A real-time expert system shell for process control.

• Main Author: Kang, Alan Montzy
• Format: Others
• Language: en
• Published: 2018
• Subjects: Expert systems (Computer science); Process control > Computer programs.; Real-time data processing.; Real-time control.
• Online Access: https://hdl.handle.net/10539/25920

A dissertation submitted to the Faculty of Engineering, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Master of Science in Engineering === A multi-layered expert system shell that specifically addresses real-time issues is designed and implemented. The architecture of this expert system shell supports the concepts of parallelism, concurrent computation and competitive reasoning in that it allows several alternatives to be explored simultaneously. An inference engine driven by a hybrid of forward and backward chanining methods is used to achieve real-time response, and certainty factors are used for uncertainty management. Real-time responsiveness is improved by allowing the coexistence of procedural and declarative knowledge within the same system. A test bed that was set up in order to investigate the performance of the implemented shell is described. It was found in the performance analysis that the proposed system meets the real-time requirements as specified in this research. === Andrew Chakane 2018