基于可分离组合式信号的Hammerstein 输出误差滑动平均系统辨识研究

THE IDENTIFICATION OF SEPARABLE COMBINED SIGNALS BASED NEURO-FUZZY HAMMERSTEIN OUTPUT ERROR MOVING AVERAGE SYSTEMS

考虑实际生产过程中的不可测噪声,提出一种基于可分离组合式信号的Hammerstein输出误差滑动平均系统辨识方法.通过可分离组合式信号源实现了Hammerstein输出误差滑动平均系统中静态非线性环节和动态线性环节的分离,解决了中间不可测变量的估计问题.基于辅助模型辨识思想和相关分析法估计Hammerstein输出误差滑动平均系统中静态非线性和动态线性环节的参数,避免了采用迭代法辨识Hammerstein模型时存在模型参数初始化和收敛性难以证明的问题,并有效补偿噪声信号的干扰.仿真结果验证了上述方法的有效性.

Considering the unpredictable noises in the actual production process, an identification method for Hammerstein output error moving average systems based on separable combined signals is presented in this paper. A separable combined signal is adopted to identify the Hammerstein output error moving average system, resulting in the identification problem of the static nonlinear element separated from that of dynamic linear part, and solving the problem of unpredictable variables estimation. Based on auxiliary model identification idea and correlation analysis algorithm, the static nonlinear and dynamic linear parameters of Hammerstein output error moving average system can be estimated. Moreover, it can not effectively compensate the interference of noise signals, but can circumvent the problem of initialization and convergence of the model parameters discussed in the existing iterative algorithm used for identification of Hammerstein model. Examples are used to illustrate the effectiveness of the proposed method.