Implemental Control Strategy for Grid Stabilization of ... - IEEE Xplore

IEEE TRANSACTIONS ON ENERGY CONVERSION, VOL. 28, NO. 3, SEPTEMBER 2013

619

Implemental Control Strategy for Grid Stabilization of Grid-Connected PV System Based on German Grid Code in Symmetrical Low-to-Medium Voltage Network Youngsang Bae, Trung-Kien Vu, and Rae-Young Kim, Member, IEEE

Abstract—In the last couple of years, the increasing penetration of renewable energy resulted in the development of grid-connected large-scale power plants. However, a high penetration harbors the risk of grid instability if the generating power plants are not able to support the grid. Therefore, grid stabilization, which depends on the system-type or grid of each country, plays an important role and has been strengthened by different grid codes. With this background, VDE-AR-N 4105 for photovoltaic (PV) systems connected to the low-voltage grid and the German Association of Energy and Water Industries (BDEW) introduced the medium-voltage grid code for connecting power plants to the grid and they are the most stringent certifications. In this paper, the control strategy of generating system is enhanced with VDE-AR-N 4105 and BDEW grid code, where both active/reactive powers are controlled. Simulation and experimental results of 100-kW PV inverter are shown to verify the effectiveness of the proposed implemental control strategy. Index Terms—BDEW, Control strategy, fault ride through (FRT), grid code, grid-connected, grid stabilization, photovoltaic (PV), power control, three phase, VDE-AR-N.

I. INTRODUCTION ECENTLY, renewable energy sources are used to improve the quality of power sources due to global warming and environmental conditions. Power electronics technology plays an important role in distributed generation and in the interconnection between renewable energy sources and the electrical grid network. Normally, photovoltaic (PV) systems can be categorized into stand-alone and grid-connected types, where, in the global market, a large proportion of PV solar power systems are grid connected [1]. The importance of large-scale grid-connected PV inverter is emphasized since the market for power plants is rapidly expanding from several to hundreds of megawatts [2]. Also, most large-scale PV inverter systems are connected to the low-voltage (LV) and/or medium-voltage

R

Manuscript received September 19, 2012; revised February 15, 2013; accepted May 12, 2013. Date of publication July 3, 2013; date of current version August 16, 2013. Paper no. TEC-00497-2012. Y. Bae and R.-Y. Kim are with the Department of Electrical and Biomedical Engineering, Hanyang University, Seoul 133-791, Korea (e-mail: [email protected]; [email protected]). T.-K. Vu is with KACO New Energy Inc., Gyeonggi-do 462-807, Korea (e-mail: [email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TEC.2013.2263885

(MV) grids, but each country has its own system-type or grid characteristics [3]. Thus, LV and MV grid-code certifications are important issues in large-scale PV inverter systems. Normally, the well-known current control scheme is used in PV systems for providing the maximum available active power from the PV array to the grid [4]–[9]. However, although the generating power plants using current control which are connected to an LV or MV grid can contribute to grid stability, it may have some weak points in practice. For example, if the grid voltage and its frequency vary, the amount of consumed power is relatively changed since the grid frequency depends on the supplying speed of the supply power. To prevent grid instability due to high penetration of renewable energy, the active power is controlled with the consideration of grid frequency variation. Furthermore, a conventional current control scheme cannot manage the desired power in case of grid voltage variations (e.g., ±10% voltage sags or swells). To solve these problems, the grid-connected PV inverter systems should not only supply active power to the system via the maximum power point tracking, but also the reactive power based on the strict standard and grid-code certifications [9]–[18]. This paper presents a power control strategy using the instantaneous power theory for three-phase PV system with consideration of VDE-AR-N 4105 and the grid-code certification of German Association of Energy and Water Industries (BDEW). Up to date, fault-ride-through (FRT) function of BDEW grid code does not require any action for active or reactive power control except connecting the system to the grid without being interrupted in unsymmetrical grid case [16], [21], [22]; therefore, in this paper, only the symmetrical grid network is considered. The proposed control strategy can not only perform power quality control (PQC), but also support reactive component demanded by grid voltage variation. Since the German LV and MV grid codes have not much difference, only in the FRT function, the simulation and experiment of 100-kW PV inverter are carried out in LV grid network and the FRT test is carried out in MV grid network. Simulation and experimental results are shown to verify the effectiveness of the proposed control strategy. II. REVIEW OF GRID-CODE CERTIFICATION [9]–[18] A. Dynamic Grid Support Fig. 1 shows the limiting curves of type 2 plants during a fault [22]. The power plants must not be disconnected during a

0885-8969 © 2013 IEEE

620


Fig. 3.

