Multi-loop PI/PID controllers design for disturbance rejection based on non-parametric effective model and non-convex optimisation

In this study, the authors address the design of the multi-loop controllers for multi-input and multi-output (MIMO) processes for disturbance rejection. Three types of disturbance rejection are analysed, and they demonstrate that those disturbance responses are directly related to the integral terms of the controllers. A concept of non-parametric effective model is introduced here to decompose a MIMO control system into several equivalent single-input and single-output control systems. Based on the effective model in each loop, a non-convex optimisation problem is established to obtain the optimal controller parameters with certain robustness indices. Considering the drawbacks of using random search algorithm, they develop a numerical algorithm to solve this non-convex optimisation problem. Simulation studies demonstrate that the proposed method improves the disturbance rejection compared with other decentralised controllers.

[1]  Fernando Morilla,et al.  Centralized multivariable control by simplified decoupling , 2012 .

[2]  Qing-Guo Wang,et al.  Auto-tuning of multivariable PID controllers from decentralized relay feedback , 1997, Autom..

[3]  Sigurd Skogestad,et al.  Sequential design of decentralized controllers , 1994, Autom..

[4]  Tore Hägglund,et al.  Design of PID controllers based on constrained optimization , 1999 .

[5]  Donghai Li,et al.  OPTIMIZATION OF DECENTRALIZED PI/PID CONTROLLERS BASED ON GENETIC ALGORITHM , 2008 .

[6]  Wenjian Cai,et al.  Simple Decentralized PID Controller Design Method Based on Dynamic Relative Interaction Analysis , 2005 .

[7]  Lihua Xie,et al.  RNGA based control system configuration for multivariable processes , 2009 .

[8]  André Desbiens,et al.  Simplified, ideal or inverted decoupling? , 1998 .

[9]  Nusret Tan,et al.  Computation of stabilizing PI and PID controllers for processes with time delay. , 2005, ISA transactions.

[10]  Fernando Morilla,et al.  An extended approach of inverted decoupling , 2011 .

[11]  Jietae Lee,et al.  One-Parameter Method for a Multiloop Control System Design , 1999 .

[12]  Chyi Hwang,et al.  Solution of a non-convex optimization arising in PI/PID control design , 2002, Autom..

[13]  Wenjian Cai,et al.  Effective transfer function method for decentralized control system design of multi-input multi-output processes , 2006 .

[14]  Moonyong Lee,et al.  Independent design of multi-loop PI/PID controllers for interacting multivariable processes , 2010 .

[15]  K. Åström,et al.  Revisiting the Ziegler-Nichols step response method for PID control , 2004 .

[16]  Chih-Hung Chiang,et al.  A direct method for multi-loop PI/PID controller design , 2003 .

[17]  Moonyong Lee,et al.  Analytical Design of Multiloop PID Controllers for Desired Closed-Loop Responses (R&D Note) , 2004 .

[18]  William L. Luyben,et al.  Simple method for tuning SISO controllers in multivariable systems , 1986 .

[19]  K.J. ÅSTRÖM,et al.  Design of PI Controllers based on Non-Convex Optimization , 1998, Autom..

[20]  W. Cai,et al.  Equivalent transfer function method for PI/PID controller design of MIMO processes , 2007 .

[21]  Wonhui Cho,et al.  Multiloop PI controller tuning for interacting multivariable processes , 1998 .