液压伺服系统建模的新方法

针对液压伺服系统中直接用测量的方法来建立伺服阀死区非线性的模型非常困难的现状，该文充分利用了现有的对液压伺服系统模型的认识，用RBF神经网络来代替伺服阀锴区非线性部分的模型，设计了一个带神经网络的辨识器。该辨识器采用了迭代最小方差（RLS）学习算法对系统进行建模，最后，用Matlab/Simulink中的S－Function模块实现了上述辨识器的编程，并进行了仿真。仿真结果表明，所设计的辨识器能较好的解决液压伺服系统的建模问题。

A New Method for Modeling the Hydraulic Servo System

It is difficult to measure the flow of the valve of the hydraulic servo system, and so we are difficult to model the non-linear valve by measuring. we design an identifier with an artificial neural network, in this identifier we use the priori knowledge of this hydraulic system and use a RBF neural network to instead the model of the non-linear dead zone. A standard recursive least-squares algorithm is used to estimate the parameters of the identifier. The identifier is fulfilled by programming S_Function module in the Matlab/simulink at last. The simulation shows that the identifier can overcome the problem of modeling the hydraulic system.