Reinforcing smart grid integrity: an enhanced cybersecurity framework for plug-in hybrid electric vehicles

PHEVs offer significant advantages in terms of reducing carbon emissions and reliance on fossil fuels, making them increasingly popular in the transition towards sustainable energy sources. However, their integration introduces vulnerabilities to cyber-attacks, such as FDIA, which can compromise the integrity and reliability of SGs. While existing approaches may offer some level of protection against cyber threats, they often suffer from limitations such as overfitting and inaccurate outputs. Acknowledging PHEVs’ dual role as energy sources and loads and their susceptibility to FDIA, the research proposes an innovative security strategy using IHHT combined with a deep learning approach. The IHHT analyses voltage and current data from smart sensors, extracting optimal features to enhance FDIA detection in PHEVs. Moreover, an RNN strategy, optimized using IGWO algorithm, is developed to classify the attacks. This improved technique significantly enhances the algorithm’s convergence and exploration capabilities, reducing its tendency to reach local optima. When tested using MATLAB, the model demonstrates a high accuracy rate of 97.35% in detecting FDIA; consequently, integrated energy systems become safer and more reliable. Thus, a robust and efficient framework for ensuring the safety of integrated energy systems is provided in this study, which contributes to the field of smart grid cybersecurity by providing accurate detection outputs.