A Natural Watermarking Approach to Cyber Attack Detection for Power Electronics- Interfaced Renewables

This paper introduces a novel approach to detecting cyber attacks for power electronics-interfaced renewable resources, e.g., solar panels. The approach leverages the inherent variability of renewable energy generation to watermark the measurements of renewable resources that are vulnerable to false data injection (FDI) attacks. By checking the existence of the watermarks imprinted by the natural fluctuations of renewables, false data injection attacks can be detected. Compared with the conventional watermarking methods, the proposed approach does not require additional noise injection which compromises control performance. The effectiveness of the proposed approach is validated by simulating a solar photovoltaic system.