论文部分内容阅读
针对存在噪声干扰与时变特性的线性系统的模型不确定性问题,提出了一种基于递推闭环子空间辨识的自适应预测控制方法.通过结合PID(proportional-integral-derivative)控制采用新的目标函数,对闭环子空间预测控制算法进行改进,推导出具有类似PID结构的闭环子空间预测控制算法;采用固定输入输出数据集大小的递推方法将改进后的算法在线实施,通过采用一种简单直观的更新方法代替LQ分解,有效提高了在线计算效率.最后,通过仿真实验验证了方法的有效性.
Aiming at the model uncertainty problem of linear systems with noise interference and time-varying characteristics, an adaptive predictive control method based on recursive closed-loop subspace identification is proposed. By using PID (proportional-integral-derivative) control, Objective function, the closed-loop subspace prediction control algorithm is improved, and a closed-loop closed-loop subspace prediction control algorithm with similar PID structure is deduced. The modified algorithm is implemented online using the fixed input and output dataset size recursion method. The simple and intuitive updating method replaces LQ decomposition, which effectively improves the online computing efficiency.Finally, the simulation results show the effectiveness of the method.