Remote diagnosis and monitoring of complex industrial systems usinga genetic algorithm approach

摘要

Remote diagnosis of industrial and manufacturing facilitiesconstitutes a feasible alternative at high-risk or remote sites whereunmanned operation is preferred. Computer-aided diagnostic tools canreduce downtime by providing support to remote monitoring centers andon-site plant operators. This paper describes a novel technique toperform on-line remote monitoring and diagnosis of industrial andmanufacturing systems based on Bayesian belief networks and geneticalgorithms. An implementation of the methodology in a chemical processindustry is presented and potential applications for different types ofindustrial systems are discussed