Active power reduction depends on grid frequency.

In case of need, the voltage control must be possible to supply the reactive current of at least 100% of rated current Fig. 1. Limiting curves of voltage at the grid connection point when grid voltage fault.

k=

ΔIRe /IRated ≥ 2 (p.u.) ΔV /VRated

(1)

where VRated is the rated voltage, VPre is the voltage prior to fault, V is the instantaneous voltage during fault, IRated is the rated current, IRe Pre is the reactive current prior to fault, IRe is the reactive current, ΔV = V − VPre and ΔIRe = IRe − IRe Pre . The dead-band limit range is 0.9VRated ≤ VRated ≤ 1.1VRated . In case of an unsymmetrical fault, the reactive current must not exceed the value that causes the magnitude of non-faulty-phase voltage to be higher than the upper value of limit (1.1VRated ). B. Frequency Dependence

Fig. 2.

Voltage support when grid fault exceeds 10% of rated voltage.

voltage drop-down to 0% of rated voltage Vrated , within duration of ≤150 ms. As shown in Fig. 1, if the after-fault voltage’s magnitude is greater than borderline 1, zone A, the power plants may not be lead to disconnection or instability. If the after-fault grid voltage’s magnitude is in zone B, the power plants should be ridden through. Also, the following options are available with the agreement of the grid operator. 1) A short-circuit current is fed-in. 2) Depending on the concept of grid connection, borderline 2 can be adjustable. 3) Short-time disconnection (up to 2 s). In zone C, a short-time disconnection can be carried out in any case. Sometime, longer disconnection times are also possible. After a short-time disconnection in the zones B and C, the active power must be increased with a gradient of at least 10% of the normal capacity per second. Underneath the blue line or zone D, there are no requirements for maintaining the grid connection. Fig. 2 shows the supporting voltage strategy if the grid voltage fault exceeds 10% of rated voltage. During failure period, the power generating plants must support the grid voltage by means of additional reactive current. The voltage control must ensure to supply a reactive current with a ratio of at least 2% (in p.u.) of the inserted reactive current percentage to dropped voltage percentage, as shown in (1), within 20 ms.

Due to German grid code, while all renewable-based power generating units are in operation, they must reduce the output instantaneous active power with a gradient of 40% of the rated power per hertz, if the grid frequency fgrid > 50.2 Hz as shown in (2). The output power is only allowed to increase again as soon as the frequency is below 50.05 Hz, but as long as the actual grid frequency does not exceed 50.2 Hz. If the grid frequency is lower than 47.5 Hz or greater than 51.5 Hz, the generating power plants must be disconnected from the grid. Fig. 3 shows the relationship between grid frequency and the output instantaneous active power 50.2 Hz − fgrid at 50.2 Hz < fgrid < 51.5 Hz 50 Hz (2) where ΔP is the power reduction, Pm is the instantaneously available power, and fgrid is the grid frequency. ΔP = 20Pm ×

C. Static Grid Support Normally, the PV systems are designed with active power control, since the reactive power control is avoided due to the losses of inverter, transmission lines, or isolation transformer. Therefore, in order to meet the grid-code certification, the capacity of PV inverter must be enlarged (e.g., a 105-kVA PV inverter is designed for a rated active power of 100 kW). Fig. 4 shows the relationship between active/reactive and apparent powers. The reactive power set point can be fixed or adjustable by a signal as follows: 1) a fixed power factor (PF); 2) a fixed reactive power value (in Var);

BAE et al.: IMPLEMENTAL CONTROL STRATEGY FOR GRID STABILIZATION OF GRID-CONNECTED PV SYSTEM

621

that the operation range is divided depending on power levels [17]. Hence, the limit range of injected reactive power can be −18.2◦ ≤ ϕ ≤ 18.2◦ or −0.31 ≤ sin(ϕ) ≤ 0.31. For example, if grid operator wants to inject ΔQ at ±5% of rated voltage, the k factor value will be k= Fig. 4.

Power’s vector diagram.

