Prognosis of Machine Health Condition Using ARMA-GARCH Model

Abstract

Appropriate strategy of machinery maintenance is one of significant factors to enable companies’ owners to improve their product quality and to reduce manufacturing cost. Among of maintenance strategies: corrective maintenance, predetermined maintenance. Condition-based maintenance (CBM) is considered to be superior to the others in being more efficient which can determine the condition of machinery as it operates, to predict and schedule the most efficient repair of problem components prior to failure, while others approaches show to be costly due to unplanned downtime as in corrective maintenance or unnecessary overhauls as in predetermined maintenance. Prognosis is one of the most important modules in a CBM system. In this study, a novel approach of machine health condition prognosis is investigated to enhance the effectiveness of maintenance strategy. Vibration is considered to be the best operating parameter to judge dynamic conditions such as imbalance (overall vibration), bearing defects (enveloping) and stress applied to components. However, a real severity of vibration may not be correctly recognized due to echanical noises as well as electrical noises from the measuring equipments. This study presents a novel application of autoregressive moving average (ARMA) and generalized autoregressive conditional heteroscedasticity (GARCH) to evaluate and predict the actual severity of vibration collecting from machines to aid in making more accurate conclusions about their health condition. In this work, ARMA and GARCH models are respectively utilized to specify conditional mean and conditional variance of vibration data. These values also respectively explain for machinery behaviors about steady development of wear and fluctuant progress of faults. The mutual combination of ARMA and GARCH models will be able to give an accurate or real severity of machinery vibration. The forecasts of combined model, so-called ARMA-GARCH model, play an important role in making decisions on machine repair or possible improvements in order to reach maximum its run-ability, before any unplanned breakdown. The promises of the proposed model are verified in empirical results as applying for a real system of a methane compressor in a petrochemical plant.