论文部分内容阅读
针对模型失配对模型预测控制算法的影响,提出了一种新的应用于单变量系统的非最小状态空间模型预测控制(NMSSPC)算法。该算法利用阶跃响应数据建立传递函数模型,将模型转换为状态空间的形式,通过模型预测控制滚动优化求得控制律。文中改进了控制量的滚动优化策略,在目标函数中引入了可调因子,并在反馈校正中对预测误差补偿进行了改进,有效的减小了模型失配时产生的误差。仿真结果表明了算法的有效性,并且具有良好的跟踪控制性能。
Aiming at the influence of model mismatch on model predictive control algorithm, a new non-minimal state space model predictive control (NMSSPC) algorithm for single variable system is proposed. The algorithm uses the step response data to establish the transfer function model, the model is transformed into the state space form, and the control law is obtained through the model predictive control rolling optimization. In this paper, the rolling optimization strategy of control variables is improved, the adjustable factor is introduced into the objective function, and the prediction error compensation is improved in the feedback correction, which effectively reduces the error caused when the model mismatch. The simulation results show that the algorithm is effective and has good tracking control performance.