Anomaly detection for multivariate times series through the multi-scale convolutional recurrent variational autoencoder

To realize the anomaly detection for industrial multi-sensor data, we develop a novel multi-scale convolutional recurrent variational autoencoder (MSCRVAE) model. It is a hybrid of convolutional autoencoder and convolutional long short-term memory with variational autoencoder (ConvLSTM-VAE). The Con...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Expert systems with applications Ročník 231; s. 120725
Hlavní autoři: Xie, Tianming, Xu, Qifa, Jiang, Cuixia
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Ltd 30.11.2023
Témata:
Anomaly detection
Inner correlation
Multi-scale convolutional recurrent variational autoencoder
Multi-sensor data
Multivariate time series
Variational autoencoder
Multi-sensor data
Variational autoencoder
Inner correlation
Multi-scale convolutional recurrent variational autoencoder
Multivariate time series
Anomaly detection
ISSN:0957-4174, 1873-6793
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:To realize the anomaly detection for industrial multi-sensor data, we develop a novel multi-scale convolutional recurrent variational autoencoder (MSCRVAE) model. It is a hybrid of convolutional autoencoder and convolutional long short-term memory with variational autoencoder (ConvLSTM-VAE). The ConvLSTM-VAE part helps the MSCRVAE model not only capture the spatial and temporal dependence of data but also learn robust high-level representations by estimating the distributions of latent variables. Moreover, a typical loss function that combines the characteristics of both autoencoder and variational autoencoder is designed for the MSCRVAE model. The experimental results on three public data illustrate the superiority of the MSCRVAE model in anomaly detection with the best average F1-score up to 0.90. The interquartile range of the boxplots on the public data also proves the robustness of the MSCRVAE model. We also illustrate the role of the two components in the loss function by exploring their changes during training. What is more, on the private data, the MSCRVAE model also performs well and the residual matrices provide reasonable interpretations to the anomaly detection results. •We propose the MSCRVAE model to detect anomaly for industrial multi-sensor data.•It is a reconstruction-based model with the hybrid of ConvAE and ConvLSTM-VAE.•A typical loss function is designed for MSCRVAE and its efficacy is illustrated.•MSCRVAE performs very well on three public data and the private industrial data.•MSCRVAE shows great interpretability in anomaly detection for multi-sensor data.
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2023.120725