A Nonlinear PID Control Design for Block-Oriented Nonlinear Models

摘要

Block-oriented nonlinear structures are popular in modelling nonlinear electrical, biological, and chemical processes. These models are characterized by a static nonlinearity component, which is in general described by polynomial form, in series with a linear dynamic submodel. The unknown parameters are estimated from real-time experiments by using the least squares identification algorithm. The conventional control strategy for this class of systems is the nonlinearity inversion method, which is including an exact inverse of the static nonlinearity into the appropriate place. Hence, any standard linear controller can be applied. In contrast, the performance of this classical control strategy becomes limited when the considered static nonlinearity is not exactly invertible. To overcome this difficulty, a novel control approach is proposed in this paper. An electrical process is considered to illustrate the efficacy of the proposed control strategy.