Fault Detection Using Vibration Analysis and Particle Swarm Optimization of the Rolling Element Bearing

The bearing's relevance, technical uses are clear in many applications. It is subjected to various types of loading. The rolling bearing may be cracked because of fatigue loading. The presence of a crack causes a change in the physical properties of a bearing and thus reducing the stiffness of the rolling bearing, where the invisible natural frequencies are being reduced. The essential signatures of vibration of bearing analysis are crack depth and location. The current study used Finite Element Analysis (FEA), experiments data and Particle Swarm Optimization (PSO) technology to create methodologies for fracture detection of a solitary crack in a rolling bearing. Different crack location effects are taken into account, and the results are compared to different rolling bearing crack depths. Then PSO algorithm has been developed using the first three relative natural frequencies taken from FE analysis and experiments data. For comparative study, both Standard PSO and APSO are used for crack diagnosis of the bearing. The feasibility of proposed PSO techniques is compared through error analysis. In the research paper, the objective has been related to the design of a Particle swarm optimization technique for more accuracy and less time consumption to the prediction of crack location and crack depth in cracked bearing.