<![CDATA[Bearings have a very important role in an industry. However, the cost of maintenance and replacement of bearings are very expensive especially for slew-bearing which operated in a very low speed. If the low-speed slew bearing shutdown suddenly, it will also cause a financial issue to the certain industries with rely on the rotating machines because the entire machine will be shut down and the production will be stop Therefore, monitoring of the low-speed slew bearing condition at all times is necessary to predict the bearing failure. There has been advance monitoring devices and systems related to the vibration condition monitoring for bearing and rotating machines, however, in certain cases those monitoring devices and systems are not sufficient. Machine learning is offered to complement and contribute in this case which aims to determine the prediction and Remaining Useful Life (RUL) of the bearing before the bearing experiences more damage. In this paper, the Random Vector Functional Link (RVFL) is used to predict RUL using low speed slew bearing data from University of Wollongong, Australia. The main evaluation matrix such as RMSE is used as an evaluation of the performance of the model used. According to the prediction results, the best modeling results are obtained using a data ratio of 80:20 and a SELU activation function that produces the best average RMSE value. The prediction value of Remaining Useful Life (RUL) of the bearing is 94.24%.]]>
