Coordinated Control of Distributed Energy Storage System With Tap Changer Transformers for Voltage Rise Mitigation Under High Photovoltaic Penetration

Transcription

1 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 2, JUNE Coordinated Control of Distributed Energy Storage System With Tap Changer Transformers for Voltage Rise Mitigation Under High Photovoltaic Penetration Xiaohu Liu, Student Member, IEEE, Andreas Aichhorn, Student Member, IEEE, Liming Liu, Senior Member, IEEE, and Hui Li, Senior Member, IEEE Abstract This paper proposes a coordinated control of distributed energy storage system (ESS) with traditional voltage regulators including the on-load tap changer transformers (OLTC) and step voltage regulators (SVR) to solve the voltage rise problem caused by the high photovoltaic (PV) penetration in the low-voltage distribution network. The main objective of this coordinated control is to relieve the tap changer transformer operation stress, shave the distribution network peak load and decrease the transmission and distribution resistive power losses under high solar power penetration. The proposed control method limits the energy storage depth of discharge in order to meet a more than ten-year cycle life. A benchmark distribution network model was developed in the Real Time Digital Simulator (RTDS) and the simulation results from the studied cases verified the proposed coordinated control strategy. The experimental implementation of proposed control algorithms were developed based on a power hardware-in-the-loop (PHIL) test bed with a 22 kwh ESS, a smart meter, Labview controller, and RTDS. The experimental results were consistent with those obtained from simulation study. Index Terms Coordinated control, ESS, high penetration PV, power hardware-in-the-loop (PHIL), tap changer transformer, voltage rise. I. INTRODUCTION P HOTOVOLTAIC (PV) is one of the fastest growing renewable energy sources in the world. Several ongoing initiative projects are targeted on developing and improving the key technology of enabling high renewable energy penetration in the electrical grid of future, such as the Intelligrid project of the Electric Power Research Institute (EPRI), the smart grid demonstration projects of the U.S. Department of Energy (DOE), the Galvin Perfect Power Initiative project, and others [1], [2]. The voltage rise issue has been reported as one of the concerns under high penetration of renewable distributed generations (DG) [3]. The reverse power flow caused by large amounts of DG may cause voltage rise to which distribution network operators (DNOs) control cannot effectively respond since the traditional grid has been planned to deliver power to the load at Manuscript received December 02, 2010; revised June 17, 2011; accepted November 10, Date of publication February 03, 2012; date of current version May 21, This work was supported by the National Science Foundation under Grant ECCS and the Department of Energy Sunshine State Solar Grid Initiative (SUNGRIN) under Grant DE-EE Paper no. TSG X. Liu, L. Liu and H. Li are with the Center for Advanced Power Systems, Florida State University, Tallahassee, FL USA ( xiaohu@caps. fsu.edu; liming@caps.fsu.edu, hli@caps.fsu.edu). A. Aichorn is with the Upper Austria University of Applied Sciences, Wels, Austria. ( andreasaichhorn@gmail.com). Color versions of one or more of the figures in this paper are available online at Digital Object Identifier /TSG satisfactory voltage range [4]. To solve this problem, many solutions have been proposed in recent literature. Reference [4] proposes a method of reactive power injection which is not to control bus voltage but to guarantee that active power generation does not cause voltage rise. The advantage is that the voltage becomes independent of the generation and the DNOs can be kept to their traditional task of voltage regulation. However, as illustrated in the paper, the disadvantages are the higher OLTC stress and feeder loss. Another drawback is that this method requires information about the upstream feeder impedance resulting in a communicating need in case of feeder reconfiguration. Reference [5] also presents a method of reactive power control where inverters decide their output reactive power autonomously at first, and continuously modify them with exchanging information between each inverter. The General Electric (GE) 2008 report [6] compares the performance with different penetration levels when using on-load tap-changing (OLTC) transformer, step voltage regulator (SVR) or PV inverter to regulate the distributed load voltage. The key conclusion from the report is that coordinated control of utility equipment and DG assets can be used to enhance the performance of distribution systems. In addition, a communication link established between service points (customer meter connections) and the utility equipment is helpful. The report investigates on the reactive power support from PV inverters. But as pointed out