Abstract
The high vibration, temperature, and pressure issues cause the failure of the rotating electrical equipments. The failures become considerable when these equipments are used in industries and in smart grid. The more common failures are because of high vibrations, and sometimes, it may lead to complete shutdown of the system. The condition monitoring system must be reliable and detects the future fault conditions of the electrical equipment. The condition monitoring (CM) system is reliable and predictive when machine learning and data analytics are implemented. There are various machine learning techniques that help to detect the fault in minimum time using the historical data of the equipment and data analytics. It also helps to avoid the permanent failure of the electrical rotating equipment. Therefore, this paper focuses on health monitoring and remaining useful life (RUL) estimation of the electrical equipment connected to the grid using principal component analysis (PCA). PCA is an unsupervised machine learning technique that is proposed in this paper for case study of high-speed wind turbine bearing.
