射电望远镜伺服系统的参数辨识与模型降阶

Parameter Identification and Model Reduction of Radio Telescope Servo System

应用基于系统模型的控制算法能够提高射电望远镜的指向精度和跟踪性能.因此,采用一种由相关分析法和特征系统实现算法组成的辨识方法来获得射电望远镜的参数模型.首先利用相关分析法对伪随机序列(PRBS)的特性进行分析,得到了系统的脉冲响应序列,依据由脉冲响应序列组成的Hankel矩阵的奇异值大小得到系统的模型阶次,并采用特征系统实现算法对Hankel矩阵进行转换,得到系统的参数模型.在保留系统主要动态特性的基础上,运用模型降阶技术对所得的传递函数进行降阶,得到一个低阶模型.最后以南山26米射电望远镜为平台进行模型辨识实验.实验结果表明:相对于子空间辨识方法,由本文提出的辨识方法得到的系统模型能够准确地描述26米射电望远镜的动态特性.

The application of advanced control algorithms based on the system model can improve the pointing accuracy and tracking performance of the radio telescope.To obtain the parametric model of the radio telescope,an identification method composed of correlation analysis and eigensystem realization algorithm is adopted in this paper.Using correlation analysis,the characteristics of the pseudo-random sequence(PRBS) are analyzed,and the impulse response sequence of the system is obtained.The model order of the system is obtained based on the magnitude of the singular values of the Hankel matrix composed of the impulse response sequence.And the Hankel matrix is transformed by using the eigensystem realization algorithm to obtain the parametric model of the system.Retaining the main dynamic characteristics of the system,the model reduction technique is used to reduce the order of the transfer function,and a low-order model is obtained.The Nanshan 26 m radio telescope is used as a platform to conduct model identification experiments.Our results show that:compared with the subspace identification method,the system model obtained by the proposed identification method can describe the dynamic characteristics of the 26 m radio telescope more accurately,and provide a valid nominal model for the next controller design.