Research On Short-term Energy Prediction And Capacity Optimization Of Wind Solar Storage System In Urban Rail Transit

With the rapid development of urban rail transit technology,its mileage has increased year by year.By December 2021,the mileage of urban rail transit in China has exceeded 8000 kilometers,and about 50 cities have opened urban rail transit systems.While urban rail transit is convenient for us to travel,its energy consumption is also huge.At present,the development of new energy power generation based on solar energy is rapid.Using solar energy to supply power for urban rail transit can effectively alleviate the problem of primary energy consumption and achieve the goal of energy conservation and environmental protection.Although the two have the characteristics of time complementarity,due to the intermittent output of new energy,volatility will lead to unstable power supply,which will seriously affect the normal operation of urban rail transit,The introduction of hybrid energy storage system can effectively solve this problem.At present,the application of solar energy storage in urban rail transit is still in the preliminary stage.This paper mainly studies the spectral output prediction and the optimal configuration of hybrid energy storage capacity.The main contents are as follows:Forecasting the output power of the wind power plant can effectively ensure the safe and stable operation of the power grid,facilitate the allocation of electric energy by relevant departments and prevent energy waste.This paper first analyzes the working principle of the wind power station,establishes the corresponding mathematical model,and analyzes the output power of the two.Because there are too many meteorological factors affecting the output of the wind power station,Kendall rank correlation coefficient and principal component analysis correlation index are used to pre-process the collected data of the wind power station and select the main influencing factors.EMD and VMD are used to decompose the output power of the wind and the wind respectively,and several intrinsic mode functions are decomposed,and comparative analysis is made to prove the superiority of the variational mode decomposition.The BP neural network method of ant colony optimization is used to predict the IMF components respectively.The IMF prediction components are integrated and compared with the time series algorithm and BP neural network.The advantages and disadvantages of various methods are described in detail to lay a foundation for the future hybrid energy storage capacity configuration in the following text.The introduction of energy storage system can enable urban rail transit to realize real local absorption,energy conservation and emission reduction of wind energy.This chapter introduces the hybrid energy storage system(HESS)composed of lithium batteries and supercapacitors to suppress and absorb the fluctuating characteristics of wind and solar output and traction load,and uses wavelet packet decomposition technology to decompose and reconstruct the traction load and wind output power signals in multi-scale to obtain low-frequency wind and solar grid-connected power and medium-high frequency components,and uses batteries and supercapacitors to absorb the medium-frequency and high-frequency components respectively.The minimum comprehensive cost of the hybrid energy storage system is taken as the goal,and the state of charge(SOC)and power limit of the hybrid energy storage system are the constraints.The differential evolution particle swarm optimization(DE-PSO)algorithm with shrinkage factor is used to solve the optimization problem to minimize the annual comprehensive cost of the hybrid energy storage system and optimize the internal power capacity configuration.Finally,based on the wind power and photovoltaic power data of a certain day,the results show that this method can effectively suppress the wind power fluctuations and effectively supply power for urban rail transit.