Multiloop FOPID Controller Design for TITO Process Using Evolutionary Algorithm

Multiloop FOPID Controller Design for TITO Process Using Evolutionary Algorithm

Lakshmanaprabu S.K. (B S Abdur Rahman Crescent Institute of Science and Technology, Chennai, India), Najumnissa Jamal D. (B S Abdur Rahman Crescent Institute of Science and Technology, Chennai, India) and Sabura Banu U. (B S Abdur Rahman Crescent Institute of Science and Technology, Chennai, India)
Copyright: © 2019 |Pages: 14
DOI: 10.4018/IJEOE.2019070107


In this article, the tuning of multiloop Fractional Order PID (FOPID) controller is designed for Two Input Two Output (TITO) processes using an evolutionary algorithm such as the Genetic algorithm (GA), the Cuckoo Search algorithm (CS) and the Bat Algorithm (BA). The control parameters of FOPID are obtained using GA, CS, and BA for minimizing the integral error criteria. The main objective of this article is to compare the performance of the GA, CS, and BA for the multiloop FOPID controller problem. The integer order internal model control based PID (IMC-PID) controller is designed using the GA and the performance of the IMC-PID controller is compared with the FOPID controller scheme. The simulation results confirm that BA offers optimal controller parameter with a minimum value of IAE, ISE, ITAE with faster settling time.
Article Preview

1. Introduction

The Controlling the Multi Input Multi Output (MIMO) interconnected process is the most challenging problems for control system engineers. Several authors have been proposed control designing procedures for multivariable systems such as decentralized (multiloop) control, decoupled control, and centralized control. The design of controller in one loop depends on other loops because its interaction effect between inputs and outputs. Hence, the tuning of multiloop controllers is difficult. There has been many designing procedures proposed in the literature such as detuning (Luyben & Jutan, 1986; Chandra & Manickam, 2016), sequential loop closing (Mayne, 1973; Hovd & Skogestad, 1994), independent loop (Lee et al., 2001; Skogestad & Morari, 1989; Vu & Lee, 2010) and relay auto tuning method (Loh et al., 1993; Tan & Ferdous, 2003). In the detuning method, the controller parameter is found for most important loop transfer function without considering the interaction effect and then the controller gains are detuned by considering the interaction effect to meet some user control specification. But the performance and stability of closed loop system have not been discussed properly in the detuning procedures. In the sequential tuning methods, controllers are tuned while closing the loop one after another, but the final controller design completely depends on the order of other controller, which may increase the order of final controller element. The independent loop design methods are most widely used method for multivariable control design where the coupled process is decoupled using decoupler and the controllers are tuned using the decoupled process.

The integer order PID is still widely used in the industrial process because of its simplicity and reliability. There have been many tuning methods available in the literature for PI/PID tuning. Recently, the fractional order controller has gained popularity in the control society because of its improved performance and less sensitive to process uncertainty (Monje et al., 2010; Chen et al., 2009; Monje et al., 2008). The fractional order PID controller provides more flexibility in control tuning than PID controller tuning because of its additional tuning parameter which is order of integrator and order of differentiator. At the same time, the tuning of FOPID controller parameters is difficult because of the additional tuning parameters. The popularity of PID controller is because of the availability of extensive rules and an auto tuning feature. There have been lots of researchers presented the tuning method of FOPID using evolutionary algorithm with time domain and frequency domain fitness function. The FOPID controller tuning procedure based on integral absolute error and maximum sensitivity function is demonstrated in (Fabrizio & Visioli, 2011).

The few advantages of algorithms such as GA, PSO and Harmony search (HS) are incorporated in the BA (She, 2010). The comparative results of various optimization techniques are demonstrated and concluded that BA is powerful than GA, PSO and HA. In (Vishal et al., 2014), PI controller designed for DC motor application using various optimization techniques such as GA, Modified particle swarm optimization (MPSO), Differential evolution (DE) and cuckoo search algorithm (CS). The authors claimed that CS based PI controller provides better controller performance than the other optimization method. In Sivakumar et al., (2016), multiloop PI controller tuned for two input two output (TITO) process using BA, Firefly Algorithm (FA), Particle Swarm Optimization (PSO) and Bacterial Forging optimization (BFO). The comparative results suggested that the BA based controller provides better control performance than FA, PSO, and BFO. In Rajabioun and Ramin (2011), GA, PSO, CS algorithm utilized to design a centralized PI/PID controller for benchmark distillation. The comparison of GA based control with a CS based controller is presented. The CS algorithm is found to be a good algorithm with faster convergence and global optimal solution. The BA for designing the constrained engineering problem is demonstrated in Gandomi et al., (2013), where the BA results are compared extensively with PSO and Harmony search algorithm. In Liu and Xiaoyong (2015), FOPID controller designed using an adaptive PSO algorithm. The improvement of adaptive PSO based FOPID controller is shown by comparing with DE and PSO based FOPID controller.

Complete Article List

Search this Journal:
Open Access Articles
Volume 9: 4 Issues (2020): Forthcoming, Available for Pre-Order
Volume 8: 4 Issues (2019): 3 Released, 1 Forthcoming
Volume 7: 4 Issues (2018)
Volume 6: 4 Issues (2017)
Volume 5: 4 Issues (2016)
Volume 4: 4 Issues (2015)
Volume 3: 4 Issues (2014)
Volume 2: 4 Issues (2013)
Volume 1: 4 Issues (2012)
View Complete Journal Contents Listing