Data Filtering-Based Maximum Likelihood Gradient-Based Iterative Algorithm for Input Nonlinear Box–Jenkins Systems with Saturation Nonlinearity

Abstract

Saturation nonlinearity exists widely in various practical control systems. Modeling and parameter estimation of systems with saturation nonlinearity are of great importance for analyzing their characteristics and controller design. This paper focuses on the identification issue of the input nonlinear Box–Jenkins systems with saturation nonlinearity. The input saturation nonlinearity is presented as a linear parametric expression through the application of a switching function, then the identification model of the system is derived by using the key term separation technique. Based on this model and the data filtering technique, the filtering identification model of the system is given by changing the system structure without changing the relationship between the input and output, which can reduce the interference of the colored noise and improve the identification accuracy. Then a data filtering-based maximum likelihood gradient-based iterative algorithm is proposed to estimate the unknown parameters. The maximum likelihood gradient-based iterative algorithm is provided for comparison. The feasibility and superiority of the proposed approach are emphasized by a simulation example.