in the report, at present, the IEEE 1547 and UL 1741 only allow PV systems to operate at a unity power factor. However, it provides a promising method if the standard can be changed in the future. Reference [7] compares three control schemes including generation curtailment, reactive power control and area-based coordinated OLTC control. It concludes that the revenue obtained from using the OLTC coordinated control is the highest. However, this coordinated control scheme was not presented in the paper. Reference [8] also compares the control schemes of generation curtailment and reactive power control. It shows that reactive power control is not efficient for small value of X/R networks and over-sized inverter is required to absorb reactive power. In this paper, a new coordinated control of distributed energy storage system (ESS) with tap changer transformers for voltage rise mitigation has been proposed. The main objective of this control is to relieve the tap changer operation stress, shave the utility peak load and decrease the transmission and distribution resistive power loss under high solar power penetration. To verify this control strategy, a benchmark distribution network model was developed in the real time digital simulator (RTDS). A power hardware-in-the-loop (PHIL) experiment test bed including a 22 kwh ESS, smart meter, and the Labview /$ IEEE

2 898 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 2, JUNE 2012 Fig. 1. One line diagram of the simple distribution network. Fig. 3. Traditional line drop compensation control of tap changer regulator. Fig. 2. Voltage range limits used in the study [6]. control interface was also developed to implement the proposed controller. The controller communicated with the RTDS model and ESS via the Distributed Network Protocol. The smart meter communicated with the ESS state-of-charger (SOC) controller via the Modbus Transmission Control Protocol (TCP). II. COORDINATED DISTRIBUTION VOLTAGE CONTROL STRATEGY A. System Description and Voltage Rise Analysis The traditional distribution system has been designed as a unidirectional power flow network. As more and more distributed renewable sources are connected to the grid, the original unidirectional network will be changed toward the bidirectional network in the future. This change brings utility operation issues such as the voltage rise problem caused by the reverse power flow from the distributed renewable energy generation. Fig. 1 illustrates the one line diagram of the simplified distribution network. There is a distributed generator connected to the load side. The generator voltage can be approximately expressed in (1) where is the substation secondary bus voltage, is the feeder line reactance and is feeder line resistance. and are the real and reactive power provided by the generator, respectively. and are the real and reactive power consumed by the load. Equation (1) shows that the generator voltage may be higher than the upper-limit if the network ratio is relatively low and there is a significant reverse power flow. One solution is that the generators can absorb a relatively large reactive power to compensate the reverse power flow. The alternative solution is that the substation secondary voltage can be correspondingly controlled or the real power injection to the grid can be decreased. The details of the solution for the voltage rise issue will be discussed below. B. Distribution Network Voltage Regulation Requirements An acceptable voltage waveform quality is one of the most important requirements for the utility daily operation. Appendix A shows the ANSI C84.1 which provides a guideline for the normal range of 120 V voltage level. Utilities may (1) choose a different standard based on their own specificcircumstances. In this paper, the voltage regulation control will follow the requirement presented by [6] which is redrawn in Fig. 2. The primary voltage refers to the voltage at the point of primary side of the step down transformer at the customer side. The service voltage means the voltage at the customer s meter, or the load side of the point of common coupling (PCC). The utilization voltage refers to the voltage at the point of use where the outlet equipment is plugged in. C. Traditional Method Voltage control in traditional distribution system is usually achieved by incorporating OLTC, switched capacitors (SC) and step voltage regulator (SVR). OLTC is typically constructed as autotransformers with automatically adjusting taps. At present, there are new solid state OLTC regulators which offer significant advantages of improved performance and reduced maintenance costs over traditional mechanical ones and they can provide more comprehensive control capability such as coordinated control with communication [9]. Similar to OLTC, SVR is also a tap changer based voltage regulator. Usually it is defined that the voltage regulator at the substation is an OLTC, and the one along the feeder is a SVR [6]. The