Computational intelligence based machine fault diagnosis

Machine fault diagnosis is a well established area where specific techniques are used to determine fault patterns or locations. In recent years, there are many studies about this issue by means of model based approach, probabilistic method, knowledge based approach and neural networks based approach et al. With the progress of the study of biology, evolutionary thought has extended into engineering problem-solving. More interests have been shown in this field. The investigation will describe two unsupervised clustering paradigms, Kohonen's self-organizing scheme and genetic algorithm (GA) based heuristic searching, for machine fault classification. In case study, a multiple faults classification problem has been attacked. Solutions generated from the GA based system are compared with that from self-organization neural networks, and the result is given, and the case study has shown that the proposed approaches are flexible enough to be used in practical fault diagnosis.