【2006.06.06】【ANFIS】基于模煳神经网络的溷合动力汽车控制策略仿真.pdfVIP

18 5 系 统 仿 真 学 报© Vol. 18 No. 5 第 卷第 期 2006 年 5 月 Journal of System Simulation May ， 2006 基于模糊神经网络的混合动力汽车控制策略仿真 钱立军，袭著永，赵 韩 ( , 230009) 合肥工业大学机械与汽车工程学院 安徽合肥 摘 要：为保证多能源系统转矩的合理分配，建立了模糊逻辑控制模型，并采用ANFIS 优化算法， 对建立的混合动力汽车模糊控制模型进行优化。通过对ANFIS 神经网络模型进行训练、测试和仿 真分析，结果表明：模糊控制模型的隶属函数得到了优化，将优化模型应用于整车仿真，与模糊逻 辑控制策略相比，燃油经济性提高5.4%。 关键词：混合动力电动汽车；控制策略；优化；神经网络；仿真 中图分类号：TP13 ；U469 文献标识码：A 文章编号：1004-731X (2006) 1384-03 Simulation of Hybrid Electric Vehicle Control Strategy Based on Fuzzy Neural Network QIAN Li-jun, XI Zhu-yong, ZHAO Han (School of Mechanical and Automotive Engineering, Hefei University of Technology, Hefei 230009, China) Abstract: In order to ensure the torque allocation of multi-power sources rationally, the fuzzy logic control mode was proposed. The fuzzy control model of the HEV was optimized based on ANFIS algorithm and then the training, test and simulation analysis were performed for the ANFIS neural network model. The results prove that the membership functions of fuzzy model obtain the satisfactory effect. Applying the optimization model in the simulation, the fuel economy of SQR-HEV raises 5.4% compared to the fuzzy logic control strategy. Key words: HEV; Control strategy; Optimization; Neural Network; Simulation 引 言1 1 多能源系统模糊逻辑控制模型介绍 1.1 控制目标 混合动力汽车(HEV)是指采用两种或两种以