Implementation of Demand Response Management in microgrids using IoT and Machine Learning

Abstract

Accounting for the increasing power demands and need to conserve natural resources, there is a necessity to develop a grid system which can allow integration of renewable power resources as well produce energy efficiently. This can be executed by using a microgrid arrangement which allows discrete management of energy and implementation of appropriate management strategies. Our paper uses the demand response management strategy to bring about a balance between the demand and generation of power. Demand response management is a system where the customer assumes an active role in the reduction of demand during peak hours in response to some financial benefits. The proposed research work has developed and simulated a demand response management system with microgrids by using potentiometers to depict the continuous variation of load and applied dynamic pricing and incentivization techniques so as to bring about active customer involvement in meeting the demand and increasing the reliability of the grid. Using IoT, all the information regarding load demand, generation and cost is continuously recorded made available for the customer readily on the web. IoT is also used for providing customers a convenient method to reduce their demand during peak hours. Machine learning algorithms are used for fruitful load forecasting for getting some insight and also making intelligent decisions in association with the system.