Design and Construction of a Low Cost Photovoltaic Generator for Laboratory Investigations M. A. Elgendy1, V. Meenakshi Sundaram and B.U. Surati School of Electrical and Electronic Engineering - Newcastle University 1
[email protected]
insolation and temperature levels when adopting the above method. Another approach for evaluating MPPT algorithms is to use a PV array emulator [2, 6, 7]. The PV array emulator is usually a dc chopper that can be programmed with a set of pre-calculated I-V curves as lookup tables for two time series inputs of solar irradiance and cell temperature [8, 9]. The main drawback of the PV array emulators is the slow response [8]. A faster response time has been achieved at the expense of the quality of the voltage response of the emulator [10]. Because of their slow response, PV array emulators may only be suitable for testing MPPT controllers operating at low sampling rates. MPPT algorithms can also be evaluated by using a low power rating converter connected to a real PV array with a controllable light source that emulates sun rays [11, 12]. Such a controllable light source is referred to as sun simulator or solar simulator. This approach is more suitable for laboratory investigations. This type of PV generator can also be used to demonstrate the output characteristics of PV arrays at different solar irradiance and cell temperature levels. In addition, it can be used to demonstrate the different operation modes of standalone PV systems comprising energy storage.
Abstract— The dependence on weather conditions of photovoltaic (PV) systems coerces a development of an indoor PV generator using sun-rays simulator to study the output characteristics of the solar panel and to evaluate maximum power point tracking (MPPT) algorithms. In this paper, a 40Wp low cost PV generator is constructed using a solar panel illuminated by switched halogen lamps. The output characteristics of the PV generator are investigated under different panel illumination levels. Both constant voltage and perturb and observe (P&O) MPPT control schemes are examined. The different operation modes of standalone PV applications with battery backup are emulated utilizing a dc-dc converter based power conditioning unit with a resistive and battery loads. Keywords— Photovoltaic generator, Standalone PV systems, Power conditioning, Battery storage, Maximum power point tracking.
I. INTRODUCTION The current-voltage (I-V) curves of a solar module/array at different solar irradiance and cell temperature levels are shown in Figs. 1 and 2, respectively. As shown, at constant cell temperature, the short circuit current is proportional to the insolation level while the open circuit voltage increases slightly with increasing insolation. The open circuit voltage has a negative temperature coefficient while the short circuit current has a positive temperature coefficient. The PV generator can be approximated to a current source if it is operated near its short circuit condition and can be approximated to a voltage source if operated near its open circuit condition. At the knee of the curve where the maximum power point (MPP) located, the PV generator cannot be accurately approximated to either type of source. The PV generator must be operated in the area of MPP to make use of the maximum possible generated power. This can be achieved by using an MPP tracker normally comprised of a simple dc-dc converter as an interface between the PV generator and the load. The duty ratio of the converter is then controlled by an MPPT algorithm. A number of different MPPT algorithms have been proposed in the literature for a large number of potential applications. These algorithms have been reviewed in a number of review papers [1-3]. MPPT algorithms can be evaluated by connecting the converter to an appropriate PV generator and measuring the PV generator output power. This power is then divided by the maximum possible output power of the PV generator calculated at the same solar irradiance and cell temperature values. In this case, an accurate model must be used to represent the utilized PV generator [4, 5]. Because the weather conditions cannot be controlled, it is difficult to evaluate the algorithm at different
978-1-4673-4430-2/13/$31.00 ©2013 IEEE
Fig. 1. Influence of solar irradiance on the I-V characteristics of a solar module at constant cell temperature
Fig. 2. Influence of cell temperature on the I-V characteristics of a solar module at constant solar irradiance
79
Solar simulators that closely simulate the sun’s spectra with good uniformity and high level of irradiance are available on the market. However, the high cost of these simulators limits their use to PV module testing facilities of line production and the tests that require precise matching to the solar spectrum. For most of laboratory demonstrations/ investigations with PV generators, maintaining the exact solar spectrum is not of high importance. These include studying PV generator characteristics, testing MPPT algorithms and investigating the different operation modes of standalone PV systems [13, 14]. In these cases, a PV array with variable illumination facility that ensures an acceptable degree of uniformity is more than adequate. In this paper, a low cost PV generator with manually controlled halogen lamp based illumination is designed and constructed. The output characteristics of the generator are measured at different illumination levels. A power conditioning unit based on synchronous buck converter is constructed and used for MPPT control and for demonstrating the different operation modes of a standalone PV system with battery storage. Both constant voltage and perturb and observe MPPT algorithms are investigated.
