The low cost and high reliability of bearings make them the most widely used elements in the industry. Companies continually make bearing replacement decisions when it is estimated that it will fail, either by statistics or by manufacturer recommendations. However, this type of maintenance can represent an unnecessary expense when replacing the bearing prematurely and, therefore, constitutes a waste of its useful life. A "precision" maintenance program may be recommended to extend bearing life 5-10 times longer than traditional programs.

 Bearings are mechanical components that allow rotary machines to exert radial or axial movements and help reduce friction between the moving elements of a system. Under normal operating conditions, bearings have a very long service life, although they have a probability of 40%. Bearing failures can be associated with the operating environment, lack of precision in tolerances when machining shafts and housings, manufacturing defect, poor assembly, misalignment, irregular rolling, corrosion, indentation. inadequate lubrication, electrical damage, overheating, overload, fatigue, wear and fracture. These failures can generate losses due to machine downtime, unwanted vibration, noise and damage to other components, but if they are detected in time, the repair costs and downtime are minimal.

The time for bearing failures identification before interrupting operation or affecting product quality is the basis for most predictive maintenance programs. Taking readings, keeping history of failures and evaluating these results in the operation of rotating equipment on a regular basis, allows to identify possible failures before they become catastrophic. In this way, the damages or defects that are detected before a failure occurs, reduce the repair costs and the time that a rotating machine will be inactive.

This article shows in detail the different detection and classification techniques most used to identify faults in a bearing. There are many analysis techniques that allow us to detect and classify failures in industrial machinery, either offline or online. Take in to account  the literature review, different techniques are identified; which can be broadly classified as vibration analysis, ultrasound analysis, artificial neural networks (i.e ANN), convolutional neural networks (i.e CNN) and support vector machine (i.e SVM). In order that, vibration analysis is the most accepted technique due to its ease of application; However, currently artificial intelligence, through the analysis of neural networks, is being implemented in this field with greater frequency given its ease of programming and variety of applications, achieving a performance in the classification in the test stage of 75-100%.


Bearing Fault Detection and Classification: A Framework Approach

Division / Sponsors: 
Aerospace, Bioengineering, Applied Mechanics, Design Engineering Division, Advanced Energy Systems, Manufacturing Engineering Division, Noise Control and Acoustics, Nondestructive Evaluation, Diagnosis, and Prognosis (NDPD), Nuclear Engineering Division, Process Industries Division, Fluids Engineering, Heat Transfer, Materials, Microelectromechanical Systems, Safety Engineering and Risk Analysis, Computers & Information in Engineering


Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2020/volume/84478

IMECE2020 - Virtual Conference

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.