SOURCE:

Faculty: Engineering
Department: Electronic And Computer Engineering

CONTRIBUTORS:

Nwobodo, H. N.
Inyiama, H.C.

ABSTRACT:

This dissertation involves the design of Intelligent Process Control Systems(IPCS) Using Fuzzy Neural Network. It presents work on the combination of advantages of Fuzzy Logic and Neural Network techniques in developing intelligent control systems for processes having unknown and non linear dynamics. Its main objective is to monitor, detect, diagnose and indicate faults in control systems using Fuzzy Neural logic principles. A Fuzzy Neural Network implementing a Proportional Integral Derivative (PID) control system was developed and simulated using Proteus. MATLAB was used to solve the PID equation while Visual Basic.net (VB.net) was used to generate the error data. This dissertation was first implemented using fuzzy logic, optimized using Neural Network to realize a Fuzzy Neural PID controller. It was ascertained that “Intelligent Process Control Systems using Fuzzy Neural Network” performed best with a rise time of 1s compared to 3s for the classical PID. Furthermore, the Fuzzy Neural process showed a smoother overall performance beyond the initial rise time of 1s.The approach to the realization of the much improved performance of the Fuzzy Neural model obviated the need for the rigorous mathematical methods involved in developing a transfer function for a linear PID control system.