On-line industrial supervision and diagnosis, knowledge level description and experimental results

This paper presents a detailed description of a knowledge-based system for on-line supervision and diagnosis of industrial continuous processes: TURBO LID. The system has been developed for a Spanish beet sugar factory, and it is in use in two plants. The system supports three main tasks; monitoring, operation mode and diagnosis. A detailed knowledge level account of the sy stems is presented in this paper, using CommonKADS methodology. The knowled ge level description presented here will allow the reuse of the TURBOLID pr oblem solving approach to supervision and diagnosis in other continuous pla nts, even in other domains. The main purpose of this paper is to illustrate how the basic tasks proposed are able to cope with supervision and diagnos is of a complex plant. Particularly significant are the separation of opera tion mode and diagnosis, and the approach of TURBOLID to diagnosis, that ob tains on-line causal explanations for detected problems. TURBOLID is able t o differentiate among competing hypotheses looking for historical data, cur rent data and even future data, tracking the plant evolution. Results of th e activity of the system during a working month an also presented. (C) 2001 Elsevier Science Ltd. All rights reserved.