the mains. They have also acceptable start-up and transient response speed. Halogen bulbs operate at high temperatures, thus their lamp holders and connection wires must be carefully chosen to cope with these high temperatures. Cooling fans should also be used so that the solar panel is not affected by such high temperatures. The PV generator adopted in this paper comprised a BP Solar BP-140 40Wp PV array illuminated by 60 low cost halogen lamps rated 240V, 50W. Each lamp generates 700lumen. The number of lamps is decided so that the maximum light intensity on the PV array is about 1000W/m2. These lamps are divided into two overlapped 13A mains-fed groups. Fig. 3 illustrates electrical distribution wiring for these lamps. As shown, each mains-fed group consists of five branches of six lamps that can be switched on and off separately to provide various PV panel illumination levels. Therefore illumination can be varied in ten different levels to study panel I-V characteristics. All wiring is done using high temperature cables and terminal blocks and the frame box is earthed to avoid any shock hazard. A power diode is connected at the output terminal of the panel to prevent reverse current flow to the PV array. To extract the heat generated by the lamps, forced fan cooling is provided using four ac fans. Each two of these fans are connected to one of the two 13A mains supply. Moreover, to avoid any high temperature hazard, two thermal switches are connected in series with the mains ac inputs to switch off all the lamps if temperature exceeds 67°C. The lamps are fixed to the inner side of the top cover of an aluminium box in which the PV panel is fixed at the bottom plane as shown in Fig. 4. The four fans are fixed at one side of the box facing holes on the opposite side of the box. These holes are drilled to allow flow of air for cooling the solar panel. The inner sides of the aluminium box are of the reflector type to ensure uniform illumination distribution. The box is made up of high temperature aluminium so that it can withstand the heat generated by the lamps.
II. PV GENERATOR DESIGN Three main factors are considered in the choice of an illumination source for the solar panel utilized in the PV generator under consideration. Firstly, the initial cost of the illumination source and the setup used in varying illumination level should be taken into account. Secondly, the start-up time and the response to changing illumination level should be as fast as possible in order to allow testing fast MPPT algorithms. Finally, the efficiency of the illumination source should be high enough to allow adequate illumination within the limited space area of the utilized PV panel. The rest of the electrical and optical characteristics of illumination sources are secondary in this paper. High-brightness light-emitting diodes (HB-LEDs) have long lifetime and high luminous efficacy of up to 100lm/W or even higher [15, 16]. They have also fast transient response compared to other types of illumination sources. However, HB-LEDs cannot be connected directly to the mains. To provide sufficient light intensity, they must be grouped together and connected to the mains via converters that fit their characteristics. In addition, the high initial cost of HB-LEDs makes them unattractive option for this project. High pressure sodium lamps have high efficiency and long lifetime but their poor transient characteristic makes them unsuitable for this application [16, 17]. Halogen bulbs are less expensive to purchase and they can be fed directly from
L2
L1
Fig. 3. Lamp wiring diagram
1-Solar Module 2-Halogen Lamps 3-Light Switches 4-Cooling Fans 5-Aluminium Box (Reflector Type)
2
5 4 3
1 Fig. 4. Different views of the PV generator
80
III.
POWER CONDITIONING UNIT
analogue and digital circuits. The algorithm perturbs the operating point of the PV generator by increasing or decreasing a control parameter (K) by a small amount and measures the PV array output power before and after the perturbation. If the power increases, the algorithm continues to perturb the system in the same direction; otherwise the system is perturbed in the opposite direction (Fig. 10). In this paper, the duty ratio of the MPPT converter is used directly as the control parameter.
