Leveraging Deep Learning for Cyberattack Detection: Advancing SCADA System Security in the IIoT Era
Djahida Belayadi, Haithem Lebkara, Narimane Ouar, Kheira Lakhdari · 2025
In today’s industrial landscape, Supervisory Control and Data Acquisition (SCADA) systems are essential for managing critical infrastructures. However, these systems are increasingly targeted by sophisticated cyber threats that jeopardize their integrity and availability. This study leverages deep learning techniques to enhance SCADA security by detecting and classifying cyberattacks. We evaluate Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Long Short-Term Memory (LSTM) networks on benchmark SCADA datasets, including WUSTL-IIOT-2018 and WUSTL-IIOT-2021. Our results show that deep learning models achieve outstanding accuracy, reaching up to 99.99% on WUSTL-IIOT-2021. These findings highlight the potential of deep learning for developing intelligent, adaptive intrusion detection systems, strengthening SCADA security against evolving cyber threats.