Design decentralized sliding mode controllers with dynamic decouplers for TITO processes

Abstract

A sliding mode controller design method is proposed for multivariable systems. The method is based on dynamic decouplers to improve performance of multi loop control systems by compensating process interactions. A Proportional-Integral-Derivative (PID) sliding surface is used for sliding mode control. A process is decoupled through dynamic decouplers and each higher order element of decoupled subsystem is represented in the form of first-order-plus-dead-time (FOPDT) model using model reduction method. The sliding mode control law is derived from the Lyapunov stability approach and tuning parameters are computed for the sliding mode controller and PID sliding surface. Typical multivariable industrial processes including those have weak, moderate or strong interaction modes are employed to demonstrate the effectiveness and simplicity of the proposed method.