EE461 -Project

Figure: Visualization of 13-time domain features. This figure enabled us to identify the similar visualization graph which will be used to set thresholds and diagnose each stage by marking the date and time whereby this was compared amongst each features which showed similar date and time of change in stage. The features which have similar visual are Root Mean Square, Square Root Mean, Kurtosis, Variance and Standard Deviation.

 

The figure below  shows a concise depiction of both the process flow and the proposed system, serving the aim of concluding.

 Rotating gear is a major component of the industrial world, yet unplanned malfunctions can result in costly downtime and safety risks. By analysing vibration data from machines, deep learning algorithms can identify minute patterns that point to wear and tear, enabling us to anticipate and stop breakdowns before they happen.  The focus of this project will be to develop and execute an intelligent system for diagnosis and prognosis. The system will have the capability to detect the degradation of RMs components, predict the RUL, and subsequently classify the nature of the defect that will happen. The system will improve its diagnostics and prognostics system by using DL techniques for analysing vibration data generated by the bearings. The accomplishment of this objective will be facilitated by the development of the system. After detecting the deterioration and classifying the issue, the system will proceed to calculate the RUL. Preventative maintenance can be implemented thr...

 Presented by Vishal Kumar - S11184965 and Khetan Lal - S11185268