一种多输入单输出Hammerstein系统的集成辨识方法

Integrated identification method for multi-input/single-output Hammerstein system

针对多输入单输出(MISO)Hammerstein系统提出了一种稳态与动态辨识相结合的集成辨识方法.该方法利用稳态信息获取稳态模型的强一致性估计,并通过稳态模型以神经网络获得其非线性逼近函数,再利用动态信息辨识获取多输入单输出(MISO)Hammerstein系统的线性子系统未知参数的一致性估计.仿真结果表明了该方法的有效性和实用性.

An integrated identification method combining steady-state and dynamic identification is introduced for the multiinput/single-output（MISO） Hammerstein system. The strong consistent estimates of steady-state model are obtained by using steady-state data. The neural network can be used to approach nonlinearity function by using steady-state model. The consistent estimation of the linear unknown parameter of the the multi-input-single-output（MISO） Hammerstein system are obtained. The efficiency and applicability of this estimation technique is demonstrated by simulation results.