traditional control system of a tap changer based voltage regulator measures the voltage and load current, estimates the voltage at the remote point, and triggers the tap change when the estimated voltage is out of boundary [6]. Fig. 3 shows the traditional line drop compensation (LDC) control strategy [10] for a tap changer regulator. This method feeds back the voltage at the secondary side of the OLTC transformer, the secondary side transformer current, estimated load bus current and line impedance between the transformer and the load to estimate the voltage drop. The compensation function for controlling the Bus 4 voltage is shown in (2) can be locally measured on the transformer secondary side. The load bus current and can be estimated. The line impedance and can also be obtained. This method can provide voltage control of Bus 4 at a nominal load. However, it may fail under high penetration PV. The reason is that it is difficult to predict the load current and with the intermittent renewable energy source. This problem can be solved by adding the communication network to send the real-time load current information back to the controller or the communication network can send all (2)

3 LIU et al.: COORDINATED CONTROL OF DISTRIBUTED ENERGY STORAGE SYSTEM 899 In this paper, the proposed method also needs the communication network to realize the control algorithms. In order to have a fair comparison between the proposed method and the traditional method, the traditional method of tap changer controller in this paper refers to the controller with the communication capability which means its control strategy will not fail when the grid has reverse power flow. Fig. 4. Proposed method: (a) system control diagram; and (b) flow chart of tap changer regulator, coordination controller and SOC controller. the node bus voltage information back to the control center in order to maintain the voltage profiles along the feeder within the limits. D. Proposed Method This paper proposes a coordinated control of distributed ESS with tap changer transformers for voltage rise mitigation. Fig. 4(a) shows the proposed system control diagram. The coordination control strategy is as follows: during the nonpeak load time, the tap changing actions response first when the voltage rise happens. Then the centralized coordination controller will broadcast the coordination charging signal to the distributed SOC controllers. The distributed ESS SOC controllers will charge the battery to lessen the transformer operation stress by absorbing the reverse power flow. During the peak load time, the centralized coordination controller will broadcast the coordination discharging signal and the distributed ESS will discharge the battery to shave the grid peak load. The control flowchart for the tap changer regulator, the centralized coordination controller and the SOC controller are shown in Fig. 4(b). The tap changer regulator employs the LDC method with communication capability as illustrated in Fig. 3. The control variable Flag is used to indicate if there is any voltage rise happening. If the voltage rise happens, the transformer only regulates the Bus 1 voltage to be within the limits. The distributed ESS SOC controllers will charge the battery based on the charging coordination signal to mitigate the reverse power flow. Through the coordination control, the additional tap changing operations, caused by the voltage rise, can be minimized. If there is no voltage rise happening, the tap changer regulator still uses the traditional method to keep the whole feeder to be within the limits. The coordination controller s function is to coordinate the distributed ESS charging or discharging operation. It needs the Bus 1 voltage and control variable Flag informationfromtap changer regulator. If Bus 1 voltage is not within the limits, it indicates that the transformer tap changing operation has not been finished. Thus, the energy storage charging operation will not be issued. If the Bus 1 voltage is within the normal range and Bus 2 or Bus 3 voltage are out of range, the charging operation will be issued. The discharging operation is based on the command from the peak load estimator. The distributed ESS SOC controllers receive the coordination charging/discharging operation command from the coordination controller. If coordination charging operation is commanded and the local PCC voltage is out of boundary, the charging operation can be issued if the current energy storage SOC is smaller than its maximum SOC limit. The discharging operation can be enabled if the SOC controller receives the discharging command and its SOC is bigger than its minimum SOC limit. The proposed SOC controllers will always keep the energy storage SOC to be within the limits in order to extend the energy storage cycle life. A dedicated state of charge control scheme is desirable in order to extend the energy storage cycle life. Reference [11] presents that the relationship between the typical battery cycle life and the depth of discharge is logarithmic. Therefore the number of cycles yielded by a battery goes up exponentially

