scholarly journals Multi-task learning based Encoder-Decoder: A comprehensive detection and diagnosis system for multi-sensor data

2021 ◽  
Vol 13 (5) ◽  
pp. 168781402110131
Author(s):  
Junfeng Wu ◽  
Li Yao ◽  
Bin Liu ◽  
Zheyuan Ding ◽  
Lei Zhang
Keyword(s):  
Anomaly Detection ◽  
Event Detection ◽  
Large Scale ◽  
Multivariate Time Series ◽  
Sensor Data ◽  
Unified Framework ◽  
Diagnosis System ◽  
Learning Framework ◽  
Task Learning ◽  
Detection And Diagnosis

As more and more sensor data have been collected, automated detection, and diagnosis systems are urgently needed to lessen the increasing monitoring burden and reduce the risk of system faults. A plethora of researches have been done on anomaly detection, event detection, anomaly diagnosis respectively. However, none of current approaches can explore all these respects in one unified framework. In this work, a Multi-Task Learning based Encoder-Decoder (MTLED) which can simultaneously detect anomalies, diagnose anomalies, and detect events is proposed. In MTLED, feature matrix is introduced so that features are extracted for each time point and point-wise anomaly detection can be realized in an end-to-end way. Anomaly diagnosis and event detection share the same feature matrix with anomaly detection in the multi-task learning framework and also provide important information for system monitoring. To train such a comprehensive detection and diagnosis system, a large-scale multivariate time series dataset which contains anomalies of multiple types is generated with simulation tools. Extensive experiments on the synthetic dataset verify the effectiveness of MTLED and its multi-task learning framework, and the evaluation on a real-world dataset demonstrates that MTLED can be used in other application scenarios through transfer learning.

Machine Anomaly Detection and Diagnosis Incorporating Operational Data Applied to Feed Axis Health Monitoring

2011 ◽  
Author(s):  
Linxia Liao ◽  
Radu Pavel
Keyword(s):  
Anomaly Detection ◽  
Principal Component ◽  
Operating Conditions ◽  
Self Organizing Map ◽  
Test Bed ◽  
Accurate Analysis ◽  
Operating Condition ◽  
Detection And Diagnosis ◽  
Feed Axis ◽  
Operational Data

Solutions for machinery anomaly detection and diagnosis are typically designed on an ad hoc, custom basis, and previous studies have shown limited success in automating or generalizing these solutions. Reusing and maintaining the analysis software, especially when the machine usage pattern or operating condition changes, remains a challenge. This paper outlines a strategy to make use of operational data obtained from the machine’s controller and signals obtained from external sensors to provide an accurate analysis within each operating condition. Operational data collected from the controller is used both for labeling datasets into different operating conditions and for analysis. Principal component analysis (PCA) is adopted to identify critical sensors that can provide useful information. Self-organizing map (SOM)-based anomaly detection and diagnosis methods are used to automatically convert data to easily understandable machine health information for operators. Experiment trials conducted on a feed-axis test-bed demonstrated the effectiveness of incorporating operational data for anomaly detection and diagnosis.

Anomaly Detection and Diagnosis Using Pre-Processing and Time-Delay Autoencoder

2021 ◽  
Author(s):  
Bryan Liu ◽  
Jianlin Guo ◽  
Toshiaki Koike-Akino ◽  
Ye Wang ◽  
Kyeong Jin Kim ◽  
...  
Keyword(s):  
Time Delay ◽  
Anomaly Detection ◽  
Detection And Diagnosis
Sign in / Sign up

Export Citation Format

Share Document