基于BP神经网络的复合储能式装载机工况识别及控制器匹配  

Working Condition Identification and Controller Matching of Composite Energy Storage Loader Based on BP Neural Network

作　　者：李钊 林慕义[1,2] 曹海岐 陈勇 LI Zhao;LIN Mu-yi;CAO Haiqi;CHEN Yong(School of Mechanical and Electronic Engineering,Beijing Information Science and Technology University,Beijing 100192;Collaborative Innovation Center of Electric Vehicles in Beijing,Beijing 100192)

机构地区：[1]北京信息科技大学机电工程学院,北京100192 [2]北京电动车辆协同创新中心,北京100192

出　　处：《液压与气动》2020年第2期36-43,共8页Chinese Hydraulics & Pneumatics

基　　金：国家自然科学基金(51275053);科技创新服务能力建设-科研基地建设-新能源汽车北京实验室(市级)(PXM 2019_014224_000005)

摘　　要：复合储能式装载机具备多种能量回收再利用功能,但由于工作工况复杂多变,无法选择与之匹配的最优控制器,使得其控制性能和经济性能并未达到最优。利用基于BP神经网络的识别模块进行工况识别,并依此匹配相应控制器,然后通过整车仿真模型进行仿真,结果表明,整车控制性能和经济性能均得到显著提升。通过dSPACE进行硬件在环试验,试验与仿真结果基本一致,验证了优化切实有效,为整车控制器的设计优化提供了参考。The composite energy storage loader has a variety of energy recovery and reuse functions,because of the complex and variable working conditions,it is impossible to select the optimal controller to match it,so that its control performance and economic performance are not optimized.In this paper,the recognition module based on BP neural network is used to identify the working conditions,and the corresponding controller is selected in this way.Then use the whole vehicle simulation model to simulation.The results show that the control performance and economic performance of the vehicle are significantly improved.The hardware-in-the-loop test is carried out by dSPACE.The test and simulation results are basically consistent,which verifies that the optimization is effective and provides a reference for the optimal design of the vehicle controller.

关 键 词：能量回收 BP神经网络 工况识别 控制策略 硬件在环 

分 类 号：TH137[机械工程—机械制造及自动化] TP391.4[自动化与计算机技术—计算机应用技术]