A power conditioning circuit is designed and constructed to test MPPT algorithms and the different operation modes of standalone PV systems. The circuit is basically a synchronous buck converter with two parallel loads; 6V-5Ah lead-acid battery and variable resistive load. The control algorithm connects/disconnects these loads, independently, based on the operation mode of the system. This is achieved by activating/deactivating a solid-state relay for each load. The converter employs MOSFET switches with very low on state resistance (28mȍ). This increases the overall efficiency of the converter especially when employed in the synchronous mode. Voltages are measured using simple non-isolated voltage divider circuits with buffered output while currents are measured using Hall-Effect sensors. The Amicus18 development board employing PIC18F25K20 microcontroller is used for system control. A four-channel oscilloscope with voltage and current probes is used in the measurements. Simplified circuit diagram of the power conditioning unit is shown in Fig. 5.
iPV
il
T1
S2
L
iL
S1 PV Gen.
vPV
CL
iPV
C
T2
S 1 S2
T1 T2
R
vb
vb
vPV MCU Based MPPT and Operation Control
iL
Fig. 5. Simplified circuit diagram of the power conditioning unit
IV. PV ARRAY CHARACTERISTICS
3.5
The I-V curves of the PV generator are measured using a variable resistor connected directly to the array. The resistance is varied manually from zero to its maximum value and the array’s current and voltage are recorded with high accuracy digital multimeters. The resistance is then disconnected and the open circuit voltage is measured. Measurements should be started after few seconds of each variation in illumination level to allow Cell temperature to reach a steady state. Cell temperature can be controlled by varying the number of ac fan in duty. Fig. 6 and Fig. 7 show the measured I-V and P-V characteristics of the PV generator at two different illumination levels when all the ac fans are operating. Lower maximum power is obtained from the PV generator due to light leakage from the fan and cooling openings and due to the old age of the utilized PV array. The effect of cell temperature on the I-V characteristics of the array is demonstrated in Fig. 8. Higher cell temperature results in lower open circuit voltage and lower maximum power output of the array.
60% Illumination 100% Illumination
Array Current (A)
3 2.5 2 1.5 1 0.5 0 0
5
10 Array Voltage (V)
15
20
Fig. 6. Measured I-V characteristics at different illumination levels 35
Array Power (W)
30
100% Illumination 60% Illumination
25 20 15 10 5 0 0
V. MAXIMUM POWER POINT TRACKING Two MPPT algorithms are examined in this paper; constant voltage algorithm and P&O algorithm. With constant voltage MPPT algorithm, the PV array is operated at constant voltage equal to the MPP voltage of the array at the standard test conditions which is usually provided by the manufacturer. This value is used as a reference for a feedback control loop that usually employs a PI controller to adjust the duty ratio of the MPPT converter as shown in Fig. 9. Constant voltage MPPT algorithm ignores the effects of insolation and temperature variations on the MPP voltage. However, it is simple and very cheap to implement which make it more suitable for low power applications. Higher energy utilization efficiency can be achieved (at the cost of a small increase in implementation cost) by employing P&O MPPT algorithm. This is a simple algorithm that does not require previous knowledge of the PV generator characteristics or the measurement of solar intensity and cell temperature and is easy to implement with
5
10 Array Voltage (V)
15
20
Fig. 7. Measured P-V characteristics at different illumination levels 3.5
At Lower Cell Temperature At Higher Cell Temperature
Array Current (A)
3 2.5 2 1.5 1 0.5 0 0
5
10 Array Voltage (V)
15
20
Fig. 8. Measured I-V characteristics at different cell temperature levels
Vref -
PI Controller
D
PV System
VPV
+ VPV
Fig. 9. Block diagram of constant voltage MPPT control
81
step size is high. The relatively high step sizes used to achieve three-level operation result in high steady state oscillations in system waveforms. Steady state oscillations can be reduced by using lower step sizes. However, this slows down the starting transient of the MPPT algorithm as well as its response to irradiance and temperature variations. The slow transient response can be compensated for by using a higher perturbation rate. If the perturbation is increased so that the sampling period becomes shorter than the settling time of the system response, the system will never reach a steady state. There is always trade off in the choice of perturbation rate and step size. Fig. 12 shows array voltage and current responses to a step increase in illumination level when employing P&O algorithm. A low perturbation frequency and high step size were used to explore the threelevel operation of the algorithm. These values are not necessarily the optimum values that should be used with a practical implementation.
