Design of a control system for computer aided SISO and MIMO intelligent control

The design of control systems for chemical engineering manufacturing processes is of paramount importance, since such parameters as temperature, concentration, pressure and flow often have to be constrained within certain bounds as a means of attaining preset process optimization criteria. In order to achieve the above we often seek the aid of controllers, which may be conventional PID, adaptive, or intelligent in their design, and these would then work as part of a complete control system comprised of various components of control hardware and software. This work deals with the design of such a complete control system. The system designed in this work is a computer-based control system which has been successfully implemented for the control of the feed flow rate, feed pre-heater temperature, reactor inlet temperature, and hot-spot temperature and position in a contact process pilot plant through the use of the following neural network based controllers: neural network self-tuning PID (NNW PID), and neural network generalized predictive control (NNWGPC) philosophies. An IBM computer, employed as the central processing unit (CPU), runs a single main module Microsoft Quickbasic(c) command program for the simultaneous intelligent control of all the process parameters. In the contact process, sulfur dioxide is oxidized on vanadium pentoxide catalyst to produce sulfur trioxide which is a raw material for sulfuric acid production