On Adversarial Vulnerability of PHM algorithms: An Initial Study
This addresses a security gap in PHM systems, which are critical for industrial and safety applications, by highlighting their vulnerability to adversarial threats, representing an initial but incremental study in this domain.
The paper investigates the vulnerability of Prognostics and Health Management (PHM) algorithms to adversarial attacks, focusing on time-series sensor data, and demonstrates through two real-world applications that these algorithms are indeed susceptible to such attacks.
With proliferation of deep learning (DL) applications in diverse domains, vulnerability of DL models to adversarial attacks has become an increasingly interesting research topic in the domains of Computer Vision (CV) and Natural Language Processing (NLP). DL has also been widely adopted to diverse PHM applications, where data are primarily time-series sensor measurements. While those advanced DL algorithms/models have resulted in an improved PHM algorithms' performance, the vulnerability of those PHM algorithms to adversarial attacks has not drawn much attention in the PHM community. In this paper we attempt to explore the vulnerability of PHM algorithms. More specifically, we investigate the strategies of attacking PHM algorithms by considering several unique characteristics associated with time-series sensor measurements data. We use two real-world PHM applications as examples to validate our attack strategies and to demonstrate that PHM algorithms indeed are vulnerable to adversarial attacks.