Meyer, MelinaGergeleit, MartinKrechel, DirkKlein, MaikeKrupka, DanielWinter, CorneliaGergeleit, MartinMartin, Ludger2024-10-212024-10-212024978-3-88579-746-32944-7682https://dl.gi.de/handle/20.500.12116/45091This study investigates anomaly detection methods in supermarket refrigeration systems. A transformer-based model is introduced in this field and compared with LSTM autoencoders. The models are trained and evaluated using preprocessed refrigeration data, with parameters optimized for accuracy, recall, precision, and F1 score. Our findings aim to enhance system monitoring and maintenance strategies, ultimately improving reliability, energy efficiency, and operational excellence in supermarket refrigeration technology.enAnomaly DetectionTransformerTime Series AnalysisIndustrial Use CaseText/Conference Paper10.18420/inf2024_1191617-54682944-7682