0.31 ΔQ = = 6.2. ΔV 0.05

(3)

III. CONTROL STRATEGY FOR GRID STABILIZATION IN THE MEDIUM-VOLTAGE THREE-PHASE GRID-CONNECTED PV SYSTEM A. Conventional Control Strategy

Fig. 5.

Reactive power curve depending on grid voltage’s variations.

Fig. 6.

PF depends on active power characteristic.

3) an adjustable reactive power depends on the grid voltage condition, as shown in Fig. 5; 4) an adjustable PF depends on the active power as shown in Fig. 6. Fig. 5 shows the reactive power control with respect to the grid voltage variations, is named as Q(V ), based on the BDEW grid code in Fig. 2. This reactive control Q(V ) function controls the reactive power infusion even when the grid voltage is in normal operating range (90–110% of rated voltage), and the dead band is reduced to be ±1% of rated voltage. The slope of Q(V ) can be controlled by k factor (k = ΔQ/ΔV ) where ΔQ = sin(ϕ) is the inserted reactive power and ΔV = V − VPre (VPre is the voltage prior to fault and V is the instantaneous voltage during the fault, respectively). It is noted that the power generating plants must provide reactive power in every operating points within a range of −0.95 ≤ cos(ϕ) ≤ 0.95, as shown in Fig. 6, but it is noted

The purpose of the VSI is to connect a dc bus, which is fed with the power from a renewable power source to ac grid through a switching device. As the voltage of the dc bus is desired to be constant and the ac grid usually has low impedance, a filtering inductor is used to connect the switching device to the grid. Fig. 7 shows the normal control scheme for PV inverter systems [4]–[9], where Fig. 7(a) shows the well-known current control, where the difference between the measured PV voltage and its reference voltage value is regulated for generating the reference active current (q-axis component) in synchronous controlling frame. Also, the current generated by PV inverter is synchronized with grid voltage using phase-locked loop (PLL) control. Normally, the reference value of reactive current component (d-axis component) is zero. Fig. 7(b) shows the modified current control scheme, where the input power is feedforward into the dc voltage control for improving the dynamics of the PV system. This control can continuously control output power based on PV power better than the conventional current control, since the feedforward power control part can generate the output current reference due to the change of grid voltage. However, the feedforward power control method cannot satisfy the German grid code or grid operator because of the current limiter. Therefore, this kind of controller cannot provide a constant power output according to the grid voltage. Fig. 8 shows an example of relationship between grid voltage and available reactive power. As shown, the reactive power can be raised or decreased proportional to the grid voltage although the reactive current is kept constant in current controller, as illustrated in Fig. 8(a). Hence, the grid-connected PV inverter must have a constant power control which can satisfy the German grid code, as shown in Fig. 8(b). B. Proposed Control Strategy According to [19], a generic xabc (t) three-phase positive sequence signal is a function of time that can be described as follows: ⎡ ⎤ cos (ωt) √ ⎢ ⎥ xabc (t) = 2X ⎣ cos (ωt − 2π/3) ⎦ . cos (ωt + 2π/3)

(4)

622

Fig. 7.


Current control strategy with grid-code certification: (a) conventional; (b) with feedforward power.


Fig. 8. Relationship between grid voltage and available reactive power: (a) current control and (b) power control.

By choosing dθ/dt = ω for the Park reference angle, the transformed signal becomes a vector of constants: xdq 0 ≡ T (θ) xabc .

(5)

According to [20], active and reactive powers can be calculated using the voltage and current in the qd-form as follows: 1 0 3 3 p = (vq iq + vd id ) = [vq d ]T iq d 2 2 0 1 0 1 3 3 q = (vq id − vd iq ) = [vq d ]T (6) iq d . 2 2 −1 0 Assuming vd to be zero, which can be ensured by an adequate tracking of the grid voltage measure using a PLL, (6) implies that active and reactive powers are proportional to iq and id , respectively. One significant advantage of using the pq theory in designing the controller is the possibility of selecting the portions of active, reactive, and zero-sequence powers. Sometimes, it is convenient to separate these powers into their average and oscillating parts as p = p¯ + p˜ and q = q¯ + q˜, where “-” and “∼” mean the average and the oscillating components, respectively.

