Design and Analysis of an Intelligent PID Controller for Sugar Industry Process

Abstract

The proposed paper focuses on the identification of system model with different investigation on real time data obtained from industrial process using neural networks along with design of PID controller using intelligent controllers. The real time process in sugar industry has been taken as reference for modeling and controller design for controlling the process parameters such as level, temperature, flow, pressure, pH etc that associate with the process. The real time process constitutes multi input and output with external disturbance which makes the conventional mathematical modeling is complex when compared to algorithm based design. The process model representation in terms of mathematical model, transfer function, state space model holds the process characteristics and order of the system reflects the input and output to the system. The initial process in sugar industry is juice extraction from sugar cane called crushing unit which requires continuous monitoring and control for maximum juice extraction. The author clearly explains the model identification for cane crushing process with the aid of neural network for data fitness, estimation and validation, model identification. The classical PID controller plays the vital role in process control industries which controls the major closed loop process. The influence of external disturbance during run time in any process or change in set point leads to non-linearity. In order to improve the control action, intelligent controller like fuzzy logic controller and genetic algorithm has been designed for tuning of PID controller. The response of proposed controllers are compared with classical controller and discussed briefly in conclusion.