Sürdürülebilir Kalkınma Amaçları
Yayınlar
- BildiriYalnız künye2023Anomaly Detection in Predicted Water Treatment Data using Hybrid CNN-LSTM Network Model
Özgenç, Busra; Ayas, Selen; Dogan, Ramazan Ozgur; Çavdar, Bora; Sahin, Ali Klvanc; Ayas, Mustafa Sinasi
Water treatment systems are among the industrial control systems where it is essential to detect anomalies accurately and efficiently due to the potential threat to public health. With advances in computer science, machine learning models have been successfully used in the anomaly detection process in recent years. In this paper, a hybrid CNN-LSTM network model is proposed to detect anomalies in water systems. Using a statistical window-based anomaly detection approach, the performance of the proposed model in detecting different types of attacks is analyzed on the open SWaT dataset. Experimental results show that the precision, recall and F1-score values of the proposed model are 0.994, 0.973 and 0.983, respectively, and can be successfully used to detect anomalies in water treatment systems.
- BildiriYalnız künye2023A Hybrid CNN-LSTM Framework for Unsupervised Anomaly Detection in Water Distribution Plant
Sahin, Ali Kivanc; Cavdar, Bora; Dogan, Ramazan Ozgur; Ayas, Selen; Ozgenc, Busra; Ayas, Mustafa Sinasi
To reduce potential threats to public health and economic losses, water distribution systems, which are critical parts of industrial control systems, require accurate and effective anomaly detection. The successful incorporation of machine learning models into anomaly detection procedures has been made possible by recent developments in computer science. In this research, a hybrid CNN-LSTM framework for anomaly detection in water distribution systems is presented using Python programming language in the JupyterLab environment. The performance of the proposed CNN-LSTM network model is carefully evaluated on the publicly available WADI dataset, which contains several types of attacks for a water distribution testbed, using a statistical window-based anomaly detection approach. The experimental results show acceptable precision, recall and F1 scores of 0.8619, 0.7041, and 0.775, respectively, demonstrating the efficiency of the framework in detecting anomalies in the water distribution system.