Predictive Maintenance of High-Velocity Oxy-Fuel Machine Using Convolution Neural Network

Abstract

Maintenance activities and programs have reached a critical factor for competitive advantage in the manufacturing world. This is due to the increasing complexity of the interactions between different production activities and increasing extended manufacturing environment along with the increasing cost of maintenance. Existing practices namely preventive maintenance (PvM) which scheduled maintenance is no longer feasible. With the advancement in the machine and artificial intelligence (AI) technology, predictive maintenance (PdM) is the future. In the predictive maintenance, Prognostic and Health Management (PHM) emerged as a technique that is widely used in order to anticipate machine breakdown through Remaining Useful Life (RUL) determination. The aim of this study is to apply machine learning technique in order to predict the Remaining Useful Life (RUL) of the High-Velocity Oxyfuel machine (HVOF) machine. As proposed by many researchers, convolutional neural network (CNN), which primarily used in image processing has been adopted to predict the RUL due to the characteristics and its ability to achieve some level of acceptable functionality and precision using few parameters as input. High-Velocity Oxyfuel machine’s data are input to the convolutional neural network (CNN) algorithm to predict breakdown as an output. Historical data as in maintenance reporting system (MRS) from the existing preventive maintenance (PvM) program is also used to understand the time-since-breakdown of the machine. The expected outcome of this study is to study the feasibility to predict HVOF machine breakdown using CNN.