Stealthy False Data Injection (FDI) Attacks on Oil and Gas Smart Sensor Networks Using Wasserstein GAN with Gradient Penalty (WGAN-GP): Evaluation and Impact Analysis

The oil and gas industry is increasingly reliant on smart sensor networks to monitor and control critical infrastructure. This dependence also increases vulnerability to cyberattacks, especially False Data Injection (FDI) attacks. This paper proposes a novel FDI attack model using Wasserstein GAN with Gradient Penalty (WGAN-GP) to generate hidden data modifications that evade traditional detection methods. We assess the effectiveness of the WGAN-GP-based attack in a simulated oil and gas smart sensor network, comparing it to conventional FDI attacks in terms of detection rate, impact score, and statistical similarity. The results show that WGAN-GP-based attacks are significantly harder to detect and cause more severe disruptions. This highlights the inadequacy of current detection mechanisms and underscores the need for enhanced protection methods to defend oil and gas infrastructure against advanced cyber threats.