Machine condition prognosis based on regression trees and one-step-ahead prediction

Predicting the degradation of working conditions of machinery and trending of fault propagation before they reach the alarm or failure threshold is extremely important in industry to fully utilize the machine production capacity. This paper proposes a method to predict the future conditions of machines based on one-step-ahead prediction of time-series forecasting techniques and regression trees. In this study, the embedding dimension is firstly estimated in order to determine the necessarily available observations for predicting the next value in the future. This value is subsequently utilized for the predictor which is generated by using regression tree technique. Real trending data of low methane compressor acquired from condition monitoring routine are employed for evaluating the proposed method. The results indicate that the proposed method offers a potential for machine condition prognosis.