, "Benchmarking of Electric and Hybrid Vehicle Electric Machines, Power Electronics, and Batteries" Electrical Machines & Power Electronics (ACEMP), 2015 Intl Conference on Optimization of Electrical & Electronic Equipment (OPTIM) & 2015 Intl Symposium on Advanced Electromechanical Motion Systems (ELECTROMOTION), IEEE, 2015, pp.519-526. [6] G. Nanda and N.C. Kar, "A survey and comparison of characteristics of motor drives in electric vehicles", IEEE CCECE, 2006, pp. 811-814. [7] Mi, M. A. Masrur and D. Gao, "Hybrid Electric Vehicles—Principles and Applications with Practical Perspectives", 2011, Wiley. [8]A. Emadi, S. S. Williamson and A. Khaligh, "Power electronics intensive solutions for advanced electric hybrid electric and fuel cell vehicular power systems", IEEE Trans. Power Electron., vol. 21, no. 3, May 2006, pp. 567-577. [9] A. Khan, "DC-to-DC Converters for Electric and Hybrid Vehicles", Proceedings of the IEEE Workshop on Power Electronics in Transportation, 1994, pp. 113-122. [10] D. M. Bellur and M. K. Kazimierczuk, "DC-DC converters for electric vehicle applications", Electrical Insulation Conference and Electrical Manufacturing Expo,2007 IEEE, pp. 286-293. [11] H. R. Karshenas, H. Daneshpajooh, A. Safaee, P. Jain and A. Bakhshai, "Bidirectional DC-DC converters for energy storage systems", Chapter 8 in energy storage in the emerging era of smart grids INTECH Open Access Book, Sept. 2011. [12] M. Elsied et al., "Efficient Power-Electronic Converters for Electric Vehicle Applications," Vehicle Power and Propulsion Conference (VPPC), 2015 IEEE, Montreal, QC, 2015, pp. 1-6.

- 75 -

[13] T.W. Ching, "Soft-Switching Converters for Electric Propulsion", Journal of Asian Electric Vehicles, 2007, pp. 1019-1026.

Vehicle

[14] H. van Hoek, M. Boesing, D. van Treek, T. Schoenen and R. W. De Doncker, "Power electronic architectures for electric vehicles," Emobility - Electrical Power Train, 2010, Leipzig, 2010, pp. 1-6. [15] M. D. Bellar, T. S. Wu, A. Tchamdjou, J. Mahdavi and M. Ehsani, "A review of soft-switched DC-AC converters," in IEEE Transactions on Industry Applications, vol. 34, no. 4, July-Aug. 1998, pp. 847-860. [16] L. M. Tolbert, F. Z. Peng and T. G. Habetler, "Multilevel inverters for electric vehicle applications," Power Electronics in Transportation, 1998, Dearborn, MI, 1998, pp. 79-84. [17] PENG, Fang Zheng. "Z-source inverter" IEEE Transactions on industry applications, 2003, 39.2: 504-510. [18] YILMAZ, Murat; KREIN, Philip T. "Review of battery charger topologies, charging power levels, and infrastructure for plug-in electric and hybrid vehicles", IEEE Transactions on Power Electronics, 2013, 28.5: 2151-2169.

- 76 -

Suggest Documents