4 900 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 2, JUNE 2012 Fig. 5. Distribution system model: (a) the single line diagram of the distribution system model; and (b) Feeder 2 model. Fig. 6. Distributed load and solar power profile. with decreased depth of discharge (DOD). This holds for most battery cell chemistries. In order to meet a more than ten-year cycle life, the proposed SOC control in this paper will limit DOD to 20% which means the battery can have 3300 cycle life according to Appendix B. The main advantages of the proposed method compared to the traditional method are summarized below: The operation of tap changer transformers will be optimized with less switching operation times and stress. Therefore the proposed method can optimize utility asset utilization. Distributed ESS can provide real power support in the peak load period. This peak load shaving function can optimize the utility operation because the energy during peak period of the day can be several times more expensive and sometimes hard to acquire. The transmission and distribution resistive losses will be decreased because of the peak load shaving function. Transformer tap changers may not be adequately rated to accept significant reverse real power and the voltage control schemes for tap changers may also be affected by the flows of reverse power flow [12], [13]. However, this potential problem will not happen if the proposed method is used because the distributed ESS will help to decrease the reverse real power flow effectively. Fig. 7. Baseline profile at 11 A.M.: (a) voltage profile; and (b) power flow profile. III. CASE STUDY A. Network Description Arepresentative distribution system based on the previous distributed generation study conducted by General Electric (GE) [6] was selected in this paper. The model is a fictional study system with two feeders that can be looped. It includes the fundamental distribution system components for investiga-

5 LIU et al.: COORDINATED CONTROL OF DISTRIBUTED ENERGY STORAGE SYSTEM 901 Fig. 8. The baseline scenarios of voltage profiles at 11 A.M. with %. tion of voltage regulation. There are on-load tap-changers on the substation transformer, step voltage regulators, switched capacitors and distribution transformers. Thus this model was selected to verify the proposed control algorithm. In order to explicitly investigate the voltage at the customer service entrance, service transformers and secondary circuits were added to the original model developed in [6]. Fig. 5(a) shows the final model adopted in this paper. The system base is 10 MVA. Bus 999 represents the infinite bus and it is the slack bus in the power flow model. This distribution model was investigated in the Real Time Digital Simulator (RTDS) with a time step of 50 s. The study will focus on the Feeder 2. Feeder 2 model is therefore redrawn in Fig. 5(b) with more details. All the load buses have solar PV and energy storage connected to it with the capacity relative to that of the load on the same bus. The energy storage capacity is rated at the two-hour peak load [14]. The Feeder 2 is about six miles in length. The total load is 11 MVA. Seven aggregated loads represent a mixture of the distributed loads and commercial loads ranging from 0.3 MW to 5 MW. The primary feeder base voltage is 12.5 kv. The secondary feeder base voltage is 240 V for residential loads. The load power factor is 0.92 which is representative of the mixture of distributed and commercial loads. The service transformers capacity is rated at 1.5 per unit (p.u.) related to the served load. The impedance of transformers is 2.5% and X/R ratio is 1.5. An average feeder length of 200 ft was selected. The secondary feeder impedance is calculated based on the conductor with 200 A of thermal capacity. More detailed feeder information can be found in Appendix D. B. Baseline Scenario Fig. 6 shows the aggregate residential daily load profile and solar power profile. The load data is referred to [15] and the solar power data is provided by the school energy data located on the Florida Solar Energy Center website [16]. From the previous analysis, the voltage rise happens when there is reverse real power flow.asshowninfig.6,at11 A.M. the reverse real power flow reaches the maximum value. Fig. 7 shows the primary feeder and service entrance baseline profiles along the feeder at 11 A.M.. Fig. 7(a) shows that the primary feeder voltage values decrease with increased feeder length at both primary and Fig. 9. Bus 205 voltage profile under %: (a) at primary feeder; and (b) service entrance. Fig. 10. ESS daily state-of-charge under %. the secondary side of SVR. SVR boosts the Bus 203 voltage up to the upper-limit value. Fig. 7(b) shows that there is a sudden increase in the reactive power at 3.5 miles due to the connection of the capacitor bank.