623

Fig. 9 shows the proposed power control scheme with the grid-code certification taken into account. As shown in Fig. 9, the active/reactive powers of the grid side are continuously measured using (6) and instantaneously separated into their average and oscillating parts. In real implementation, this separation is realized through a low-pass filter. The cutoff frequency of a low-pass filter must be selected carefully as to the inherent dynamics that lead to compensation errors during transients. Unfortunately, the unavoidable time delay introduced by the low-pass filter may degenerate the entire performance of the system during transient. However, due to the delay caused by the low-pass filter does not break the German grid-code’s requirement of response time. As shown in Fig. 10, the transient time of active power control is 50 s [21]. If the response time is greater than 50 s, an additional test needs to be carried out with a step change from 100% to 15% of rated power. Fig. 11 shows an example of active power test based on German grid code, where the set points are changed from 100% to 10% of rated power. Also, a dc voltage regulator should be added to the control strategy in a real implementation. In fact, a small amount of average real power (Ploss ) must be drawn continuously from the power system to supply switching and losses in the pulse width modulation (PWM) converter. Otherwise, this energy would be supplied by the dc capacitor, which would discharge continuously. This means that the dc voltage must be kept higher than the peak value of the ac-bus voltage, in order to guarantee the controllability of the PWM current control. The PV inverter must be able to reduce its output active power in several cases; among them, some unexpected situations are described as follows: 1) unsafe operation; 2) islanding phenomena; 3) static/dynamic grid instability; 4) instable system due to frequency increase. Based on the grid-code certification, which is aforementioned in Section II, the active power must be reduced with a gradient of 40% of the rated power per Hertz, if the grid frequency fgrid > 50.2 Hz as shown in (2). In case of using second- or third-order output low-pass filter, the reactive power component in filter capacitors, which is shown in (7), must be considered in the control design step

qc = 3ic vc = 3(vc zc )vc = 3ωCvc2

(7)

where ω is the grid frequency (rad/s), and C and vc are the filter capacitance and voltage, respectively. Due to German grid code, it is noted that the reactive power reference value can be either fixed or adjusted by a signal from the grid operator. Based on this fact, the PF is defined as the ratio of the active power “p” to apparent power “s” as follows: PF = cos(ϕ) =

p or ϕ = a cos(PF). s

(8)

624

Fig. 9.

Fig. 10.


Proposed power control strategy based on German grid code.

Response time of power control corresponding to German grid code.

And the relationship between active and reactive power can be obtained as follows: Fig. 11.

q = p tan(ϕ) or q = p tan [a cos(PF)] .

(9)

As shown in (9), the reactive power can be consider as a function of PF. Hence, the reactive power reference can be chosen as follows: q ∗ = p∗ tan [a cos (PF∗ )]

(10)

where “∗” denotes the reference values. The PF is adjusted within the appropriate range (−0.95 to 0.95, corresponding to

Active power response test based on German grid code.

capacitive or inductive load cases) Gop (s) =

Kpc Kpp s2 + (Kpp Kic + Kpc Kip )s + Kip Kic Td Lf Cf s5 + Lf Cf s4 + Td s3 + (Kpc + 1)s2 + Kic s (11)

Gcl (s) =

3 Vq 2

grid

Kpc Kpp s2 + (Kpp Kic + Kpc Kip )s + Kip Kic D (12)


Fig. 12.

625

Equivalent control block diagram of power control structure.

where D = Td Lf Cf s5 + Lf Cf s4 + Td s3 3 + Kpc + 1 + Vq grid Kpc Kpp s2 2 3 + · · · + Kic + Vq grid (Kpp Kic + Kpc Kip ) s 2 3 + Vq 2

grid Kip Kic .

(13)

Hence, the reactive power control can be considered as a closed-loop control so far, where both reactive power and PF can be controlled. Fig. 12 shows the equivalent power control block diagram and its open-loop and closed-loop transfer functions can be obtained in (11), (12), and (13), respectively, where Kpp and Kip are power controller gains, Kpc and Kic are current controller gains, Lf is filter inductance, Cf is filter capacitance, Td is the computational PWM delay, Ts is switching period, and Vq grid is the grid voltage magnitude. The corresponding frequency response is shown in Fig. 13(a) and its Nyquist diagram is illustrated in Fig. 13(b). As shown, the system using power control is stable. In addition, the compensation values for other reactive components (e.g., the stray components, energy transmission line, etc.) depend on exact conditions and they only can be obtained at the installed field. Therefore, in our inverter control program, the information of aforementioned reactive components can be entered with HMI screen as add-in parameters

Fig. 13. (a) Frequency response of closed loop and (b) Nyquist diagram of power control structure.

