Research On Energy-saving Optimal Control Of Urban Rail Trains Based On Regenerative Braking

Urban rail transit trains have frequent start and stop,high traction energy demand and high regenerative braking energy.Nowadays,the energy consumption expenditure of the traction power supply system of urban rail transit accounts for a considerable part of the expenditure of operating enterprises,and the full application of regenerative braking energy is of great significance.The application of supercapacitor energy storage devices and adjustment of train operation diagrams can effectively utilize regenerative braking energy.Supercapacitors deploy regenerative braking energy in the time dimension,and adjust train schedules so that regenerative braking energy flows from brake trains to traction trains.Dimensional deployment of regenerative braking energy.This thesis discusses the application of supercapacitor energy storage system,designs supercapacitor intelligent control strategy and optimizes its parameters,analyzes the impact of supercapacitor and line schedule on line energy consumption,and uses intelligent algorithms to supercapacitor parameters and line Timetable integration optimization.The work done in this article mainly includes:(1)A model of urban rail transit traction power supply and supercapacitor energy storage system is establish,use small signal analysis method to derive the transfer function of supercapacitor during charging and discharging,and adopt double closed-loop control strategy as supercapacitor control strategy.(2)The traction power supply system of urban rail transit is a complex time-varying system,and the line schedule and line regenerative braking utilization rate have a very complicated coupling relationship.Through simulation experiments,the effects of supercapacitor energy storage system,line departure interval,and train stopping time on line traction station energy consumption were verified.Experiments show that the supercapacitor can allocate regenerative braking energy in time,and the train can increase the overlap time between traction and braking between trains by adjusting the timetable to improve the utilization rate of regenerative braking energy.(3)Aiming at the problems of DC traction network voltage fluctuations and stored regenerative braking energy caused by train traction braking,a dual population immune clone selection algorithm(DPICSA)optimized urban rail train supercapacitor fuzzy neural network(FNN)is proposed)Control Strategy.First,DPICSA is used to coordinate and optimize the membership function and quantization and scale factors of the main fuzzy controller;on this basis,a fuzzy parameter self-corrector is designed to adjust the quantization and scale factors online;finally,two RBF neural networks are used The network memorizes main fuzzy inference and parameter self-tuning fuzzy inference separately,and uses neural network high-speed parallel distributed computing capabilities to speed up fuzzy inference.Through simulation experiments in three different scenarios,it is verified that the strategy is superior to genetic algorithm-optimized PI control and conventional fuzzy control in suppressing network pressure fluctuations and energy saving.(4)Aiming at the demand for low line energy consumption,short running time,and small investment.Convert line energy consumption and investment into line operating cost.For the two goals of operating cost and running time,the line supercapacitor parameters and line departure interval 3.Multi-optimization of train stopping time to get Pareto frontier solution.The effectiveness of the NSGA-Ⅱ algorithm is verified by experiments.