Research On Frequency Control For Distribution Network With Distributed PV-ES Combined Power Generation System

Frequency,which reflects the supply and demand balance between power and loads,is a key index in the operation of distribution network.Any sudden changes of loads in the system may lead to power deviations and frequency fluctuations,and even trigger serious accidents that affects the safety of distribution network.Therefore,it is quite urgent for the system to effectively improve the power quality,safety and stability of the distribution network by solving frequency fluctuation problems.In recent years,with the extensive accesses of distributed generators represented by Photovoltaic(PV),the distribution network is gradually developing from passive to active,greatly enriching the choices of the system for the modes of frequency modulations at the same time.In the diversity of distributed generation modes,the combined operation as well as mutual promotion of PV and energy storage(ES)has become an increasingly important developing trend.There have been so many advantages in the combined operation of PV and ES that it is not only a key means for distributed generations to actively participate in suppressing frequency fluctuations of power grids and effectively improving the safety and stability of the distribution network,but also an important direction to promote the wide distributed power consumptive abilities by solving the inherent randomness and fluctuations in the distributed generations.This topic is a key research point coming from the opening topic "Research of PV and Energy Storage Adaptive Control Technology for Active Distribution Network" of Qinghai Electric Power Research Institute.It is mainly to conduct corresponding research on frequency control for the distribution network with distributed PV and ES combined power generation system,whose purpose is to realize the active participation of distributed power generation system in power grid frequency regulation by studying and controlling the combined operation of PV and ES,thus making a positive contribution to promoting the stability of the distribution network.The major researches of this thesis are mainly divided into the following:Firstly,Making research on output power forecasting of PV.This thesis studies the optimization method of BP neural network model for PV output power forecasting based on genetic algorithm.Also,the thesis establishes the BP neural network model for PV output power forecasting after optimization.And then,making analysis and verification on theeffectiveness of the proposed output power forecasting method of PV based on weather classifications and a practical example in Qinghai.Secondly,Making research on the estimating strategy of the state of charge(SOC)of battery energy storage.Since the extended kalman filter(EKF)is susceptible to be disturbed by noise interference in the process of SOC estimating of battery energy storage,this thesis studies the dynamic correction and noise reduction model based on adaptive algorithm.And then,proposing a strategy for estimating the battery SOC after noise reduction optimization.Thirdly,Making research on frequency control strategy for distribution network with distributed PV and ES combined power generation system.In view of control modes of grid-connected inverter for distributed PV and ES combined power generation system,this thesis studies the nonlinear relational model between power shortage of distribution network,output of distributed PV,the SOC of battery energy storage and the droop coefficient of the inverter.Also,this thesis proposes a frequency control strategy for PV and ES combined power generation system based on variable droop coefficient.And then,making research on the establishment of a typical PSCAD model for distribution network system containing distributed PV and ES combined power generation system in Qinghai Province to verify whether the frequency control strategy is feasible and effective or not.