IV. SIMULATION RESULTS To confirm the validity of the proposed control strategy with the German grid-code certification taken into account, a simulation study using 100-kW three-phase transformer-type PV inverter system was performed using MATLAB/Simulink, as shown in Fig. 14. And the system parameters for simulation study are listed in Table I, and Fig. 15 shows the simulation model of power control structure in detail. The steps-changing’s response of active power is illustrated in Fig. 16, where the active power is ramp up from 10% to 100% of rated power. Some reactive power components still remain due to the filter’s reactive component. Also, its corresponding output grid voltages and currents are shown in Fig. 17, where the reference current is automatically generated based on the change of reference power value. Fig. 18 shows the dependence on frequency of active power. 50.25, 0.7, 51.15, 50.07, and 50 Hz. Based on the German grid-

Fig. 14.

Overall simulation model of the 100-kW PV inverter.

code requirements, the active power is kept to be constant after 51.15 Hz until grid frequency reaches to the original value for the safety reason. Fig. 19(a) shows the active power response of normal grid voltage case with step changes from 100% to 70% then increasing 80% and 90% of rated power. The simulation result of a dip grid voltage case is shown in Fig. 19(b). As shown in the figure, the power control can follow the tracks of reference value whether or not the grid voltage dips. Fig. 20 shows the reactive power response when changing from 0% to 100% of reactive power in case of capacitive and

626


TABLE I SYSTEM PARAMETERS

Fig. 17. (Top) Grid voltage, (middle) grid current, and (bottom) grid current’s q-axis component and its reference (normal: left, partial enlarged: right).

Fig. 15.

Basic simulation model of power control.

Fig. 18.

Active power reduction corresponding to grid frequency’s variation.

V. EXPERIMENTAL RESULTS

Fig. 16. Active and reactive power when step-changing from 10% to 100% of rated power.

inductive cases. The simulation result of PF control is shown in Fig. 21, where the error between the average value of feedback response and its reference is acceptable (≤5%). Fig. 22 shows an example of supporting voltage in the case of grid voltage dips. As shown, the controller can support a reactive power component within 0.02 s, after grid voltage dips.

A 100-kW three-phase grid-connected transformer-type PV inverter, which is shown in Fig. 23(a), was used to verify the proposed control method with the same parameters listed in Table I. The power control, which is integrated with the conventional current control part, is an adopted fully digital control system using 150-MHz floating point TMS320C6711 DSP chip and the PWM pulses are generated through the internal pulse generator of another TMS320F2812 DSP. The experimental testbed is illustrated in Fig. 23(b), where a PV simulator is used as PV power source. Fig. 24 shows the ramping-up of active power from 10% to 100% of rated power. As shown, the output power response mostly tracks its reference value with a small error. The grid frequency dependence of active power is shown in Fig. 25, where the active power is reduced with a gradient of 40% per hertz. Due to the BDEW requirements, the active power is kept constant at the latest value after grid frequency is increased to 50.15 Hz. The active power is increased again when grid frequency reduces to 50 Hz.


Fig. 19.

Fig. 20.

Fig. 21.

PF response in capacitive and inductive load cases.

Fig. 22.

Reactive power supporting when grid voltage dips.

627

Active power response in case of (a) normal and (b) dip grid voltage.

Reactive power response in capacitive and inductive load cases.

With a current control technique, the active power will be reduced when voltage dips, since the reference current is fixed. However, the use of power control can remain approximately the rated value during voltage dips, since the power is controlled in this case, so the current reference value is automatically generated. If the grid voltage sagged, the current reference is increased and vice versa. An example of power response in 10% dip of grid voltage case is illustrated in Fig. 26, where Fig. 26(a) shows the power response using current control. As shown, since the current reference is constant, the active power will depend on the grid voltage. Fig. 26(b) shows the response using power control, where the active power is controlled to track the reference value. The reference current is automatically generated due to the active power set point, so it will be increased if grid voltage sags and vice versa. The reactive power response is also tested, as shown in Fig. 27, where the reactive power set value changes from –100% to 100%

628


Fig. 26. Active power response when 10% voltage dips in case of (a) current control and (b) power control.

Fig. 23. testbed.

