PSO Based Multi-Objective Approach for Controlling PID Controller

Abstract

CSTR (Continuous stirred tank reactor) is employed in process control and chemical industries to improve response characteristics and system efficiency. It has a highly nonlinear characteristic that includes complexities in its control and design. Dynamic performance is compassionate to change in system parameters which need more effort for planning a significant controller for CSTR. The reactor temperature changes in either direction from the defined reference value. It is important to note that the intensity of chemical actions inside the CSTR is dependent on the various levels of temperature, and deviation from reference values may cause degradation of biomass quality. Design and implementation of an appropriate adaptive controller for such a nonlinear system are essential. In this paper, a conventional Proportional Integral Derivative (PID) controller is designed. The conventional techniques to deal with constraints suffer severe limitations like it has fixed controller parameters. Hence, A novel method is applied for computing the PID controller parameters using a swarm algorithm that overcomes the conventional controller's limitation. In the proposed technique, PID parameters are tuned by Particle Swarm Optimization (PSO). It is not easy to choose the suitable objective function to design a PID controller using PSO to get an optimal response. In this article, a multi-objective function is proposed for PSO based controller design of CSTR.

Review on PID, fuzzy and hybrid fuzzy PID controllers for controlling non-linear dynamic behaviour of chemical plants

“…Goud et al (Goud et al 2022) have introduced the tuned PID for controlling the nonlinear temperature in CSTR. In the suggested model, the temperature was controlled by the auto-tuned PID via a feedback loop.…”

Feedback Temperature Controller Design to a Semi-batch Hydrogenation Industrial Reactor - Part A: Modeling, Experimental Validation and Current Control System Evaluation

“…Additionally, when a process is run in a semi-batch mode, more nonlinearities are introduced, and the basic PID control typically fails to operate in a highly nonlinear environment (Goud et al 2022). PID controllers are still useful when a highly nonlinear process is run within a small range or when the nonlinearity level is not too high (Seaborg et al 2017).…”

“…Similarly, the volumetric flow of the refrigerant fluid through the control valve can be calculated by the Driskell equation(Garcia 2005), through Equation (32). 𝐹 𝐽 = 27339.6𝐶 𝑣,𝐽 .…”

Optimal Decentralized PI Control Design for a TimeDelay Benchmark Process Control System : AModified Whale Optimization Algorithm Approach

“…Optimization techniques work well for estimating parameters and optimising PID controllers, but control system engineers frequently face difficult optimization issues with traditional conventional numerical approaches. Normally, conventional optimization algorithms do not provide adequate efficient closed loop performances [20]. Recent research has used multiple evolutionary optimization algorithms since evolutionary optimization algorithm-based design does not require a strong mathematical model of the plants [21].…”