The development and evaluation of an intelligent supervisory system for process control.

[摘要] As industrial plants become more complex. there is a growing need for new approachesto control and supervision. This research investigates the issues involved inapplying Artificial Intelligence (AI) techniques in the real-world engineering problemof process control supervision. Current AI theory is examined and some techniquesmodified to design a general-purpose, reactive planner. The planner forms the basisof a supervisory control system. The system is implemented and interfaced withan existing Laboratory plant, so that its performance can he tested and evaluatedby comparing with a conventional feedback controller This real life testing necessitatesexplicit treatment of issues such as data: sampling. situation assessment andCPU scheduling. The case study shows that by combining AI techniques with conventionalcontrol, a system can be built which displays superior performance undernormal operating conditions and which can deal with abnormal conditions such asequipment failures.

[发布机构] University of the Witwatersrand