A review on hybrid methods of fault diagnosis, necessity and challenges

  To increase industrial machines reliability and to prevent shut downs due to faults which cause production loss, fault detection and isolation is very important. There is an increasing request for finding a solution by researchers from industry. Different methods can be used as a solution. Some of them are based on system’s model while some others use experimental data collected from the system. In recent studies hybrid methods are suggested to compensate limitations of individual methods. This approach is briefly reviewed in this article.