Multi-innovation recursive identification methods for nonlinear sandwich systems using the auxiliary model

In this paper, an auxiliary model-based multi-innovation least squares algorithm is used to solve the identification problem existing in the nonlinear sandwich system. Nonlinear sandwich systems are widely used in real industrial systems, and effective system identification is beneficial for designing effective controllers to control them. The nonlinear sandwich system consists of two linear modules and one nonlinear module in which there are unmeasurable internal variables. To solve the problem, an auxiliary model is constructed in this paper to replace the unmeasurable variables with their outputs. Then the multi-innovation theory is introduced to improve the accuracy of the algorithm and the convergence of the algorithm is proved. Finally, a numerical example and one physical simulation example are used to prove the effectiveness of the algorithm proposed in this paper.