“…While some traditional methods for fault detection require feature engineering (Kato, Yairi, & Hori, 2001;Su, Sun, Gao, Qiu, & Tian, 2019;J. Wang et al, 2020), recent work has shown that end-to-end autoencoders can outperform traditional approaches (Zong et al, 2018;Maleki, Maleki, & Jennings, 2021;Heistracher, Jalali, Suendermann, et al, 2021). Autoencoders are promising and have been for minimal-configuration fault detection (Heistracher, Jalali, Suendermann, et al, 2021;Hood et al, 2021), however this unsupervised methods lack the ability for fault classification or fault location.…”