scholarly journals A PCC-Ensemble-TCN model for wind turbine icing detection using class-imbalanced and label-missing SCADA data

2021 ◽  
Vol 17 (11) ◽  
pp. 155014772110577
Author(s):  
Shenyi Ding ◽  
Zhijie Wang ◽  
Jue Zhang ◽  
Fang Han ◽  
Xiaochun Gu ◽  
...  
Keyword(s):  
Data Acquisition ◽  
Supervisory Control ◽  
Wind Farm ◽  
Pearson Correlation ◽  
Cold Climate ◽  
Data Driven ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Convolutional Network ◽  
Original Dataset

Blade icing problems are ubiquitous for wind turbines located in cold climate zones. Data-driven indirect icing detection methods based on supervisory control and data acquisition system have shown strong potential recently. However, the supervisory control and data acquisition data is annotated through manual observation, which will cause the data between normal condition and icing condition to be unlabeled. In addition, the amount of normal data is far more than icing data. The above two issues restrict the performance of most current data-driven models. In order to solve the label missing problem, this article proposes a Pearson correlation coefficient–based algorithm for measuring the degree of blade icing, which calculates the similarity between the unlabeled data and the icing data as its label. Aiming at the class-imbalance problem, this article constructs multiple class-balanced subsets from the original dataset by under-sampling the normal data. Temporal convolutional networks are trained to extract features and make predictions on each subset. The final prediction result is obtained by ensembling the prediction results of all temporal convolutional network models. The proposed model is validated using the actual supervisory control and data acquisition data collected from a wind farm in northern China, and the results indicate that ensuring the consecutiveness and class-balance of the data are quite advantageous for improving the detection accuracy.

scholarly journals An Adaptive-Data-Driven Attack Detection Framework on ADS-B Data

2021 ◽  
Author(s):  
Tengyao Li
Keyword(s):  
Traffic Management ◽  
Contextual Information ◽  
Attack Detection ◽  
Training Data ◽  
Data Driven ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Sequential Data ◽  
Security Measures ◽  
Data Share

<p>With the variety and quantity of flights increasing, accurate and efficient surveillance methods are in great demands for the next generation air traffic management. Relying on high accuracy, wide coverage, low deployment cost and data share support, Automatic Dependent Surveillance – Broadcast (ADS-B) is becoming the primary surveillance method in 2020. However, ADS-B data is lacking of sufficient security measures to ensure data integrity and authentication, which makes it face with various attack threats. To detect the malicious data caused by attack behaviours accurately, an adaptive-data-driven attack detection framework is proposed, which is utilized to establish the consistent framework for predictive discriminant detection methods. It is composed of sequential predictor, behaviour discriminator and dynamic updater, enhancing adaptive sequential detection performances. According to the framework, an effective implementation is designed to improve attack detection accuracy: (I) The sequential predictor identifies flight phases to predict sequential data effectively and accomplish model fusion to generate ADS-B predictive data sequences. (II) The behaviour discriminator utilizes value differences and contextual information to distinguish attack data from ADS-B data sequences. (III) The dynamic updater is designed to update the training data sets and discriminate threshold dynamically, improving the adaptation in face of concept drifts for ADS-B data. By experiments on real ADS-B data with diverse attack patterns, the feasibility and efficiency of the framework are validated.</p>

scholarly journals An Adaptive-Data-Driven Attack Detection Framework on ADS-B Data

2021 ◽  
Author(s):  
Tengyao Li
Keyword(s):  
Traffic Management ◽  
Contextual Information ◽  
Attack Detection ◽  
Training Data ◽  
Data Driven ◽  
Detection Methods ◽  
Detection Accuracy ◽  
Sequential Data ◽  
Security Measures ◽  
Data Share

<p>With the variety and quantity of flights increasing, accurate and efficient surveillance methods are in great demands for the next generation air traffic management. Relying on high accuracy, wide coverage, low deployment cost and data share support, Automatic Dependent Surveillance – Broadcast (ADS-B) is becoming the primary surveillance method in 2020. However, ADS-B data is lacking of sufficient security measures to ensure data integrity and authentication, which makes it face with various attack threats. To detect the malicious data caused by attack behaviours accurately, an adaptive-data-driven attack detection framework is proposed, which is utilized to establish the consistent framework for predictive discriminant detection methods. It is composed of sequential predictor, behaviour discriminator and dynamic updater, enhancing adaptive sequential detection performances. According to the framework, an effective implementation is designed to improve attack detection accuracy: (I) The sequential predictor identifies flight phases to predict sequential data effectively and accomplish model fusion to generate ADS-B predictive data sequences. (II) The behaviour discriminator utilizes value differences and contextual information to distinguish attack data from ADS-B data sequences. (III) The dynamic updater is designed to update the training data sets and discriminate threshold dynamically, improving the adaptation in face of concept drifts for ADS-B data. By experiments on real ADS-B data with diverse attack patterns, the feasibility and efficiency of the framework are validated.</p>

scholarly journals Alarms management by supervisory control and data acquisition system for wind turbines

2021 ◽  
Vol 23 (1) ◽  
pp. 110-116
Author(s):  
Isaac Segovia Ramirez ◽  
Behnam Mohammadi-Ivatloo ◽  
Fausto Pedro García Márquez
Keyword(s):  
Data Acquisition ◽  
Wind Turbine ◽  
Wind Turbines ◽  
Supervisory Control ◽  
Principal Component ◽  
Data Acquisition System ◽  
Fault Detection And Diagnosis ◽  
Acquisition System ◽  
False Alarms ◽  
Original Dataset

Wind energy is one of the most relevant renewable energy. A proper wind turbine maintenance management is required to ensure continuous operation and optimized maintenance costs. Larger wind turbines are being installed and they require new monitoring systems to ensure optimization, reliability and availability. Advanced analytics are employed to analyze the data and reduce false alarms, avoiding unplanned downtimes and increasing costs. Supervisory control and data acquisition system determines the condition of the wind turbine providing large dataset with different signals and alarms. This paper presents a new approach combining statistical analysis and advanced algorithm for signal processing, fault detection and diagnosis. Principal component analysis and artificial neural networks are employed to evaluate the signals and detect the alarm activation pattern. The dataset has been reduced by 93% and the performance of the neural network is incremented by 1000% in comparison with the performance of original dataset without filtering process.

Design of supervisory control and data acquisition system based on GIS technology

2013 ◽  
Vol 33 (2) ◽  
pp. 567-570
Author(s):  
Zeping YANG ◽  
Deqiang LIU ◽  
Qian WANG ◽  
Qiangming XIANG
Keyword(s):  
Data Acquisition ◽  
Supervisory Control ◽  
Data Acquisition System ◽  
Acquisition System ◽  
Gis Technology

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.

Sign in / Sign up

Export Citation Format

Share Document