Secondary Frequency Control Of Islanded Microgrid Considering Wind And Solar Stochastics

Under the background of " carbon emissions peak and carbon neutrality ",the penetration of new energy in the power system is increasing,but the power generation characteristics,the frequency regulation ability and inertia response ability of the system are reduced,which brings challenges to the system frequency control.However,if standby control is adopted for wind and photovoltaic power generation,existing some power in new energy,it can participate in frequency regulation and improve the system frequency control capability.This paper focuses on the research of model predictive control method that considers the participation of new energy power reserve control in secondary frequency regulation for wind and photovoltaic stochastic distributed generation,and the main work is as follows.An island microgrid load frequency control model with the participation of new energy in frequency regulation is established.Based on the characteristics of each distributed power generations,the mathematical model of diesel engine,which can reflect the frequency regulation characteristics in the traditional power system,is firstly constructed,and then the model of the generator-power system is derived,and then the load frequency control model of the traditional single-area power system is constructed;what’s more,the models of photovoltaic(PV)arrays and wind turbines(WT)are obtained,and the new energy power generations are controlled by power reserve,leaving power to participate in system frequency adjustment.The PV array is controlled by changing the tracking voltage to achieve power reserve control,and the WT is controlled by changing the rotor speed and the pitch angle to achieve power reserve control;next,the mathematical model of the energy storage system is derived.A model predictive controller considering the stochastic of WT and PV output is designed for application to island microgrid.An extended state matrix containing stochastic power perturbations is established and Kalman filtering is used to estimate the stochastic unknown distribute generation.The state space expression can be obtained from the previously derived island microgrid model.The unknown state variables and known state variables in the expression are decoupled using mathematical methods to provide preliminary estimate of the unknown variables using the known variables of the system and the controller output.Variable constraints are further set to take into account random changes in the scenery.Based on the maximum available power of WT and PV,real-time variable constraints are established to avoid the power of crossing the limit.Reasonable weighting coefficients are set to ensure that new energy resources can be fully exploited,so that the priority output of new energy,WT and PV can participate in secondary frequency regulation.The estimated unknown variables are fed into the model prediction link in the model prediction controller to improve the control accuracy of the controller.The model predict control objective function is designed,which is zero to frequency deviation,and the objective function with constraints is transformed into a quadratic programming problem to obtain the optimum power for each unit.A series of optimal values in the predicted time domain are found,and only the first optimal value is input into the microgrid system before the next rolling optimization to ensure that the system is optimally controlled at all times.A microgrid model containing several photovoltaic and wind power sources is established and simulated in comparison with conventional secondary frequency regulation methods under different scenarios.The effectiveness of the proposed control strategy in the frequency control response of island microgrids is investigated.The simulation results show that the proposed method can improve the system frequency recovery speed and reduce the system frequency fluctuation especially under the scenario of severe fluctuation of wind and photovoltaic power generation.The performance and advantages of the proposed model predictive control strategy considering stochastic power output to participate in the frequency response of the microgrid system are verified in different scenarios.