(a) Front and inside of 100-kW PV inverter and (b) its experimental

Fig. 27. Reactive power response when step-changing from 100% of capacitive case to 100% of inductive case.

Fig. 24.

Ramp-up of active power.

Fig. 28.

Fig. 25.

Frequency dependence of active power.

PF response due to capacitive and inductive loads.

of rated power. Although there are some delays caused by the use of low-pass filter, these delays do not exceed the rule of transient time given by German BDEW certification. Fig. 28 shows the experimental result of PF control. The PF reference is set from –0.8 (capacitive load case) to 0.8 (inductive load case). Since the reactive power is a closed-loop control with a feedback component, the reactive power and/or the PF can be control easily with good accuracy in responses. The error


Fig. 32.

629

Configuration of testbed for FRT. TABLE II FRT TEST CONDITIONS

Fig. 29.

Reactive power response when 10% voltage sags/swells.

Fig. 30.

Grid current (r-phase) using power control and its THD value.

TABLE III FRT TEST SEQUENCES

Fig. 31.

System efficiency using power control.

between the response and reference value is within in a range of 5%, which is accepted by German BDEW certification. Fig. 29 shows the case of voltage supporting with reactive component. When grid voltage sags or swells, the controller will generate a capacitive or inductive component to remain an approximately constant value of grid voltage magnitude. Fig. 30 shows the grid current (r-phase) and its THD value using power control scheme. As shown, a good output response can be obtained with the proposed control structure. Also, the efficiency of whole system can be greater than 96% at full power, as illustrated in Fig. 31. Fig. 32 shows the testbed configuration for FRT test in “Frounhofer ISE,” Germany, where the FRT test part (the dotted-

Fig. 33.

Proposed control strategy for FRT test.

line rectangle) is based on the standard FGW TR3 [22] and its test conditions are listed in Table II. Also, the FRT test sequences are listed in Table III. Based on Table III, the proposed FRT test strategy for PV application is

630


Fig. 34. Experimental result of Test 1: (a) Starting-fault and (b) ending-fault (Symmetrical fault, V /Vo = 0%, 150 ms fault duration, 90% of rated power).

Fig. 35. Experimental result of Test 2: (a) Starting-fault and (b) ending-fault (Symmetrical fault, V /Vo = 0.2–0.25%, 550 ms fault duration, 90% of rated power).

illustrated in Fig. 33, where each section is described in detail as follows 1) Section 1 : If the grid voltage drops below FRT value standard (0.9 p.u.) which is set in initial parameter, the PV inverter will switch to the FRT operation mode. The overcurrent value due to an inrush current is decided by the slope and depth of voltage sag. Then, when overcurrent occurs, the inverter will stop PWM modulation within one cycle. After that, inverter will infuse the reactive power to grid according to the grid code. Since the reactive power value is proportional to the parameter setting value, grid operator can infuse the reactive power to the grid as much as possible. 2) Section 2 : Infuse both active and reactive powers to grid according to grid code. 3) Section 3 : The overcurrent value due to the inrush current is decided by the slope and depth of voltage sag. Then, when overcurrent occurs, the inverter will stop PWM modulation within one cycle. After that, the inverter will infuse the reactive power to grid according to grid code. 4) Section 4 : If grid voltage rises above FRT value standard (0.9 p.u.) which is set in initial parameter, the PV inverter will switch to normal operation and stop the reactive current infusion. The different FRT test results according to the test sequences in Table II are shown in Fig. 34–37 in the case of symmetrical fault at 90% of rated power operation, where the left-side figure



is the result at starting time of a grid fault and the right-side figure is the ending-time case, respectively. The inverter stops PWM generation within one cycle after overcurrent is detected, as shown in section 1 of figures. Then, a section reactive component is infused to keep inverter operation during the fault. As shown in the figures, the inverter can ride-through the fault, even in the worst case, as shown in Fig. 34.

VI. CONCLUSION In this paper, an optimal control strategy scheme for the threephase grid-connected PV system, with the BDEW certification taken into account, is proposed. Due to the BDEW requirements, the proposed control strategy method not only can perform the PQC, but also can compensate the reactive power demanded by a voltage sag/swell for a grid stabilization purpose. The proposed control strategy does not require any external hardware modification when compared with the use of conventional current control. Simulation and experimental results of 100-kW PV inverter are shown to verify the theoretical analysis and effectiveness of the proposed control strategy. For a future work, the proposed control strategy still needs to be investigated for the PV system in unsymmetrical grid network.


