Font Size: a A A

Research On Energy Management Strategy Of Series-parallel Hybrid Electric Vehicles Based On Model Predictive

Posted on:2019-04-20Degree:MasterType:Thesis
Country:ChinaCandidate:X F ShiFull Text:PDF
GTID:2382330566477803Subject:Vehicle Engineering
Abstract/Summary:PDF Full Text Request
The energy management strategy,which has great impact on the fuel economy of the vehicles,is the core part in the Hybrid Electric Vehicle(HEV)research.Due to the model predictive control is very suitable for solving nonlinear and uncertain problems,and can be combined with different optimization algorithms to obtain the optimal solution in a short time,it can be applied to the energy management of HEV,and there is a great improvement in the fuel economy,at the same time,it has a certain potential for real-time application.Focusing on Toyota Prius,the Model Predictive Control(MPC)energy management strategy of hybrid electric vehicle was emphatically searched.(1)The working principle of the hybrid system was analyzed,Several working modes in the hybrid system were analyzed by using the lever method.The dynamic models of each working mode was performed and the backward simulation model of the HEV was established.(2)The paper established a rule-based energy management strategy and a global optimal energy management strategy: Dynamic Programming(DP),and performed simulation analysis under UDDS cycle conditions.Due to the impact of human experience on the use of real vehicles in the rule-based energy management strategy,and the existence of “dimension disasters” in the global optimization energy management strategy that cannot be applied to real-time application,energy management strategy based on MPC was researched and its control principle and hierarchical prediction control method were introduced.(3)Combined with the dynamic programming algorithm,the energy management strategy of HEV based on model predictive control was established,and the methods for improving the prediction accuracy was researched.The first-order,second-order and third-order Markov as well as the Radical Basis Function(RBF)neural network and the Generalized Regression Neural Network(GRNN)were applied respectively to do the cycle prediction and the effect of each method was compared.By comparing the simulation results of energy managements based on different predictive methods,the GRNN neural network predictive method can improve at least 4.4% fuel economy and this certifies the effectiveness of the prediction precision on fuel economy improvement.(4)Optimal speed model was established based on the traffic light,the GRNN neural network was used to predict the front vehicle speed,the traffic light and the front vehicle speed was used to solve the target vehicle speed,which is also called the back vehicle,and the MPC energy management strategy of the front and back vehicles based on the traffic light was simulated and analyzed.The result shows that compared with the ordinary HEV(the front car),the HEV with communication function has better trafficability characteristic and fuel economy.Through the research of the paper,the energy management strategy based on model predictive control has a good effect on improving the fuel economy of HEV,and has a certain potential for real-time application.
Keywords/Search Tags:Energy Management, Model Predictive, Markov, Neural Network, Traffic Light
PDF Full Text Request
Related items