RAKSHAK: Resilient and Scalable Demand Response Management Scheme for Smart Grid Systems
Aparna Kumari, Sudeep Tanwar
Abstract
To provide electricity in a controlled and smart way, one of the emerging technologies, Smart Grid (SG), is being used widely. It necessitates enabling real-time monitoring and control of bidirectional flows of electricity and essential data. The persistent connectivity of electricity and communication introduced a massive volume of data that demands techniques faraway greater than conventional methods for appropriate decision-making and analysis. Big Data Analytics (BDA) and Machine Learning (ML) techniques play a crucial role in procuring these benefits. So, this paper propose, RAKSHAK, A Resilient and Scalable Demand Response Management (DRM) scheme for SG system, which addresses the issue of load forecasting on demand side. Initially, it classifies the electricity load using Support Vector Machine (SVM) classifier and then employs the ML-based approaches to predict the electricity demand. The experimental results show the efficacy of RAKSHAK scheme as it obtained better prediction result compared to pre-existing expensive demand response schemes in terms of Root Mean Squared Error (RMSE) value.