REFERENCES [1] J. M. Carrasco, J. M. Carrasco, L. G. Franquelo, J. T. Bialasiewicz, ´ M. Prats, J. I. León, and N. MorenoE. Galván, R. C. P. Guisado, M. A. Alfonso, “Power-Electronics system for the grid integration of renewable energy source: A survey,” IEEE Trans. Ind. Electron., vol. 53, no. 4, pp. 1002–1016, Aug. 2006. [2] PV Power Plants 2011 Industry Guide, Renewables Insight (2011). [Online]. Available: http://www.pv-power-plants.com/fileadmin/user_ upload/pdf/PVPP12_low.pdf [3] J. C. Boemer, K. Burges, P. Zolotarev, J. Lehner, P. Wajant, M. Fürst, R. Brohm, and T. Kumm, “Overview of German grid issues and retrofit of photovoltaic power plants in Germany for the prevention of frequency stability problems in abnormal system conditions of the ENTSO-E region continental Europe,” in Proc. Solar Integration Workshop, Aarhus, Denmark, 2011. [4] T. Kerekes, M. Liserre, R. Teodorescu, C. Klumpner, and M. Sumner, “Evaluation of three-phase transformerless photovoltaic inverter topologies,” IEEE Trans. Power Electron., vol. 24, no. 9, pp. 2202–2211, Sep. 2009. [5] Z. Ye, R. Walling, L. Garces, R. Zhou, L. Li, T. Wang, “Study and development of anti islanding control for grid connected inverters,” General Electric Global Research Center, Niskayuna, NY, USA, Subcontractor Rep. NREL/SR-560-36243, May 2004. [6] R. Gonzalez, E. Gubia, J. Lopez, and L. Marroyo, “Transformerless singphase multilevel-based photovoltaic inverter,” IEEE Trans. Ind. Electron., vol. 55, no. 7, pp. 2694–2702, Jul. 2008. [7] M. Ciobotaru, R. Teodorescu, and F. Blaabjerg, “Control of single-stage single-phase PV inverter,” in Proc. Eur. Conf. Power Electron. Appl., 2005, pp. 1–10. [8] E. Paál, Z. Weitzl, and C. S. Choi, “Grid management functions built in PV inverters for distributed power generation,” in Proc. ICPE 2011-ECCE Asia, 2011, pp. 2637–2644. [9] M. Garc´ıa-Gracia, N. El Halabi, H. Ajami, and M. Paz Comech, “Integrated control technique for compliance of solar photovoltaic installation grid codes,” IEEE Trans. Energy Convers., vol. 27, no. 3, pp. 792–798, Sep. 2012. [10] L. Ma, H. Liao, J. Li, X. Yang, K. Tschegodajew, and F. Tang, “Analysis of Chinese photovoltaic generation system low voltage ride through characters,” in Proc. IEEE 7th Int. Power Electron. Motion Control Conf., Jun. 2012, pp. 1178–1182. [11] F. Iov, A. D. Hansen, P. Sørensen, and N. A. Cutululis, “Mapping of grid faults and grid codes,” Risø Nat. Lab., Tech. Univ. Denmark, Copenhagen, Denmark, Rep. Risø-R-1617-(EN), Jul. 2007. [12] T.-N. Preda, K. Uhlen, and D. Eirik Nordg˚ard, “An overview of the present grid codes for integration of distributed generation,” in Proc. CIRED Workshop, May 2012, pp. 1–4. [13] M. Mohseni and S. M. Islam, “Transient control of DFIG-Based wind power plants in compliance with the Australian grid code,” IEEE Trans. Power Electron., vol. 27, no. 6, pp. 2813–2824, Jun. 2012. [14] IEEE Recommended Practices and Requirements for Harmonic Control in Electrical Power Systems, IEEE Standard 519-1992, 1993. [Online]. Available: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?&arnumber= 210894 [15] H. Berndt, M. Hermann, H. D. Kreye, R. Reinisch, U. Scherer, and J. Vanzetta, “Network and system rules of the German transmission system operators,” Transmission Code, 2007. [Online]. Available: http://www. vde.com/de/fnn/dokumente/documents/transmissioncode%202007_engl. pdf [16] W. Bartels, F. Ehlers, K. Heidenreich, R. Hüttner, H. Kühn, T. Meyer, T. Kumm, J.-M. Salzmann, H.-D. Schäfer, and K.-H. Weck, “Generating plants connected to the medium-voltage network,” Technical Guideline of BDEW, 2008. [Online]. Available: http://www.bdew.de/internet.nsf/id/ A2A0475F2FAE8F44C12578300041C92F / $ file / BDEW_RL_EA-amMS-Netz_engl.pdf [17] Technical Requirements for the Connection to and Parallel Operation With Low-Voltage Distribution Networks, VDE-AR-N 4105, 2011. [18] E. Troester, “New German grid codes for connecting PV systems to the medium voltage power grid,” in Proc. 2nd Int. Workshop Concentrating Photovoltaic Power Plants: Opt. Design, Prod., Grid Connection, 2009, pp. 1–4. [19] C. L. Fortescue, “Method of symmetrical co-ordinates applied to the solution of polyphase networks,” Amer. Inst. Electr. Eng., Trans., vol. 37, no. 2, pp. 1027–1140, 1918.

