Coordinated Voltage Control Methods For Distribution Networks With High Penetrations Of Photovoltaic

With the development and application of renewable energy sources,a large number of photovoltaic(PV)systems have been connected to distribution networks in recent years.However,the output power of PV has strong random fluctuations,which will cause voltage fluctuations in the distribution network,thereby reducing the power quality and power supply reliability of the distribution network.Therefore,investigating the coordinated voltage regulation methods for the distribution network with high penetrations of PV has important theoretical value and practical application significance for promoting the utilization and consumption of PV power,and improving the economics and power quality of the distribution network.Firstly,the K-means clustering algorithm is utilized to analyze the characteristics of PV power fluctuations,and then the theoretical analysis of the problem of nodal voltage fluctuations in distribution networks caused by PV power fluctuations is performed.Eventually,two feasible methods are given to solve the problem of nodal voltage variations of the distribution network with high penetrations of PV.The first one focuses on the “source” side and can reduce the nodal voltage variations of the distribution network by reducing the fluctuations of the active power injected into the distribution network from PV plants.The second one focuses on the “distribution network” side and can control the the voltage variations of all nodes in the distribution network by cooperatively optimizing the operation of each controllable devices in the distribution network.The following research is carried out regarding the voltage control methods for distribution networks:1.In order to solve the key problems faced by using the battery energy storage system(BESS)to smooth PV power fluctuations,such as excessive charging and discharging cycles of batteries,higher capacity requirements of batteries,and deviations of the state of charge(SOC)of batteries from the ideal range,this dissertation proposes a control strategy for the active power curtailment of PV and the BESS to cooperatively smooth grid-connected power fluctuations of a PV plant.Dynamically limiting the real-time active power of PV can not only cooperate with the BESS to achieve better smoothing effects,but also effectively reduce the voltage variations at the PV connection point as well as the utilization and capacity requirements of batteries.In addition,this dissertation proposes an adaptive adjustment of Kalman filter parameters based real-time charging and discharging control strategy for the BESS.This strategy can accurately control the charging and discharging power of the BESS.At the same time,it can effectively control the BESS SOC by dynamically adjusting the Kalman filter parameters.Finally,the effectiveness of the strategies proposed above has been verified via simulations.2.Since supercapacitors have the advantages of large charging and discharging power rate and long cycle life,it is more appropriate to use supercapacitors to smooth the high-frequency power fluctuations of PV,and use batteries to smooth the low-frequency power fluctuations of PV.In order to solve the difficulties in determining the frequency demarcation point of high-frequency and low-frequency power fluctuation components of PV,which is faced by the existing filtering algorithm-based method for the hybrid energy storage system(HESS)to smooth PV power fluctuations,this dissertation first proposes a control strategy for the active power curtailment of PV and the HESS to cooperatively smooth grid-connected power fluctuations of a PV plant.Based on this,this dissertation proposes a multi-objective nonlinear real-time optimal dispatching strategy for the HESS to control its charging and dispatching power.This strategy can correctly control the charging and discharging power of batteries and supercapacitors without considering the frequency characteristics of PV power fluctuations.An optimal power dispatching model of the HESS is established with the aim of minimizing the operating losses of the HESS and the SOC deviations of the supercapacitors.By dynamically adjusting the weight coefficient of the sub-objective functions of the above model,the ability of the HESS to cope with a sudden drop in PV power can be significantly improved.The simulation results show that the proposed smoothing strategy can not only suppress the fluctuation of the active power injected into the distribution network from a PV plant within a given range,but also effectively reduce the voltage fluctuation at the connection point of the PV plant,the operating loss of the HESS,and the number of charging and discharging cycles of the batteries.3.To solve the problem of allocating the capacity of batteries and supercapacitors in the HESS,this dissertation proposes an optimal capacity allocation method for the HESS based on the particle swarm optimization(PSO)algorithm with inertia weight.A nonlinear optimal allocation model of the capacities of batteries and supercapacitors is established with the aim of maximizing the net income of the PV and HESS system.The PSO algorithm with inertia weight is employed to solve the established allocation model.Since the proposed allocation method not only considers the optimization of the real-time charging and discharging power of the HESS,but also considers a large number of typical PV power fluctuation scenarios when optimizing the HESS capacity,the obtained capacity allocation results of the HESS are not only extremely economical,but also ensure that the HESS can effectively deal with various power fluctuations that may occur in the actual operation of PV.4.To solve the problem of voltage variations of all nodes in the distribution network,this dissertation proposes a two-stage coordinated voltage regulation method for distribution networks considering the optimization of the substation.This method can take into account the control of nodal voltage fluctuations in the distribution network and the optimization of substation operation.In the centralized optimization stage,a multi-objective mixed integer nonlinear programming(MOMINP)model for the distribution network is developed to optimize the active and reactive power flow of feeders,aiming at reducing the nodal voltage fluctuations and operating losses of the distribution network,the number of adjustments of the on-load tap changer(OLTC)and capacitor banks(CBs),and the PV power curtailments.In addition,it aims to improve the power factor of the OLTC.The established MOMINP model is solved by using the improved non-dominated sorting genetic algorithm II(NSGA-II).Meanwhile,a fast decision-making method for selecting the most feasible solution from the Pareto optimal solution set is given.In the local optimization stage,this dissertation proposes a real-time active power control algorithm of PV and a droop control based real-time reactive power control algorithm of PV.Simulation results show that the proposed two-stage coordinated voltage regulation method can not only control the voltage variations of all nodes within a given range,but also improve the economics of the substation and distribution networks.