Nonparametric time series modelling for industrial prognostics and health management

Prognostics and health management (PHM) methods aim at detecting the degradation, diagnosing the faults and predicting the time at which a system or a component will no longer perform its desired function. PHM is based on access to a model of a system or a component using one or combination of physical or data driven models. In physical based models one has to gather a lot of knowledge about the desired system, and then build analytical model of the system function of the degradation mechanism that is used as a reference during system operation. On the other hand data-driven models are based on the exploitation of symptoms or indicators of degradations using statistical or Artificial Intelligence (AI) methods on the monitored system once it is operational and learn the normal behaviour. Trend extraction is one of the methods used to extract important information contained in the sensory signals, which can be used for data driven models. However, extraction of such information from collected data in a practical working environment is always a great challenge as sensory signals are usually multidimensional and obscured by noise. Also, the extracted trends should A. Mosallam FEMTO-ST Institute AS2M department 25000 Besançon, France. E-mail: [email protected] K. Medjaher FEMTO-ST Institute AS2M department 25000 Besançon, France. E-mail: [email protected] N. Zerhouni FEMTO-ST Institute AS2M department 25000 Besançon, France. E-mail: [email protected] 2 Ahmed Mosallam, Kamal Medjaher and Noureddine Zerhouni represent the nominal behaviour of the system as well as should represent the health status evolution. This paper presents a method for nonparametric trend modelling from multidimensional sensory data so as to use such trends in machinery health prognostics. The goal of this work is to develop a method that can extract features representing the nominal behaviour of the monitored component and from these features extract smooth trends to represent the critical component’s health evolution over the time. The proposed method starts by multidimensional feature extraction from machinery sensory signals. Then, unsupervised feature selection on the features domain is applied without making any assumptions concerning the number of the extracted features. The selected features can be used to represent the nominal behaviour of the system and hence detect any deviation. Then, empirical mode decomposition algorithm (EMD) is applied on the projected features with the purpose of following the evolution of data in a compact representation over time. Finally, ridge regression is applied to the extracted trend for modelling and can be used later for remaining useful life prediction. The method is demonstrated on accelerated degradation dataset of bearings acquired from PRONOSTIA experimental platform and another dataset downloaded form NASA repository where it is shown to be able to extract signal trends.