液压伺服系统状态估计和参数辨识的鲁棒算法

针对液压伺服系统不易进行状态估计和参数辨识的问题,提出了一种鲁棒算法,把液压伺服系统的动态行为当作一个具有时变参数的线性随机状态空间模型来描述,把故障当作系统参数变化,将参数公式中重要项进行泰勒级数展开,推导线性状态方程和线性测量方程,从而得出状态向量和参数向量的估计。在液压伺服系统中实验结果表明:该鲁棒算法能很好地对液压伺服系统进行状态估计和参数辨识;并且相比于其他算法,收敛速度快,对非高斯噪声和系统参数故障的存在敏感性较低,鲁棒性好。

Robust Algorithmfor State Estimation and Parameter Identification of Hydraulic Servo System

In order to solve the problem of state estimation and parameter identification in hydraulic servo system, a Robust Algorithm (RA) is proposed in the paper. In the robust algorithm, the dynamic behavior of the hydraulic servo system is described as a linear stochastic state space model with the time-varying parameters, the fault is regarded as the change of system parameters, and the important terms in the parameter formula are expanded by Taylor series. The linear state equation and linear measurement equation are derived, and the estimation of state vector and parameter vector is obtained. The experimental results in the hydraulic servo system shows that the robust algorithm could estimate its state estimation and parameter identification very well. Compared with other algorithms, the robust algorithm has faster convergence speed, lower sensitivity to non-gaussian noise and system parameter faults, good robustness.