Domain Adaptation via Alignment of Operation Profile for Remaining Useful Lifetime Prediction

Abstract

Effective Prognostics and Health Management (PHM) relies on accurate prediction of the Remaining Useful Life (RUL). Data-driven RUL prediction techniques rely heavily on the representativeness of the available time-to-failure trajectories. Therefore, these methods may not perform well when applied to data from new units of a fleet that follow different operating conditions than those they were trained on. This is also known as domain shifts. Domain adaptation (DA) methods aim to address the domain shift problem by extracting domain invariant features. However, DA methods do not distinguish between the different phases of operation, such as steady states or transient phases. This can result in misalignment due to underor over-representation of different operation phases. This paper proposes two novel DA approaches for RUL prediction based on an adversarial domain adaptation framework that considers the different phases of the operation profiles separately. The proposed methodologies align the marginal distributions of each phase of the operation profile in the source domain with its counterpart in the target domain. The effectiveness of the proposed methods is evaluated using the New Commercial Modular Aero-Propulsion System (N-CMAPSS) dataset, where sub-fleets of turbofan engines operating in one of the three different flight classes (short, medium, and long) are treated as separate domains. The experimental results show that the proposed methods improve the accuracy of RUL predictions compared to current state-of-the-art DA methods.

DARE-GRAM : Unsupervised Domain Adaptation Regression by Aligning Inverse Gram Matrices

“…This framework encompasses popular approaches, including domain-invariant representation learning, reweighting, and adaptive Gaussian processes. Recent endeavors have focused on tackling the unique challenges within the unsupervised setting for domain adaptation for regression [6,27,28]. For instance, RSD [6] addresses the challenge of feature scale changes during domain adaptation by aligning representation subspaces using a geometrical distance measure.…”

Unsupervised Physics-Informed Health Indicator Estimation for Complex Systems

“…The accuracy and reliability of prognostic prediction techniques depend highly on the quality and representativeness of the available time-to-failure data. Therefore, these methods may not perform well when applied to data from new units that operate under different conditions than those used during training (Nejjar, Geissmann, Zhao, Taal, & Fink, 2023).…”

“…The N-CMAPSS dataset consists of engines classified into three distinct flight classes, and it is reasonable to expect that the degradation patterns between engines of different flight classes may differ significantly. Furthermore, the sensor measurements of units from different flight classes may also be affected as the units reach different altitudes and speeds (Nejjar et al, 2023).…”