Planning and Verification of a Cloud-Based Monitoring DC Microgrid System Using the Greedy Algorithm

This paper aims to implement a cloud-based monitoring DC microgrid system suitable for communities by integrating a simulated utility grid system (SUGS), battery energy storage system (BESS), solar power generation system (SPGS), and cloud-based front-end monitoring interface and technology. Additionally, the paper utilizes Long Short-Term Memory (LSTM) model to predict the next day's load curve and employs a solar simulator to simulate the daily variations in solar irradiance. Furthermore, to enhance economic benefits during peak and off-peak time-of-use (TOU) pricing periods, this paper adopts the concept of local selection using a greedy algorithm to optimize energy allocation between SUGS, BESS, and SPGS through cloud computing. Finally, a derating case study is conducted through simulations and experiments to verify the economic value and feasibility of the proposed greedy algorithm in a community-based DC microgrid.