Analysis of Dissatisfaction Factor on Customer’s Bills in Smart Microgrids Using Reinforcement Learning

Smart cities leverage advanced technologies and data analytics to optimize energy consumption and demand response (DR) strategies. By integrating microgrids (MGs) into the smart city framework, it becomes possible to monitor and manage energy consumption. MGs can respond to real-time data and adjust energy supply and demand accordingly, allowing for load balancing and peak shaving. This article delves into the intricate dynamics of DR management within an MG context, focusing on the implementation of the Q-learning (QL) algorithm to optimize energy utilization. The MG under examination comprises renewable energy sources such as photovoltaic (PV) and wind turbine (WT), alongside traditional components like battery storage and diesel generators (DGs). Additionally, the MG is seamlessly integrated with the main grid to ensure energy balance and reliability. Within this operational framework, customers assume a pivotal role in DR by adapting their energy consumption patterns to curtail costs effectively. Finally, this study explores how customers can actively engage in energy markets to reduce their expenditures while enhancing the overall stability and efficiency of the MG and smart city.