Research On Hybrid Energy Storage System To Stabilize Wind Power Forecasting Error

With the rapid development of science and technology and the increasing scarcity of traditional fossil fuels,the research and use of new energy sources have received widespread attention from human beings.Wind energy has become one of the important resources for energy transformation due to its abundant resources and mature development technology.Wind energy is converted into daily electricity through wind power generation technology,which greatly reduces the burden of traditional energy.However,due to the randomness and intermittence of wind power generation,if wind power is directly connected to the grid,it will cause great impact on the power grid and affect the power quality.Therefore,it is of great practical significance to study and improve the stability of wind power grid connection.Firstly,the access mode of hybrid energy storage system is studied and analyzed.The hybrid energy storage system is connected in parallel to the AC bus of wind farm in a centralized access mode.Secondly,the different connection modes of lithium battery and super capacitor in hybrid energy storage system are analyzed.The topology of each DC-DC converter in parallel is used as the simulation basis.According to the equivalent circuit model of lithium battery and super capacitor,the charging and discharging of hybrid energy storage system are simulated and analyzed.It is verified that lithium battery and super capacitor can deal with power changes according to their own instructions.Finally,according to the energy flow relationship of the wind-storage combined system,the wind power forecasting value is taken as the grid-connected target,and the wind power forecasting error is compensated and released by the hybrid energy storage system to reduce the grid-connected fluctuation and provide a certain method for enhancing the stability of wind power grid connection.In terms of wind power forecasting technology,a CNN-LSTM hybrid forecasting model constructed by Convolutional Neural Networks(CNN)and Long Short Term Memory(LSTM)is used.The LSTM and CNN algorithms are used to forecast the power of a wind farm in Xinjiang,and the forecasting results are compared with the results of the CNN-LSTM hybrid forecasting model.The results show that CNN-LSTM has higher forecasting accuracy.In order to solve the influence of sample data on CNN-LSTM forecasting results,a Variational Mode Decomposition(VMD)data processing method is proposed to optimize the CNN-LSTM hybrid model.The example results show that the CNN-LSTM forecasting accuracy is further improved.In terms of stabilizing wind power forecasting error of hybrid energy storage system,according to the characteristics of lithium battery and super capacitor,the low frequency component and high frequency component of stabilizing wind power forecasting error of lithium battery and super capacitor are determined by combining Aquila Optimizer(AO)with Fourier transform,and the primary power distribution of lithium battery and super capacitor energy storage equipment is realized.In order to avoid the over-charge and over-discharge behavior of energy storage equipment,the SOC state of charge limit strategy is further implemented for lithium battery and super capacitor energy storage equipment to realize the secondary distribution of power.The reasonable distribution of power ensures the reasonable operation of the hybrid energy storage system while suppressing the wind power forecasting error.