Intelligent Supervisory Controller for Gas Distribution System

This paper describes the application of artificial intelligence techniques in the supervisory controller of a gas distribution network. The main idea was to use symbolic models of the distribution network in the knowledge-based system components. The symbolic models represented various quantitative and qualitative features of the process which made it possible to support the monitoring, diagnostic and intervention planning procedures with high-level tools. The spectrum of Al techniques used in the system architecture included associative databases, object-oriented programming and planning techniques. The system has been implemented in a distributed computing environment.