FUZZY LOGIC-BASED CONTROL OF A PH PROCESS

Abstract

Difficulties in pH control are predominantly due to the nature of the disturbances and nonlinearity present in the system. The changing process dynamics in pH neutralization process has made it one of the benchmark control problems. In this study, a Computer-Aided-Engineering (CAE) fuzzy control platform, known as FlexControl, is applied to design and test a real-time fuzzy logic controller for controlling pH. Integrating both simulation and real-time implementation capabilities, FlexControl allows the tedious and time-consuming design and testing of the controller to be carried out conveniently and efficiently. A 7x7 fuzzy rule table and bell-shaped membership functions were designed and implemented on-line to test the applicability and efficiency of both the platform and the controller. An experimental evaluation of the fuzzy control system has been carried out using a computer-controlled laboratory scale continuous stirred tank reactor (CSTR) for pH neutralization. Real-time experimental results are provided to demonstrate the robustness of the newly designed fuzzy controller in handling a variety of disturbances that occur in a pH process. The performance has been compared with that of a digital PI controller. As a model-free control algorithm, fuzzy controller is found to successfully imitate the actions of an operator and provide better control performance than the PI controller. In all cases studied, fuzzy control outperformed PI in terms of load rejection and set-point tracking capabilities.