Scalable multiple linear model RTDA controller formulation for constrained nonlinear processes

Abstract

Robustness, tracking, disturbance rejection and overall aggressiveness (RTDA) controller is a next generation control strategy with good robustness, tracking, and regulatory performance. This article proposes a scalable multiple linear model based RTDA (M‐RTDA) control scheme to handle nonlinear processes with or without constraints effectively. Multiple linear models at different operating regions are used to represent the nonlinear dynamics of the system. The optimal local linear models are identified using the gap metric based technique. The linear model and the corresponding RTDA controller at each operating region are combined to get a model‐controller unit. The output of each model controller unit is integrated using a fuzzy scheduling technique to ensure bumpless transfer. The effectiveness of the proposed technique is demonstrated through simulation using nonlinear benchmark processes like pH neutralization and continuous stirred tank reactor (CSTR) process with constraints. The block diagram representation of the scalable M‐RTDA controller is developed, and the step‐by‐step procedure (algorithm) to implement the proposed concept is discussed. The proposed scalable multiple linear model scheme is easily reconfigurable as per the performance/operator requirement without affecting the other local RTDA controllers. The qualitative and quantitative performances are compared with the results of the direct extended multiple model based RTDA (DE‐RTDA) controller and dynamic matrix controller(DMC).