Dates: 2022
Principal Investigator: Dr. Md Shafiullah
Description: The global smart-grid market is expected to grow exponentially over the coming years. However, like traditional power system networks, smart grids may experience various kinds of faults that will cause customer-minute loss (CML). Precise fault information plays a vital role in expediting the power restoration process to reduce the CML after being subjected to any kind of fault. Besides, artificial intelligence (AI) garners many front-page headlines daily as the technology enables machines to learn from experience and perform human-like tasks. Though the term 'AI' was coined in the 1950s, the field is still evolving due to the invention of new high-speed computing devices. Therefore, the advantages of AI can be incorporated into diagnosing smart grid faults to reduce the outage duration after being subjected to faults. This research proposes an AI-based fault diagnosis approach for smart grids considering the integration of intermittent renewable energy resources. Among different variations of AI tools, artificial neural networks (ANN), transparent neural networks (TNN), deep learning machines (DLM), support vector machines (SVM), extreme learning machines (ELM), etc., will be deployed. Besides, it models the smart grids, renewable energy resources, and dynamic load profiles in a real-time digital simulator (RTDS) machine. Finally, it develops the fault diagnosis (detection, classification, and location) schemes using advanced signal processing-based machine learning tools. It also investigates the sensitivity and robustness of the proposed approach under renewable intermittency, dynamic loading, and the presence of measurement noises. The outcomes of this research will assist in reducing the CML that eventually positively impacts society.