Energy Control System Optimization Of Hybrid Vehicle Based On Genetic Algorithm

Hybrid electric vehicle is a kind of highly efficient energy-saving andenvironment-friendly vehicles which gets more and more attention of the internationalcommunity. United States and Japan and other countries have made an example for thedevelopment of hybrid vehicles. The huge success Toyota Prius obtained showed thatthe hybrid vehicle technology is widely accepted by people. The vehicle controlstrategy as one of the core technology of hybrid electric vehicle is the hot area for theresearch institutions and universities research. Based on the analysis of the researchstatus of different control strategies and the advantages and disadvantages of thosestrategies, a fuzzy control strategy optimized by improved genetic algorithm ispresented which can be used on hybrid electric vehicle. The main work of this thesis is:①By analysis energy saving mechanism of a hybrid electric vehicle, found thatenergy management strategy has a very large impact on the performance of vehicleimpact. The domestic and overseas present development situation on energymanagement strategy for hybrid electric vehicle was researched. Two intelligentalgorithms which are genetic algorithm and fuzzy logic were determined to be used toachieve vehicle energy management strategy.②The parallel hybrid electric vehicle was chose to be the research object. In orderto determine the model and parameters of main components in the power system ofhybrid electric vehicle, traditional automobile power performance parameters wasreferenced. Above which the basic knowledge of automobile theory and dynamics wasneeded. And then the vehicle and power system model was built based on ADVISORsoftware by using the backward simulation and forward simulation theory.③Based on the vehicle model established, the working model of hybrid car wasanalysis, and the method for realizing control strategy of the power system was presentwhich was using fuzzy controller with two inputs and one output. According to thefuzzy control theory and the characteristics of vehicle models, the membership functionof fuzzy controller and control rule and other related parameters were set, and thenestablished the control strategy of vehicle model.④The control rules of the fuzzy control strategy have a stronger subjectivedependence, and the control results can’t be optimal, so genetic optimization which wasused for optimization of control rules was present. According to the theory of genetic algorithm, vehicle fuel consumption and emissions were selected as the objectivefunction, and then determine the fitness function by using the weighted method. In orderto simplify the coding method, and better genetic operation process, decimal integercoding was used for variable coding rules. Genetic optimization results showed that itwas reach the global optimum when the30generation or so, and new control rules wasgot after optimization.⑤The simulation results were compared before and after optimization, theoptimized control strategy has a better economic and emission performance thanoptimization without optimization. What’s more, the fuel consumption decreased by14.5%, and emissions values were decreased by7.4%（HC）、10.7%（CO）、16.4%（NOx）.