重度混合动力汽车油耗和排放的多目标随机最优控制策略  

Multiple-Objective Stochastic Optimal Control Strategy for the Fuel Consumption and Emissions of Full Hybrid Electric Vehicle

作　　者：隗寒冰[1,2] 秦大同[1] 陈淑江[1] 

机构地区：[1]重庆大学,机械传动国家重点实验室,重庆400044 [2]重庆交通大学机电与汽车工程学院,重庆400074

出　　处：《汽车工程》2014年第8期923-928,949,共7页Automotive Engineering

基　　金：国家自然科学基金(51305472);重庆交通大学高层次人才科研启动经费项目资助

摘　　要：鉴于采用通常在特定工况下开发的控制策略的混合动力汽车在实际路况下的性能未必能达到最优的问题,将实际道路下的混合动力汽车能量管理策略问题转化为标准路况下的随机线性最优控制问题。建立了包含三效催化转化器热状态的混合动力汽车二次型状态空间方程,以发动机燃油消耗和三效催化转化器出口处的排放最小为优化目标,对蓄电池SOC、车速、三效催化转化器温度和出口排放等实际状态进行卡尔曼滤波估计,以对电机功率和发动机功率等输出变量进行最优反馈。仿真结果表明,与规则控制相比,所建立的随机最优控制策略能在满足车辆动力性要求的前提下,三效催化转化器的起燃时间约缩短160s,HC和CO的转化率明显提高。In view of the problem that the hybrid electric vehicle（ HEV） adopting the control strategy commonly developed with specific driving cycle may not necessarily achieve optimal performance in real operation condition,the issue of energy management strategy for HEV in real road condition is converted into the issue of stochastic linear optimal control. The quadratic state-space equation for HEV is set up with consideration of the thermal state of three-way catalytic converter（ TWC）,and with minimizing the fuel consumption of engine and the emissions at the exit of TWC as optimization objectives,a Kalman filtering estimation is conducted on the real states including battery SOC,vehicle speed,TWC temperature and the emissions at TWC exit,so as to do the optimal feedback for output variables such as motor power and engine power. The simulation results show that compare with rule control,the linear quadratic Gaussian（ LQG） optimal control strategy set up can shorten the light-off time of TWC by around160 s and increase the converting rate of HC and CO obviously,while meeting the power performance requirements of vehicle.

关 键 词：混合动力汽车 油耗 排放 线性二次型高斯最优控制 三效催化转化器 

分 类 号：U469.7[机械工程—车辆工程]