Early weak anomaly detection and uncertainty quantification of equipment based on multi-task and multi-domain temporal memory autoencoder
To avoid severe malfunctions of industrial equipment, it is necessary to perform accurate detection in the early stages of abnormal occurrences. However, early anomalies are usually weak, difficult to model anomalous features, and affected by data uncertainty and training uncertainty. To address the...
Uloženo v:
| Vydáno v: | Engineering applications of artificial intelligence Ročník 162; s. 112735 |
|---|---|
| Hlavní autoři: | , |
| Médium: | Journal Article |
| Jazyk: | angličtina |
| Vydáno: |
Elsevier Ltd
26.12.2025
|
| Témata: | |
| ISSN: | 0952-1976 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Shrnutí: | To avoid severe malfunctions of industrial equipment, it is necessary to perform accurate detection in the early stages of abnormal occurrences. However, early anomalies are usually weak, difficult to model anomalous features, and affected by data uncertainty and training uncertainty. To address these limitations, we propose a multi-task and multi-domain temporal memory autoencoder (MTMAE). In the encoding stage, we design a temporal feature learning convolutional encoder and a frequency-aware temporal block to fuse time-domain and frequency-domain anomalous features, thereby creating a cloud-enhancement feature modeling (CEFM) approach based on cloud model theory to mitigate uncertainty in the data. After obtaining the latent features of the encoder, the reconstruction task uses the deconvolution network to recover the data, forming an encoder–decoder adaptive matching. The encoding memory task uses an external attention memory unit scorer to memorize potential patterns in the data. In addition, we design an optimized regularization uncertainty weighting (UW) method to balance the two tasks and penalize training uncertainty. The experimental results of five public datasets demonstrate the superiority of MTMAE in anomaly detection, with an average F1 score of 0.871 and an area under the precision–recall curve of 0.887. In the actual anomaly detection of private marine diesel engine data, MTMAE can detect early weak anomalies fastest and has the lowest false alarm rate. In addition, we also demonstrated the contribution of CEFM and UW methods to the model’s resistance to uncertainty through noise set detection and model-independent detection experiments.
•We propose a multi-domain autoencoder to enhance time series modeling.•We design a memory scorer based on external attention to score anomalies.•Multi-task learning with uncertainty minimizes reconstruction error and score.•We use cloud model theory to quantify data uncertainty and assist the encoder.•We design noise set and model independent tests to evaluate model robustness. |
|---|---|
| ISSN: | 0952-1976 |
| DOI: | 10.1016/j.engappai.2025.112735 |