Start Initialize
Measure; VPV(T) and IPV(T) PPV(T) = VPV(T) × IPV(T) ǻPPV(T) = PPV(T) í PPV(Tí1) No
Yes
ǻPPV(T) >= 0
ǻK(T) = íǻK(Tí1)
ǻK(T) = ǻK(Tí1)
K(T) = K(Tí1) + ǻK(T) PPV(Tí1) = PPV(T), K(Tí1) = K(T)
Fig. 10. Flowchart of P&O MPPT algorithm
VI. OPERATION MODES OF STANDALONE PV SYSTEMS The energy available from a PV array varies during the day time according to weather conditions. Also PV arrays cannot supply power to the load during night time. Therefore, storage batteries are required to store the excess energy available during periods with higher power generation to supply the load during down time. Depending on the energy available from the PV array, a standalone PV system with battery storage operates in four different modes as follows [13, 14]:
PV Array Voltage
PV Array Current
A. Bulk Charging mode In this mode, maximum power output from the PV generator is higher than load demand. The excess power is used to charge the battery. With Bulk Charging mode, the PV array should be operated at/near the MPP to extract the maximum possible power from the solar array.
Fig. 11. Response of constant voltage MPPT to a step increase in illumination level (from 60% to 100% illumination)
B. Trickle Charging mode This mode is also known as Floating mode as the battery voltage floats around its maximum voltage rating. Power output from the panel is still higher than the load but now the battery is charged to its full capacity. MPPT algorithm will be stopped and the array generates only the power required to feed the load and compensate for battery losses.
PV Array Voltage
PV Array Current
Fig. 12. Response of P&O MPPT to a step increase in illumination level (from 60% to 100% illumination)
C. Dual mode: During the period of low solar irradiation when the power available from PV array is lower than the load demand stored energy in the battery is used to supply the load deficit. The MPPT algorithm will extract maximum power from the panel and remaining power will be supplied by battery. This mode is also known as Partial Discharge mode.
Fig. 11 shows the response of constant voltage MPPT control to a step change in illumination level. The reference voltage is set at 16V while the illumination level is changed from 60% to 100% of its maximum value. As shown, array current increases to match the new irradiance level while array voltage is kept constant at its reference value. With P&O algorithm, system waveforms fluctuate around their MPP values even if solar irradiance and cell temperature are constants. These usually fluctuate between three levels in the steady state when the perturbation frequency is low and the
D. PV Off mode During night time when no power is available from the panel, stored energy in the batteries is used to supply the load. The size of the batteries should be sufficient to avoid any load interruption. These different operation modes of standalone PV system are emulated by the proposed system. Bulk Charging mode is emulated when the battery has low state of charge and panel illumination is set at its maximum. As shown in Fig. 13, the excess energy generated by the PV array than the
82
load demand is used to charge the battery (battery current is positive). For Floating mode, the state of charge of the battery should approach 100%. In this case, to avoid overcharging the battery, MPPT control must be stopped. The operating point of the array will be set so that the array feeds the load and compensates for battery losses. Fig. 14 shows that array voltage in this mode is much higher than the MPP voltage. Array power is wasted in order to protect the battery and the load. Dual operation mode can be emulated when the state of charge of the battery is high and the illumination level is low. In this case, the shortage in load demand which cannot be supplied by the PV array is supplied by the battery as shown in Fig. 15 (battery current is negative). For PV Off mode, all load power is supplied from the storage battery.
