Abstract
The modernization of electricity networks and the integration of renewable energy resources in Internet of Things (IoT) based smart grids have led to increased variability in market prices, necessitating effective demand response (DR) strategies. To address this challenge, this paper proposes a novel Balanced Q-Learning based Demand Response System (BQL-DRS) that combines both optimistic and pessimistic targets in the Q-learning algorithm to achieve a balanced decision-making process in IoT based smart grids. It optimizes DR actions by efficiently managing consumer demand in real-time, considering IoT data from grid conditions, energy prices, and consumer preferences. The significance of the BQL-DRS lies in its ability to handle dynamic and uncertain IoT based grid environments, enabling it to make informed and cautious decisions while pursuing energy efficiency and cost-effectiveness. By effectively addressing both pessimistic and optimistic scenarios, the BQL-DRS ensures grid stability, load balancing, and substantial cost savings compared to representative models.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