6 902 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 2, JUNE 2012 Fig. 11. OLTC and SVR tap position with %: (a) traditional method; and (b) proposed method. Fig. 12. Daily real power flow at Bus 201 under with %: (a) real power flow; and (b) reactive power flow. TABLE I T&D LOSSES DURING PEAK AND OFF-PEAK PERIOD Fig. 13. T&D losses during peak and off-peak period with different solar power penetration levels. To investigate the voltage rise issue, the baseline scenarios of voltage profiles at 11 A.M. with different penetration levels were investigated where penetration level is definedin(3) % (3) The study shows that the voltage profiles no longer decrease when feeder length increases and the voltage rise will exceed the upper-limit when the penetration level is more than 30%. Fig. 8 shows the voltage profiles when %. Since Bus 205 primary feeder voltage has the worst voltage rise scenario under 30% and 50% penetration levels, Bus 205 voltage profiles are selected in the next section in order to verify voltage regulation performance of the proposed method. C. Performance Comparison In order to compare the voltage regulation capability of the proposed method with the traditional method, a worst case of Bus 205 voltage under 50% solar power penetration is selected. Fig. 9 shows the Bus 205 primary feeder and service entrance

7 LIU et al.: COORDINATED CONTROL OF DISTRIBUTED ENERGY STORAGE SYSTEM 903 Fig. 14. The architecture of power hardware-in-the-loop (PHIL) test. voltage regulation performance comparison with the daily voltage profiles. The baseline case shows the voltage profile if the grid voltage regulators do not have any control. Fig. 9 shows that the proposed method can achieve almost the same voltage regulation performance as the traditional method. Fig. 10 illustrates the energy storage daily SOC under 50% solar power penetration. It demonstrated that the proposed method can achieve SOC control which limits the energy storage DOD to 20% under the worst case. To verify the advantages of the proposed method, the comparisons with traditional method are illustrated in the following aspects: 1) Tap Changer Operation Performance: Although the traditional method and the proposed method can achieve similar voltage regulation performance, the proposed coordinated control with distributed ESS can relieve the stress of SVR tap changers under different penetration level. Fig. 11 shows the OLTC and SVR daily tap position comparison under 50% solar power penetration level. Although the OLTC transformer operation of two methods remains the same, the minimum and maximum SVR tap positions of the proposed method are 17 and 24; however, the minimum and maximum SVR tap positions of the traditional method are 15 and 26, respectively. Moreover, the proposed method has 11 tap changing operations. However, the traditional method has 18 tap changing operations. Therefore the proposed method has relieved the tap-changer operation stress with less switching operations and less physical stress which helps to optimize the utility asset utilization. The tap changer operation stress has also improved at 10% and 30% level but not as much as that of 50% due to the less reverse power flow. 2) Peak Load Shaving: Fig. 12 shows the daily power flow at Bus 201 under 50% solar power penetration. Fig. 12(a) shows that the proposed method has the smallest peak-to-average ratio (PAR). It verifies that the proposed method has the peak load shaving function. The gray shaded area is the shifted power. Fig. 12(b) shows the daily reactive power flow at Bus 201 under 50% solar power penetration. The reactive power flow for those two methods has a smaller difference only during the time when Fig. 15. Smart grid lab test platform. charging or discharging the batteries. The similar results were also found at 10% and 30% penetration level and therefore were not shown here. 3) Transmission and Distribution (T&D) Losses: The flow of energy from central generation sites to load centers throughout the grid involves some losses due to the resistance of wires and other equipment at the transmission, subtransmission and distribution levels. The T&D losses of traditional method and the proposed method were calculated respectively. The loss calculation method was obtained from [17] and it is shown in Appendix C. Fig. 13 compares the T&D losses of two methods during the peak and off-peak period with different penetration levels. The results proved that the proposed method has the smaller T&D losses compared to those of the traditional method. The calculated power loss and saved power data is illustrated in Table I. The data shows that the T&D losses saving can be up to 20% with the proposed method under high solar power penetration. IV. EXPERIMENT IMPLEMENTATION The experimental implementation of the proposed control algorithm was achieved using a real time power hardware-in-theloop (PHIL) test bed developed in the laboratory. The architecture of the PHIL system test diagram is illustrated in Fig. 14. The distribution system model was simulated in the RTDS. The Bus 205 voltage was sent out as voltage reference for

