有色噪声干扰下Hammerstein非线性系统两阶段辨识

针对实际工业过程中普遍存在有色噪声，提出了有色噪声干扰下Hammerstein非线性系统两阶段辨识方法。采用设计的组合式信号实现Hammerstein系统各模块参数辨识分离，简化了辨识过程。在第一阶段，基于可分离信号的输入输出数据，利用相关分析算法估计线性模块参数，减少了有色噪声对辨识的干扰。在第二阶段，基于随机信号的输入输出数据，在最小二乘算法中引入滤波技术，推导了滤波递推增广最小二乘算法，提高了非线性模块参数和噪声模型参数的辨识精度。仿真结果表明：提出的两阶段辨识方法提高了辨识精度，有效地抑制了有色噪声的干扰。

Two-stage Identification of Hammerstein Nonlinear System Corrupted by Colored Noise

In view of the color noise in actual industrial processes, a two-stage identification method of the Hammerstein nonlinear system corrupted by colored noise is proposed. The combined signals are used to separate the parameter identification of nonlinear and linear blocks for the Hammerstein system, which simplifies the identification process. In the first stage, based on the input and output data of separable signals, the parameters of the linear block are identified by adopting a correlation analysis algorithm, which reduces the impact of the unknown colored noise term on identification. In the second stage, based on the input and output data of random signals, the filtering technology is introduced into the least squares algorithm, and the filtering-based recursive extended least squares algorithm is derived, which improves the identification accuracy of nonlinear block and noise model parameters. The simulation results show that the proposed two-stage identification method improves the identification accuracy and effectively suppresses the interference of colored noise.