Executive Summary : | Advanced monitoring systems are crucial for assessing the overall performance of plants, either by accurately calculating the end of effective life cycle of components or by predicting and locating defects in in-situ machine components in operation conditions. This requires a combination of quantifying different parameters, modeling the entire set-up, and suitable statistical techniques. The fundamental problem involves determining the physical process within a machine, which cannot be measured directly and must be inferred from external measurements, such as vibration signals. Fault diagnosis and detection of mechanical equipment, along with continuous monitoring, require analyzing and extracting important parameters to calibrate the state of prime components of machineries and industries. Bearings are critical components of induction motors and often experience defects during operation. Online condition monitoring is essential for the reliability of high-speed composite bearings systems, allowing for continuous data recording and appropriate maintenance activities planning. Combining signals from various advanced sensors can recognize current conditions and predict future condition trends. The presence of defects in bearings introduces local flexibility that affects its dynamic response. Theoretical analysis can be used to calculate strain energy release rate and compliance matrix, which are then used to calculate modal parameters of healthy and defective components of bearing parts. Experimental investigations on high-speed bearing systems can measure vibration responses in both healthy and defect conditions, verifying the authenticity of the developed theoretical model. Research is directed towards developing advanced computational technology for condition monitoring using artificial intelligence (AI) techniques of bearings, which could significantly impact engineering applications. |