8 904 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 2, JUNE 2012 Fig. 16. Service voltage profile from smart meter: (a) proposed method; and (b) traditional method. Fig. 17. method. Powerflow profileatbus201: (a) proposedmethod; and (b) traditional the controllable ac voltage source. The voltage source was connected to the smart meter and 22 kwh ESS. The current through the smart meter was fed back to the RTDS model as the current reference to the ESS of the Bus 205. The proposed centralized coordination controller is developed as the DNP3 [18] master and communicates with the simulated tap changer regulator and SOC controllers in the RTDS grid model via the DNP3. This DNP3 master also communicates with the 22 kwh energy storage SOC controller via the DNP3. The SOC controller communicates with the smart meter via the Modbus protocol [19] and it controls the energy storage through the TCP/UCP communication links. Fig. 15 shows the smart grid lab test platform. The test bed includes a 22 kwh distributed energy storage system (DESS) provided by GreenSmith company and a smart meter from Electro Industries GaugeTech. The energy storage is the lithium-ion iron phosphate (LiFePO4) battery. To validate the proposed control, a real-time 8 hour hardware-in-the-loop experiment was conducted. The experimental results were shown in Fig. 16. The solar and load power profiles were the same of Fig. 6. The time scale was downsized proportionally in order to test a 24 hours performance. For example, 3 A.M. in Fig. 16 was corresponding to 9 A.M. infig.6. Fig kwh ESS state-of-charge state. Fig. 16 illustrates the service voltage profiles (rms) from the smart meter. Two hundred eight volts is the base voltage for the service voltage. It shows that the two methods have the similar voltage profile which is consistent with the simulation results. Therefore the experiment results verified that the proposed method can achieve the same voltage regulation performance as the traditional method.

9 LIU et al.: COORDINATED CONTROL OF DISTRIBUTED ENERGY STORAGE SYSTEM 905 Fig. 17 shows the power profile at Bus 201. By comparing the real power flow during the nonpeak load time and peak load time, it shows that the proposed method has larger power flow during the nonpeak load time. This is due to the battery charging operations. And during the peak load time, the proposed method has smaller power flow because of the battery discharging operations. So this experiment results demonstrate these peak load shaving function of the proposed method. Fig. 18 shows the 22 kwh ESS state-of-charge. This PHIL test was started with the 60% SOC. The experiment result validates that the SOC controller can limit the battery depth-of-discharge to 20% in the proposed method. V. CONCLUSION In this paper, a new coordinated control of distributed energy storage system with tap changer transformers for voltage rise mitigation under high PV penetration has been proposed. The simulation case study has verified the advantages of the proposed control as follows: The operation of tap changer is optimized with less switching operation times and stress. The distributed ESS can realize the peak load shaving function by getting the coordinated charging/discharging operation signal from the coordination controller. The distributed energy storage can achieve longer cycle life resulting from limiting the DOD state-of-charge control. The T&D losses under high penetration of solar power can be decreased. In addition, the proposed control has been implemented using power hardware-in-the-loop (PHIL) test bed developed in the laboratory. The experimental results were consistent with simulation results. C. T&D LOSS CALCULATION METHOD [17] Asimplified circuit shown in above was used to calculate the total T&D loss during peak and off-peak period. The loss calculation is based on the following equation: where and are load current during peak and off-peak periods, respectively. and are equivalent T&D losses during peak and off-peak periods, respectively. is the storage discharging time during the peak period and is the storage charging time during the off-peak period. D. FEEDER INFORMATION APPENDIX A. ANSI C84.1 VOLTAGE RANGE FOR 120 VOLTAGE LEVEL B. BATTERY CYCLE LIFE VS. DEPTH OF DISCHARGE [13] ACKNOWLEDGMENT The authors would like to thank Mr. Sen Zhang from Green- Smith for his support of communication solution for our experimental implementation. REFERENCES [1] J. Maire and D. Von Dollen, Profiling and mapping of intelligent grid R&D programs, IEEE Working Group on Distribution Automation, Rep , Dec [2] [Online]. Available: SG_MYPP.pdf [3] C. L. Masters, Voltage rise: The big issue when connecting embedded generation to long 11 kv overhead lines, Inst. Elect. Eng. Power Eng. J., vol. 16, no. 1, pp. 5 12, Feb [4] P. Carvalho, P. Correia, and L. Ferreira, Distributed reactive power generation control for voltage rise mitigation in distribution networks, IEEE Trans. Power Syst., vol. 23, no. 2, pp , May [5] M. Hojo, H. Hatano, and Y. Huwa, Voltage rise suppression by reactive power control with cooperating photovoltaic generation systems, in Proc.20thCIREDInt.Conf.Electr.Distrib., Jun. 8 11, 2009, pp [6] E. Liu and J. Bebic, Distribution system voltage performance analysis for high-penetration PV, Feb [Online]. Available: eere.energy.gov/solar/pdfs/42298.pdf, NREL/SR [7] S. N. Liew and G. Strbac, Maximising penetration of wind generation in existing distribution networks, IEE Proc. Gener. Transm. Distrib., vol. 149, no. 3, pp , May 2002.

