Research On Capacity Optimization Of Hybrid Energy Storage System Considering Wind Power Characteristics Of Fluctuation And Anti-Peak Shaving

Energy exhaustion and environment pollution enforce people to turn their attention to clean energy.There are abundant wind energy resources in China and wind energy has immense potential for development and utilization.However,its main disadvantages such as fluctuation,anti-peak shaving and so on,which are different from conventional energy resources,will affect the frequency,peak shaving,power quality and reliable operation of the power system.Utilizing the characteristics of energy storage system to absorb and release energy can reduce the defects of wind power and improve the utilization efficiency of wind energy.However,the energy storage system is very expensive,so it is necessary to optimize its capacity.From the perspective of wind farms,the fluctuation can be smoothed and anti-peak shaving characteristics can be alleviated by optimizing the capacity of hybrid energy storage system,while taking into account the economic benefits of wind farms.According to the sampled data of wind power and load,the fluctuation and anti-peak shaving characteristics of wind power are analyzed,and a hybrid energy storage system consisting of power-type and energy-type storage system are proposed.On this basis,the mechanisms of using power-type storage system to smooth high-frequency fluctuation and energy-type storage system to reduce the anti-peak shaving characteristic are revealed.Meanwhile,a decoupling strategy of wind power based on wavelet transform theory is proposed.On the basis of wavelet transform theory,the wavelet basis function and decomposition levels are determined by entropy method and fluctuations in this thesis.Therefore,the high-frequency component and low-frequency component of wind power can be obtained.For high-frequency component,super-capacitors are selected as power-type storage system,and an optimal model is proposed which based on the fluctuations.In the example,the low pass filtering,moving average filtering and wavelet decomposition are compared in detail,and the results show that the capacity obtained by wavelet decomposition is the smallest and the economy is the best.In terms of the anti-peak shaving characteristic,the low-frequency component of wind power is integrated with the output of LiFePO₄ battery to participate in the daily dispatching plan.Taking the harm of over charge and discharge to lifespan into account,an optimal control strategy of stage of charge for LiFePO₄ battery to extent cycle life is raised.Then,an optimal model of energy-type storage system with the maximum profits of wind farms is established,and the model is solved by CPLEX.The example results show that the energy storage system can reduce anti-peak shaving characteristic of wind power and increase the grid-connected power of wind farms.