An Intelligent Neural Network Controller for Non-Linear CSTR Process Control

The non-linear process control is the most important problem of statement in chemical industries. Stirred tanks are frequently used as industrial reactors, where a chemical component of a flow stream resides in the tank for a period of time before proceeding to other steps in a chemical process. In this research article, a computational intelligence-based controller is introduced for CSTR concentration control model. The proposed RBFNN model is optimally tuned by two-hybrid optimization strategies based on combining the best features of Particle Swarm Optimization (PSO), Gravitational Search Algorithm (GSA), and their variants such as Deterministic Particle Swarm Optimization Algorithm, and Differential Gravitational Search algorithm (DPSO-DGSA). An experiment is conducted to examine the effectiveness of the proposed controller methodology to compare it to other state-of-the-art approaches.