10 906 IEEE TRANSACTIONS ON SMART GRID, VOL. 3, NO. 2, JUNE 2012 [8] E. Demirok et al., Clustered PV inverters in LV networks: An overview of impacts and comparison of voltage control strategies, in Proc IEEE Electr. Power Energy Conf., Oct , 2009, pp [9] R. Echavarria, A. Claudio, and M. Cotorogea, Analysis, design, and implementation of a fast on-load tap changing regulator, IEEE Trans. Power Electron., vol. 22, no. 2, pp , Mar [10] C. Gao and M. A. Redfern, A review of voltage control techniques of networks with distributed generations using on-load tap changer transformers, in Proc. 45th Universities Power Eng. Conf. (UPEC), Aug. 31 Sep , pp [11] [Online]. Available: [12] L.M.CipciganandP.C.Taylor, Investigation of the reverse power flow requirements of high penetrations of small-scale embedded generation, IET Renew. Power Gener., vol. 1, no. 3, pp , Sep [13] M. Thomson, Automatic-voltage-control relays and embedded generation. II, IEEE Power Eng. J., vol. 14, no. 3, pp , Jun [14] J.P.BartonandD.G.Infield, Energy storage and its use with intermittent renewable energy, IEEE Trans. Energy Convers., vol. 19, no. 2, pp , Jun [15] [Online]. Available: [16] [Online]. Available: [17] A. Nourai, V. I. Kogan, and C. M. Schafer, Load leveling reduces T&D line losses, IEEE Trans. Power Del., vol. 23, no. 4, pp , Oct [18] IEEE Standard for Electric Power Systems Communications Distributed Network Protocol (DNP3), IEEE Standard , Jul [19] [Online]. Available: Xiaohu Liu (S 09) received the B.S. and M.S. degrees from Huazhong University of Science and Technology, Wuhan, China, both in electrical engineering, in 2007 and 2009, respectively. He is currently working toward the Ph.D. degree at Florida State University, Tallahassee. His research interests include bidirectional dc-dc converters, solid state transformers (SST) based dc microgrids, and fuel cell power conversion systems. Andreas Aichhorn (S 09) received the B.Sc. degree from the Upper Austria University of Applied Sciences/Campus Wels in automation engineering with the major of Industrial Informatics. He is currently working on the M.S. Thesis in the Center for Advanced Power Systems, Florida State University, Tallahassee. His research interests include communication networks and renewable energy systems. Liming Liu (M 08 SM 11) received the Ph.D. degree in electrical engineering from Huazhong University of Scientist and Technology, China, in He is currently an Assistant Scientist in the Center for Advanced Power Systems, Florida State University, Tallahassee. His research interests include renewable energy conversion systems, modeling and control of multilevel inverter applications, motor drive control with hybrid energy storages, and flexible ac transmission system. Hui Li (S 97 M 00 SM 01) received the B.S. and M.S. degrees in electrical engineering from Huazhong University of Science and Technology, China, in 1992 and 1995, respectively, and the Ph.D. degree in electrical engineering from the University of Tennessee, Knoxville, in She is currently an Associate Professor in the Electrical and Computer Engineering Department, Florida A&M University Florida State University College of Engineering, Tallahassee. Her research interests include PV converters, energy storage applications, and smart grid.

11 本文献由 学霸图书馆 - 文献云下载 收集自网络, 仅供学习交流使用 学霸图书馆 ( 是一个 整合众多图书馆数据库资源, 提供一站式文献检索和下载服务 的 24 小时在线不限 IP 图书馆 图书馆致力于便利 促进学习与科研, 提供最强文献下载服务 图书馆导航 : 图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具