Study On Intelligent Control Strategy Of Energy Management System For Parallel Hybrid Electric Vehicle

The target of control strategy of energy management system for parallel hybrid electricvehicle is to achieve the optimal energy management and the best balance betweenefficiency and emissions by distributing torque of the engine and the motor reasonably.Logic threshold control strategy, instantaneous optimum control strategy, global optimumcontrol strategy and intelligent control strategy based on fuzzy logical or neural network areproposed in recent years. Only logic threshold control strategy which is based onengineering experience designing has been applied widely in real commercialization ofhybrid vehicles. The combination of logic threshold control strategy and intelligentalgorithms can achieve better control and distribution of energy, which has importantpractical significance.This work is supported by Open Fund Project of Jiangsu provincial key laboratory(QK09003) and the985project of Hunan University. Intelligent algorithm is integrated intothe logic threshold control strategy and the logic threshold control strategy of ADVISORsoftware will be improved. The main work and innovations of the thesis are as follows:(1) Logic threshold control parameters are optimized by multi-level parametric sweepsalgorithm. The simulations are carried on ADVISOR of the optimized vehicle and vehiclebefore optimized and the two simulation results are compared. The results show that fuelconsumption and emissions have been significantly reduced.(2) Simulated annealing particle swarm optimization algorithm (SAPSO) is built bycombining the advantages of particle swarm algorithm and simulated annealing algorithm.Control parameters of logic threshold control strategy are optimized by SAPSO algorithmand at the same time the simulations are carried on ADVISOR software. The results showthat SAPSO algorithm is superior to multi-level parametric sweeps algorithm on theoptimization of control parameters.(3) Fuzzy BP neural network control strategy is built by combining logic thresholdcontrol strategy and instantaneous optimum control strategy. A secondary development ofADVISOR software is implemented successfully. The results of simulation experimentshow that Fuzzy BP neural network control strategy can effectively improve the fueleconomy and reduce response time.