分段非线性Hammerstein模型动态分离辨识方法

对具有分段非线性特性的Hammerstein系统,可采用动态分离重构中间输入的辨识方法.即通过引入开关函数,将线性动态部分分离处理,辨识得到新的线性模型.基于此模型,利用L延迟逆动态滤波器重构出系统的中间输入.最后由系统测试输入和重构中间输入,通过多项式回归估计出非线性的参数.数字仿真结果表明该方法有效实用.

Dynamical Separation Identification Method for Piecewise Nonlinear Hammerstein Model

A dynamic separating and intermediate input reconstructing identification method can be used for Hammerstein system with piecewise nonlinearities. That is by introducing the switch function, the linear dynamical part is separated by a simple algebraic operation, then a new linear model can be achieved. Based on the new model, L-delay inverse filter is used to reconstruct the intermediate input between the nonlinear pan and linear pan. Finally, the nonlinear term is identified by polynomial regression procedure using the test input signal and the reconstructed intermediate input. Numerical simulation results show the effectiveness and good performance of the proposed algorithm.