Electrical and Thermal Power Sharing in Modified District Cooling System for Local Energy Community

This article presents an Energy Management System (EMS) tailored for a Modified District Cooling System (MDCS) implemented within building clusters, featuring distributed cooling power plants at each building site. The primary objective is minimising electric power consumption across the cooling plant cluster while reducing cooling power losses within the water pipe network. It employs the Neural Hierarchical Interpolation for Time-Series Forecasting (N-HiTS) method, trained using simulated historical cooling load data to forecast the hour-ahead cooling power load demand for individual buildings. Subsequently, an optimal energy management system is devised using the Grey Wolf Optimization (GWO) algorithm, which determines the optimal operating frequency for the connected water pumps in the pipe network. The article demonstrates the algorithm’s effectiveness in optimizing the operation of a building-cluster cooling plant and evaluates energy consumption during an operating hour.