Diagnosing Faults in Smart Grids using Liquid Time-Constant Network

The adoption of smart grids (SGs) has become increasingly popular due to their ability to deliver automated and decentralized energy distribution to consumers. However, in order to ensure the reliable and secure provision of high-quality power from generation sources to end-users, SGs are required to be equipped with efficient diagnostic systems. This paper introduces an effective data-driven approach for diagnosing faults and failures in SGs. In particular, it is proposed to employ Liquid Time-Constant networks for the sake of diagnosing faults in an experimental power transmission system. To evaluate and verify the effectiveness of the proposed diagnostic scheme, a comprehensive study is conducted on a Three-Bus Two-Line experimental setup. Different scenarios by considering various types of faults and failures have been simulated in this case study, and the attained results verify the feasibility and efficiency of the proposed scheme.