基于随机梯度的双率系统自校正控制方法

针对输入更新频率是输出刷新频率整数倍的未知参数双率系统，设计一个损失输出估计器计算采样间输出，再根据随机梯度算法设计参数估计器并得到系统模型的估计参数，基于最小方差控制原则设计出双率系统的自适应控制器。通过与基于最小二乘方法辨识系统参数的自适应控制算法进行比较，可以看出该算法的计算量较小，尤其是在输入数据更新频率与输出数据刷新频率相差较大时，计算量的差距更加明显。最后用仿真例子说明了该算法的有效性。

Self-tuning control for dual-rate systems based on stochastic gradient

For parameters unknown dual-rate systems with different input and output updating frequency sampling periods, a missing output estimator is designed to get the inter-sampling output data. By using the measured input-output data and estimated output data, a stochastic gradient algorithm based estimator is proposed to obtain the parameters of plants, and furthermore, a minimum variance based self-tuning controller is designed for pants. Comparing to the computation load of least squares based self-tuning algorithm for dual-rate systems, the new algorithm dominates a certain advantage, especially under a larger multiple circumstances. Finally, simulation example illustrates the efficiency.