高速高精运动平台模型参数最小二乘辨识研究

Model parameters identification of motion control platform with high frequency and fast response based on least squares method

针对高速高精运动控制对象模型参数辨识的问题,对辨识对象模型、最小二乘辨识算法和滤波器等内容进行了研究,提出了一种基于滤波器的最小二乘线性回归方程的辨识算法,即加入一个稳定的二阶传递函数,对系统输入输出数据进行了滤波处理,使辨识算法在噪声环境下能够减小失真且运行稳定。仿真与实验结果表明,最小二乘在有噪声的闭环运动控制系统中的辨识结果不收敛且失真,而加入滤波器的最小二乘算法能显著提高控制系统的稳定性及精确度,且满足其高速高精的要求。

Aiming at realizing the problem of the model parameters identification of motion control with high frequency and fast response,research on the identification model,least squares identification algorithm and filter,a least squares linear regression equation on the base of filter was used,namely adding a stable second-order transfer function,it was used in input and output data filter,could achieve the effect of stable recognition.The simulation and experimental results show that the identification data of least squares in noisy close-loop motion control system is non-convergent and distorted,while is used least squares algorithm with filter can significantly improve the stability and accuracy of the control system,and can meet the requirements of high frequency and fast response.