Model-based reasoning for fault diagnosis

Conference · Thu Jan 01 00:00:00 EST 1987

OSTI ID:5929210

Venkatasubramanian, V; Rich, S H

Recent developments in Artificial Intelligence (AI) have resulted in newer approaches using knowledge-based expert systems to problems in the design of automated process fault diagnostic systems. Despite the advantages offered by these first-generation systems over conventional methods such as fault tree analysis and signed digraphs there are some serious drawbacks. Owing to their complete reliance on heuristic or experiential knowledge, the first-generation systems are not flexible to accommodate even small changes in process configuration and are incapable of diagnosing unanticipated fault combinations. In this paper, the authors discuss a methodology that aids the development of expert systems which are process-independent, transparent in their reasoning, and capable of diagnosing a wide diversity of faults. A prototype expert system, called MODEX, has been implemented incorporating these ideas. The domain knowledge of the system is based on qualitative reasoning principles and captures physical interconnections between equipment units as well as casual relationships among process state variables. The inference strategy uses model-based reasoning for analyzing the plant behavior. Using a variant of the technique adopted from fault tree synthesis, an initially observed abnormal symptom is considered to be a top level event and a tree structure is constructed as the system searches for a basic event to which the fault can be traced. The diagnostic reasoning process is driven by a problem reduction strategy. The knowledge base is process-independent, thereby enhancing the generality of the expert system. Reasoning from first-principles with the aid of causal and fault models facilitates the diagnoses of novel or unanticipated faults.

• OSTI ID: 5929210
• Report Number(s): CONF-870323-
• Resource Relation: Conference: American Institute of Chemical Engineers spring national meeting, Houston, TX, USA, 29 Mar 1987; Other Information: Technical Paper 82C
• Country of Publication: United States
• Language: English

Related Subjects

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
COMPUTERIZED CONTROL SYSTEMS
EXPERT SYSTEMS
INDUSTRIAL PLANTS
FAILURE MODE ANALYSIS
ARTIFICIAL INTELLIGENCE
KNOWLEDGE BASE
MONITORING
CONTROL SYSTEMS
SYSTEM FAILURE ANALYSIS
SYSTEMS ANALYSIS
320303* - Energy Conservation
Consumption
& Utilization- Industrial & Agricultural Processes- Equipment & Processes
990210 - Supercomputers- (1987-1989)