A Systematic Guide for Predicting Remaining Useful Life with Machine Learning

Abstract

Prognosis and health management (PHM) are mandatory tasks for real-time monitoring of damage propagation and aging of operating systems during working conditions. More definitely, PHM simplifies conditional maintenance planning by assessing the actual state of health (SoH) through the level of aging indicators. In fact, an accurate estimate of SoH helps determine remaining useful life (RUL), which is the period between the present and the end of a system’s useful life. Traditional residue-based modeling approaches that rely on the interpretation of appropriate physical laws to simulate operating behaviors fail as the complexity of systems increases. Therefore, machine learning (ML) becomes an unquestionable alternative that employs the behavior of historical data to mimic a large number of SoHs under varying working conditions. In this context, the objective of this paper is twofold. First, to provide an overview of recent developments of RUL prediction while reviewing recent ML tools used for RUL prediction in different critical systems. Second, and more importantly, to ensure that the RUL prediction process from data acquisition to model building and evaluation is straightforward. This paper also provides step-by-step guidelines to help determine the appropriate solution for any specific type of driven data. This guide is followed by a classification of different types of ML tools to cover all the discussed cases. Ultimately, this review-based study uses these guidelines to determine learning model limitations, reconstruction challenges, and future prospects.

Exposing Deep Representations to a Recurrent Expansion with Multiple Repeats for Fuel Cells Time Series Prognosis

“…In this context, components material ( Figure 1 ), gas contamination, and the operating conditions imposed by the load profile are responsible for the system durability. Accordingly, a well-structured prognosis policy is needed to extend the lifetime by incorporating the required CBM tasks [ 5 ].…”

“…Among data-based approaches, DL tools have been widely investigated. According to [ 5 ], the reason for choosing DL tools is motivated by the need to provide a meaningful representation of data patterns originally presented in a very complex feature space (see Figure 2 from [ 5 ]). In the context of PEMFCs prognosis, the entire deterioration path can be obtained under real experimental conditions.…”

“…Additionally, it is common for PEMFCs to be subject to a higher level of dynamic disturbances caused by changing operating conditions leading to more “ complex ” and “ drifted ” data. If we project these three characteristics (i.e., data is complete, complex, and drifted) onto the flowchart proposed in the systematic guide of [ 5 ] (see Figure 2 of [ 5 ]), there would be no more optimal solution than DL for such a case, and conventional machine learning will no longer be recommended. Thus, reconstructing an RUL predictive model requires meeting three main criteria: data availability , complexity , and drift .…”

Synthesizing Rolling Bearing Fault Samples in New Conditions: A Framework Based on a Modified CGAN

“…Variational autoencoders (VAE) and generative adversarial networks (GANs) are the most famous generative models and have been wildly used for bearing fault detection [ 11 , 12 ], data augmentation [ 13 ], and predicting remaining useful life. By using generative algorithms, the problem of lack of samples and patterns in industrial data can be solved [ 14 ]. CGAN is a variation of GAN, which can generate conditional new data [ 15 ].…”

A remaining useful life prediction method based on PSR-former

“…With the increase of equipment service time, various wear and defects will gradually form on internal parts, resulting in equipment performance degradation, thus affecting the service life of the equipment. Bearing, as the core component of most rotating machines, once fails, the equipment will be affected or even collapsed, making it difficult to maintain the prognostic and health management (PHM) 1 of the equipment which is a task for real-time equipment operation monitoring. This information including the equipment operation status will be reflected in the vibration data through abnormal fluctuations.…”