递推广义预测自校正控制器

RECURSIVE GENERALIZED PREDICTIVE STC

本文给出了广义预测自校正控制器的递推算法,用递推平方根法代替最小二乘法估计参数,改善了计算精度,比牛津大学克拉克等人的算法节省了计算时间.仿真实验表明,该控制器对系统时延与阶次有很好的鲁棒性,性能优于PID调节器、广义最小方差自校正器与极点配置自校正器,有较强的抑制干扰能力.

In this paper, a recursive algorithm for generalized Predictive STC(RGPC) is developed. The calculation needed is less than that of the GPC algorithm of Clarke et al. Using recursive squares root method insteat of recursive least squares method. The RGPC has higher precision than GPC. Simulation shows that RGPC has a strong robustness to the change of dead-time and model order. It is superior to the accepted techniques such as PID, STC and Pole-placement. It has a strong robustness to disturbance.