Challenging design problems arise regularly in modern fault diagnosis systems. Unfortunately, the classical analyticaltechniques often cannot provide acceptable solutions to such difficult tasks. This explains why soft computing techniquessuch as evolutionary algorithms and neural networks become more and more popular in industrial applications of faultdiagnosis. The main objective of this paper is to present recent developments regarding the application of evolutionaryalgorithms and neural networks to fault diagnosis. ; In particular, a brief introduction to these computational intelligenceparadigms is presented, and then a review of their fault detection and isolation applications is performed. Close attention ispaid to techniques that integrate the classical and soft computing methods. A selected group of them is carefully describedin the paper. The performance of the presented approaches is illustrated with the use of the DAMADICS fault detectionbenchmark that deals with a valve actuator.
|Advances in model-based fault diagnosis with evolutionary algorithms and neural networks||2018-08-10|