【作者】 陈飚；

【导师】 黄妙华；

【作者基本信息】 武汉理工大学 ， 车辆工程， 2005， 硕士

【摘要】 能源危机与环境污染给世界汽车工业的稳定发展带来了严峻挑战。电动汽车作为新能源汽车的代表,为未来汽车发展指明了方向。电动汽车仿真技术的研究与应用则缩短了电动汽车的开发周期,节省了科研经费。本文以国家“863”电动汽车重大专项“多能源动力总成能量管理仿真分析系统”子课题为依托,围绕与电动汽车前向仿真系统及课题相关的内容展开论述与研究。 在电动汽车理论研究阶段,为确定较优的整车布置结构、部件规格尺寸与能量管理方案,可通过建立相应整车数字化模型,并对其动力性、经济性及排放性进行仿真,从而对各种备选方案的优劣进行比较。 前向仿真与后向仿真作为两种最典型的电动汽车仿真结构,无论是在建模方式或是仿真运算上都各有优劣。本文通过理论结合实例的比较分析,指出了后向仿真适用于设计初期,指导整车部件及控制策略选型;而前向仿真则更适宜于设计后期,针对应用于实车的部件参数与控制策略的优化改进。 作为电动汽车仿真研究的基础平台,仿真软件的开发与应用起着重要作用。针对课题研究需要,课题组以前向结构为构架开发出一套混合动力电动汽车仿真软件系统HEVSim,该系统为实车控制策略设计及优化提供了良好环境。 前向仿真与后向仿真在部件构成方面的唯一区别在于有无驾驶员模型。驾驶员模型作为前向仿真中的起始部件,向整车控制器直接发出控制指令。本文对驾驶员模型的建模方式作了详细介绍,针对仿真需求车速与仿真所得车速之差需进行实时修正的问题,分别提出了采用常规PI控制器以及模糊控制器进行修正的方法,并在HEVSim中对所建立起的相应模型进行仿真,在实现仿真要求的前提下,模糊控制器对车速偏差的修正效果优于PI控制器。 针对课题所涉及的EQ7200HEV新型并联式混合动力电动轿车实车控制策略的开发,提出了基于需求转矩分配的控制方案。以发动机稳态下的燃油消耗率特性图为基础,根据驾驶员在车辆运行过程中的需求转矩,对驱动装置进行工作转矩实时分配;另外,对于此控制策略中的不足,提出采用模糊控制器替换部分原有门限逻辑判断,并利用遗传算法对相关控制变量的隶属函数进行优化。在前向仿真软件HEVSim中对所开发的控制策略进行建模与仿真,其控制效果符合设计要求,同时验证了前向仿真便于实车控制策略开发与优化的结论。更多还原

【Abstract】 The crisis of energy sources and pollution of the environments have brought rigorous challenge for the stable development of the automobile industry of the world. As the representative of the automobiles with new energy, the electric vehicle indicates the development direction for the automobile of the future. The research and application of the simulation technique in the electric vehicle design shorten the design period and save the research outlay. This paper discusses and researches the correlative contents on the facing-forward simulation of the electric vehicle and the task that is the sub-task of "simulation analysis system of multi-energy-source powertrain energy management" of the 863 electric vehicle important project presided by government.At the electric vehicle theoretical research stage, in order to choose the best vehicle structure, parts size and energy management strategy, we can set up the corresponding whole vehicle digital model and simulate the dynamics performance, fuel economy performance and exhaust emission performance, sequentially compare the merits and shortcomings of different designs which can be chosen.As the two most representative simulation structures of electric vehicle, both the facing-forward simulation and the facing-backward simulation have their own merits and shortcomings regardless of the modeling method or the simulation operation. By the comparison analyzing that integrates theory with example, this paper points out that the facing-backward simulation suits to be used at the primary design stage to guide the parts and control strategy choosing, but the facing-forward simulation is more suited to adjust and optimize the parts’ parameters and control strategy to meet the application requirement of prototype vehicle at the late design stage.The development and application of simulation software play an important role as the basic platform of electric vehicle simulation research. Aiming at the demand of task research, task group developed a set of software called HEVSim for the hybrid electric vehicle simulation on the base of facing-forward simulation structure. This system provides a good environment for the design and optimization of the control strategy in the prototype vehicle.The unique component differentiation between the facing-forward simulationstructure and the facing-backward simulation structure is whether has a driver model. As the origination part of facing-forward simulation structure, the driver model sends control command to the controller directly. This paper introduces the modeling method of the driver model in detail. To the problem that real-time correct the difference between the demand vehicle speed of simulation and the acquire vehicle speed of simulation, this paper puts forward that use the conventional PI controller and the fuzzy logic controller to set up the driver model respectively. The corresponding modeling work is completed and the simulation has run in the HEVSim software. The simulation result shows the control effect of fuzzy logic controller is better than the PI controller’s on the premise of achieving the demand of simulation.To the development of the controller of the new parallel hybrid electric vehicle EQ7200HEV related to our task, this paper presents the control strategy based on demand torque distribution. On the base of the optimal engine curve of steady state, the strategy distributes the work torque timely to the drive devices according to the driver’s demand torque during the driving. At the same time, to the shortage of this control strategy, using fuzzy logic controller to substitute for the partial threshold judgments is presented and the membership function of some control variable in the fuzzy logic controller are optimized by genetic algorithm. Building the controller model with the developed control strategy and running it in the facing-forward simulation software HEVSim, the control effect satisfy the design requirement and the conclusion that facing-forward simulation is more convenient for the development and optimiza更多还原