Acoustic Anomaly Detection Using Convolutional Autoencoders in Industrial Processes

2019 ◽  
pp. 432-442 ◽  
Author(s):  
Taha Berkay Duman ◽  
Barış Bayram ◽  
Gökhan İnce
Keyword(s):  
Anomaly Detection ◽  
Industrial Processes ◽  
Acoustic Anomaly

scholarly journals OutlierNets: Highly Compact Deep Autoencoder Network Architectures for On-Device Acoustic Anomaly Detection

Sensors ◽  
2021 ◽  
Vol 21 (14) ◽  
pp. 4805
Author(s):  
Saad Abbasi ◽  
Mahmoud Famouri ◽  
Mohammad Javad Shafiee ◽  
Alexander Wong
Keyword(s):  
Machine Learning ◽  
Anomaly Detection ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Network Architectures ◽  
Design Exploration ◽  
Convolutional Autoencoder ◽  
Acoustic Anomaly ◽  
Human Operators ◽  
Computational Resources

Human operators often diagnose industrial machinery via anomalous sounds. Given the new advances in the field of machine learning, automated acoustic anomaly detection can lead to reliable maintenance of machinery. However, deep learning-driven anomaly detection methods often require an extensive amount of computational resources prohibiting their deployment in factories. Here we explore a machine-driven design exploration strategy to create OutlierNets, a family of highly compact deep convolutional autoencoder network architectures featuring as few as 686 parameters, model sizes as small as 2.7 KB, and as low as 2.8 million FLOPs, with a detection accuracy matching or exceeding published architectures with as many as 4 million parameters. The architectures are deployed on an Intel Core i5 as well as a ARM Cortex A72 to assess performance on hardware that is likely to be used in industry. Experimental results on the model’s latency show that the OutlierNet architectures can achieve as much as 30x lower latency than published networks.

scholarly journals Bayesian-based anomaly detection in the industrial processes

2020 ◽  
Vol 53 (2) ◽  
pp. 11729-11734
Author(s):  
Yijun Pan ◽  
Zeyu Zheng ◽  
Dianzheng Fu
Keyword(s):  
Anomaly Detection ◽  
Industrial Processes

scholarly journals Add-On Anomaly Threshold Technique for Improving Unsupervised Intrusion Detection on SCADA Data

Electronics ◽  
2020 ◽  
Vol 9 (6) ◽  
pp. 1017 ◽  
Author(s):  
Abdulmohsen Almalawi ◽  
Adil Fahad ◽  
Zahir Tari ◽  
Asif Irshad Khan ◽  
Nouf Alzahrani ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Supervisory Control ◽  
State Of The Art ◽  
Current Approach ◽  
Industrial Processes ◽  
Parameter Choice ◽  
Detection Algorithms ◽  
Detection Approach ◽  
Unsupervised Anomaly Detection ◽  
Threshold Technique

Supervisory control and data acquisition (SCADA) systems monitor and supervise our daily infrastructure systems and industrial processes. Hence, the security of the information systems of critical infrastructures cannot be overstated. The effectiveness of unsupervised anomaly detection approaches is sensitive to parameter choices, especially when the boundaries between normal and abnormal behaviours are not clearly distinguishable. Therefore, the current approach in detecting anomaly for SCADA is based on the assumptions by which anomalies are defined; these assumptions are controlled by a parameter choice. This paper proposes an add-on anomaly threshold technique to identify the observations whose anomaly scores are extreme and significantly deviate from others, and then such observations are assumed to be ”abnormal”. The observations whose anomaly scores are significantly distant from ”abnormal” ones will be assumed as ”normal”. Then, the ensemble-based supervised learning is proposed to find a global and efficient anomaly threshold using the information of both ”normal”/”abnormal” behaviours. The proposed technique can be used for any unsupervised anomaly detection approach to mitigate the sensitivity of such parameters and improve the performance of the SCADA unsupervised anomaly detection approaches. Experimental results confirm that the proposed technique achieved a significant improvement compared to the state-of-the-art of two unsupervised anomaly detection algorithms.

scholarly journals Deep Dense and Convolutional Autoencoders for Machine Acoustic Anomaly Detection

2021 ◽  
pp. 337-348
Author(s):  
Gabriel Coelho ◽  
Pedro Pereira ◽  
Luis Matos ◽  
Alexandrine Ribeiro ◽  
Eduardo C. Nunes ◽  
...  
Keyword(s):  
Anomaly Detection ◽  
Acoustic Anomaly

scholarly journals Deep Convolutional Clustering-Based Time Series Anomaly Detection

Sensors ◽  
2021 ◽  
Vol 21 (16) ◽  
pp. 5488
Author(s):  
Gavneet Singh Chadha ◽  
Intekhab Islam ◽  
Andreas Schwung ◽  
Steven X. Ding
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Anomaly Detection ◽  
Latent Variables ◽  
Industrial Processes ◽  
Data Set ◽  
Novel Approach ◽  
Latent Space ◽  
Latent Features ◽  
Multi Class Classification

This paper presents a novel approach for anomaly detection in industrial processes. The system solely relies on unlabeled data and employs a 1D-convolutional neural network-based deep autoencoder architecture. As a core novelty, we split the autoencoder latent space in discriminative and reconstructive latent features and introduce an auxiliary loss based on k-means clustering for the discriminatory latent variables. We employ a Top-K clustering objective for separating the latent space, selecting the most discriminative features from the latent space. We use the approach to the benchmark Tennessee Eastman data set to prove its applicability. We provide different ablation studies and analyze the method concerning various downstream tasks, including anomaly detection, binary and multi-class classification. The obtained results show the potential of the approach to improve downstream tasks compared to standard autoencoder architectures.

Sign in / Sign up

Export Citation Format

Share Document