Portal de Periódicos da CAPES

Deep learning in electrical utility industry: A comprehensive review of a decade of research

2020; Elsevier BV; Volume: 96; Linguagem: Inglês

10.1016/j.engappai.2020.104000

1873-6769

Manohar Mishra, Janmenjoy Nayak, Bighnaraj Naik, Ajith Abraham,

Electricity Theft Detection Techniques

Smart-grid (SG) is a new revolution in the electrical utility industry (EUI) over the past decade. With each moving day, some new advanced technologies are coming into the picture which forces the utility engineers to think about its application to make the electrical grid become smarter. Artificial intelligence (AI) techniques such as machine learning (ML), artificial neural network (ANN), deep learning (DL), reinforcement learning (RL), and deep-reinforcement learning (DRL) are the few examples of above-mentioned advanced technologies by which large volume of collected information being processed, and deliver the solution to the complex problems associated with EUI. In recent times, DL for artificial intelligence applications has gained huge attention in the diverse research area. The traditional ML techniques have several constrained for processing the data in raw form. However, the DL provides the options to process the raw data without extracting and selecting the feature vector. The DL techniques belong to a new era of AI development. This article presents the taxonomy of DL algorithms available in the literature applied to different problems in EUI. The main objective of this survey is to provide a comprehensive idea to the researcher/utility engineer about the applications and future research scope of DL methods for power systems studies.