VII. CONCLUSIONS A low cost PV generator for indoor use was designed and constructed in this paper. Halogen lamps were chosen as illumination source for the utilized PV panel as a tradeoff between cost, transient characteristics and energy efficiency. The output characteristics of solar array were examined. The constant voltage and P&O MPPT algorithms were employed and the different operation modes were explored using a power conditioning unit with battery and resistive loads. REFERENCES [1]
[2] [3]
PV Array Voltage PV Array Current
[4]
Battery Current
[5]
[6]
Load Current
[7] Fig. 13. Bulk Charging operation mode [8] PV Array Voltage
[9]
PV Array Current
[10] Battery Current
[11] Load Current
[12]
Fig. 14. Floating operation mode
[13]
[14]
PV Array Voltage PV Array Current
[15]
[16]
Battery Current Load Current
[17]
Fig. 15. Dual operation mode
83
M. A. de Brito, L. Galotto, Jr., L. P. Sampaio, G. de Azevedo e Melo, and C. A. Canesin, “Evaluation of the main MPPT techniques for photovoltaic applications,” IEEE Trans. on Industrial Electronics, vol. 60, pp. 1156–1167, 2013. T. Esram and P. L. Chapman, "Comparison of photovoltaic array maximum power point tracking techniques," IEEE Trans. on Energy Conversion, vol. 22, pp. 439-449, 2007. D. P. Hohm and M. E. Ropp, "Comparative study of maximum power point tracking algorithms," Progress in Photovoltaics: Research and Applications, vol. 11, pp. 47-62, 2003. M. A. Elgendy, B. Zahawi and D. J. Atkinson, "Assessment of Perturb and Observe MPPT Algorithm Implementation Techniques for PV Pumping Applications,’ IEEE Trans. on Sustainable Energy, vol. 3, no. 1, pp. 21-33, 2012 M. A. Elgendy, B. Zahawi, and D. J. Atkinson, "Comparison of Directly Connected and Constant Voltage Controlled Photovoltaic Pumping Systems," IEEE Trans. on Sustainable Energy, vol. 1, pp. 184-192, 2010. A. Pandey, N. Dasgupta, and A. K. Mukerjee, "High-performance algorithms for drift avoidance and fast tracking in solar MPPT system," IEEE Trans. on Energy Conversion, vol. 23, pp. 681-689, 2008. H. Nagayoshi, "I-V curve simulation by multi-module simulator using I-V magnifier circuit," Solar Energy Materials and Solar Cells, vol. 82, pp. 159-167, 2004. H. Matsukawa, K. Koshiishi, H. Koizumi, K. Kurokawa, M. Hamada, and L. Bo, "Dynamic evaluation of maximum power point tracking operation with PV array simulator," Solar Energy Materials and Solar Cells, vol. 75, pp. 537-546, 2003. P. Sanchis, J. Lopez, A. Ursua, and L. Marroyo, "Electronic controlled device for the analysis and design of photovoltaic systems," IEEE Power Electronics Letters, vol. 3, pp. 57-62, 2005. S. H. Lloyd, G. A. Smith, and D. G. Infield, "Design and construction of a modular electronic photo-voltaic simulator," IEE Conference Publication, pp. 120-123, 2000. W. T. Chee, T. C. Green, and A. H.-A. Carlos, "Analysis of perturb and observe maximum power point tracking algorithm for photovoltaic applications," PECon 2008 - 2008 IEEE 2nd International Power and Energy Conference, pp. 237-242, 2008. C. W. Tan, T. C. Green, and C. A. Hernandez-Aramburo, "A currentmode controlled maximum power point tracking converter for building integrated photovoltaics," 2007 European Conference on Power Electronics and Applications, EPE, pp. 1-10, 2007. Z.M. Salameh, B.S. Borowy, “Methodology for optimally sizing the combination of a battery bank and PV array in a wind/PV hybrid system,” IEEE Trans. on Energy Conversion, vol. 11, pp. 367–375, 1996. J. Chiang, H.-J. Shieh, and M.-C. Chen, “Modeling and control of PV charger system with SEPIC converter,” IEEE Trans. Ind. Electron., vol. 56, no. 11, pp. 4344–4353, Nov. 2009. M. Arias, D. G. Lamar, J. Sebastián, D. Balocco, A. A. Diallo, “High-Efficiency LED Driver Without Electrolytic Capacitor for Street Lighting,” IEEE Trans. on Industry Applications, vol. 49, no. 1, pp. 127-136, 2013. R. A. Pinto, M. R. Cosetin, T. B. Marchesan, M. Cervi, A. Campos, R. N. do Prado, “Compact Lamp Using High-Brightness LEDs", Industry Applications Society Annual Meeting, 2008. IAS '08. IEEE, pp. 1 - 5, Oct. 2008. A. Martín, N. Bordel, C. Blanco, J. C. A. Antón, and G. Zissis, “Comparison of the Emission of a High-Pressure Sodium Lamp Working at 50 Hz and at High Frequency,” IEEE Trans. on industry applications, vol. 46, no. 5, pp. 1740-1745, 2010.
