Application of Artificial Neural Networks to the Optimum Control of pH in Chemical Process

emuoyibofarhe, o. j.
Department of Computer Science and Engineering, Ladoke Akintola University of Technology, Ogbomoso, Nigeria.
oladipo, babajide
Department of Computer Science and Engineering, Ladoke Akintola University of Technology, Ogbomoso, Nigeria.
ogunleye, o. o.
Department of Chemical Engineering, Ladoke Akintola University of Technology, Ogbomoso, Nigeria.
Abstract
In chemical process, the influence of external disturbance contributes to the instability of the process, thus leading to unsafe plant operation and hence the need for control. Optimizing the control of pH in chemical process is not only of economic importance, but also assures product safety and product quality. In this paper we use an adaptive neuro-fuzzy based system to simulate the control and monitoring of pH in a Continuous Stirred Tank Reactor (CSTR), the result obtained shows that this methodology provides an alternative solution. The control of pH is an automatic process and hence requires automatic and very sensitive monitoring methods. Artificial neural networks due to their ability to learn from experience are best suited for monitoring such processes. We achieve a 96% monitoring accuracy.

Keywords: Continuous Stirred Tank Reactor (CSTR), pH-value, Adaptive-based – based Fuzzy Inference System (ANFIS)

Comments are closed.