Adaptive prescribed performance estimator for Hammerstein-like system identification based on quantized observations and its application

Prescribed performance technology is a promising methodology that has received wide attention in control communities owing to its quantitative description for the steady-state and transient performance of control systems in recent years. Simplifying controller design, reducing system regulation time and preventing system divergence can all be achieved through the improved transient performance of parameter estimation. Unfortunately, in system identification communities, few papers on the transient performance of parameter identification are published because of difficulties in designing the error variable reflecting this performance. To resolve the above problem, this study provides an available solution by integrating the prescribed performance technology into the design of the estimator. We introduce an adaptive prescribed performance parameter identification of Hammerstein-like systems subject to the quantised observations. Firstly, a low-pass filter and forcing variables are developed to construct the transient performance error expression. An improved prescribed performance function that characterises the error bound of the parameter estimation is then introduced. Secondly, the identification error transformation is used to obtain a new system by transforming the raw system such that a constraint condition is avoided. A novel adaptive law is proposed to guarantee the original parameter identification with prescribed performance. Finally, simulation and process examples are given to state the finding results.