631

[20] H. Akagi, E. H. Watanabe, and M. Aredes, Instantaneous Power Theory and Applications to Power Conditioning (IEEE Press Series on Power Engineering). Hoboken, NJ, USA: Wiley-IEEE Press, 2007. [21] Technical Guidelines for Power Generating Units. Part 8 Certification of the Electrical Characteristics of Power Generating Units and Systems in the Medium-, High- and Highest-Voltage Grids, Revision 3, FGW, Kiel, Germany, Mar. 2010. [22] Technical Guidelines for Power Generating Units. Part 3 Determination of Electrical Characteristics of Power Generating Units Connected to MV, HV and EHV Grids, Revision 20, FGW, Kiel, Germany, Oct. 2009.

Youngsang Bae was born in Kunsan, Korea, in 1979. He received the M.S. degree in control and instrumentation engineering from the Seoul National University of Science and Technology, Seoul, Korea, in 2005. He is currently pursuing the Ph.D. degree in the Energy Electronics Control System Lab, Hanyang University, Seoul, Korea. From 2008 to 2012, he was a Senior Researcher at the KACO New Energy R&D Center, Gyeonggi-do, Korea. Since 2012, he has been with the Destin Power Inc., Gyeonggi-do, Korea, where he is currently a Senior Researcher at R&D Center. His main research interests include analysis of power quality; modeling, design, and control of power distribution system for renewable energies; and energy storage system.

Trung-Kien Vu received the B.S. degree in electrical engineering from the Hanoi University of Science and Technology, Hanoi, Vietnam, in 2001, the M.S. and Ph.D. degrees from Chungnam National University, Daejeon, Korea, in 2005 and 2011, respectively, all in electronics engineering. From 2001 to 2003, he was with Automation Mechanic Application Company Limited, Vietnam, where he was involved in the development of automatic control system. Since 2011, he has been with KACO New Energy Inc., Gyeonggi-do, Korea, where he is currently a Senior Researcher at R&D Center. His main research interests include analysis of power quality; modeling, design, and control of power distribution system for renewable energies.

Rae-Young Kim (S’06–M’10) received the B.S. and M.S. degrees from Hanyang University, Seoul, Korea, in 1997 and 1999, respectively, and the Ph.D. degree from the Virginia Polytechnic Institute and State University, Blacksburg, VA, USA, in 2009, all in electrical engineering. From 1999 to 2004, he was a Senior Researcher at the Hyosung Heavy Industry R&D Center, Seoul. In 2009, he was a Postdoctoral Researcher at National Semiconductor Corporation, involved in a smart home energy management system. Since 2010, he has been with Hanyang University, where he is currently an Assistant Professor in the Department of Electrical and Biomedical Engineering. His research interests include modeling and control of power converter systems, soft-switching techniques, energy management systems in smart grid applications, power converter systems for renewable energies, and motor drive systems. Dr. Kim received the First Prize Paper Award in Industry Applications Society (IAS) 2007. Since 2009, he has been a member of the IEEE IAS Industry Power Converters Committee, and also served as a Reviewer for the IEEE TRANSACTION ON INDUSTRIAL ELECTRONICS and the IEEE TRANSACTION ON INDUSTRY APPLICATIONS.