Infrared Thermography Based Fault Diagnosis and Prognosis for Rotating Machines

Rotating machines are the crucial components of any industrial firm. The abrupt malfunctioning of such machines doesn’t merely affect the process downtime moreover it leads to production loss. Therefore to minimize the operation and maintenance costs, early detection of faults is very essential. As of late infrared thermography (IRT) has picked up consideration amongst the non-destructive condition monitoring techniques for early fault diagnosis of rotating machines. IRT is one of the contactless and non-invasive condition monitoring (CM) tool with very high accuracy and reliability. Real-time temperature measurement is done in a non-contact manner with this technique. IRT has discovered its applications in paper, aerospace, nuclear, wood, plastic, and various other industries. This paper presents a state of art review describing the fundamentals of IRT, diverse induction motor (IM) faults, and their diagnosis. The paper summarizes some of the machine learning methods such as artificial neural network, fuzzy logic, adaptive neuro-fuzzy inference system, and support vector machine for the detection of bearing faults. Additionally, some deep learning methods have been discussed in this paper due to their superiority over machine learning methods in terms of feature extraction as well as their selection. Also, the future scope has been proposed for developing a complete and self-learning package for fault diagnosis, and prognosis.