基于MPC的无级变速器控制优化策略研究

Study on Control Optimization Strategy of Continuously Variable Transmission Based on Model Predictive Control

金属带式无级变速器(Continuously variable transmission,CVT)夹紧力是影响CVT系统传动效率及整车燃油经济性的重要因素,针对传统控制策略难以准确、合理地获得最优夹紧力控制范围的问题,提出了基于模型预测控制(Modelpredictive control,MPC)的控制策略。根据传动系统运动学原理,建立CVT动态方程及状态空间表达式,设计CVT模型预测控制器。考虑夹紧力具体约束限制,将无约束优化问题转化为二次规划问题,利用预测控制器求解每一时刻的夹紧力最优控制序列,构造最优夹紧力输入。在CVT系统输出侧施加干扰转矩,结果显示系统鲁棒性良好。同时,为验证MPC控制策略的可行性,利用Simulink和AMEsim进行联合仿真,并在紧急加速、紧急制动及综合工况下进行实车验证。结果表明:相较于传统控制,基于MPC的控制系统能够将传动效率提高8.72%~9.23%、油耗降低5.61%,为进一步解决CVT控制问题提出新思路。

The clamping force of continuously variable transmission is an important factor affecting the transmission efficiency of Continuously variable transmission(CVT)system and fuel economy,the optimal clamping force control range is difficult to be obtained accurately and reasonably by the traditional control strategy,a control strategy based on model predictive control is presented.According to the kinematics principle of transmission system,the CVT dynamic equation and state space expression are built,and the MPC controller of CVT is designed.The specific constraint of clamping force is considered,and the problem of unconstrained optimization is transformed into a quadratic programming problem,the optimal clamping force input is got through the use of predictive controller to solve optimal control sequence of clamping force at each moment.The disturb torque is applied to output side of CVT system,test results show that system has good robustness.Moreover,in order to verify the feasibility of clamping force control strategy,co-simulation with Simulink and AMEsim,real vehicle is tested in emergency acceleration,emergency braking and integrated conditions.Results show that compared with the traditional clamping force control strategy,the transmission efficiency of the control system based on MPC is improved by 8.72%-9.23%,the fuel consumption is reduced by 5.61%,and a new method is proposed to solve the problem of the control of clamping force of CVT.