expert control system

expert control system

Expert control system

A type of intelligent control system which can emulate the reasoning procedures of a human expert in order to generate the necessary control action. Expert control systems seek to incorporate knowledge about control system design, practical operations, and abnormal system recovery plans to automate tasks normally performed by experienced control engineers (the experts). Techniques relating to the field of artificial intelligence are usually used for the purpose of acquiring and representing knowledge and for generating control decisions through an appropriate reasoning mechanism. As it operates essentially on a knowledge base, an expert control system is often referred to as a knowledge-based control system. One of the most important benefits associated with the use of an expert control system is the inherent capability of the system to deal with uncertainty in information. Information provided to these systems can be general, qualitative, or vague because, like humans, expert control systems possess functionalities to perceive, reason, infer, and deduce new information. They can learn, gain new knowledge, and improve their performance through experience.

The principal components almost certain to be present in all expert control systems are the knowledge source, the database, the inference engine, the control algorithms, and the interface between the expert control system and humans (see illustration).

Central to an expert system is the knowledge source (or knowledge base), which contains knowledge in a specific domain (control in this context). This knowledge consists of domain-specific facts and heuristics, usually in the form of rules or frames, useful for solving problems in the domain.

The database is a short-term memory component which contains the current problem status, inference states, and the history of solutions to date for reference purposes.

The inference engine operates on the information from the knowledge source, from the associated database, or from the user; guides the search process according to the programmed strategy and search algorithms; and uses the inferencing mechanisms, usually in the forms of rules of logic, to solve problems, arrive at conclusions, and activate the final control actions.

The control algorithms are the tools for the expert systems to perform the final control action. A rich library of control algorithms, with advanced features, is usually made available.

The user interface allows the user to interact with the overall system, browse through the knowledge source, edit rules, and perform many other interactive tasks. See Artificial intelligence, Control systems, Expert systems, Process control