Data-driven approaches to achieve energy transition: Statistical analysis of demand-side management and smart decision making in urban energy hub networks

Abstract

Energy hubs (EHs) have emerged as a crucial solution to address resource scarcity and environmental concerns by integrating various technologies for energy production, conversion, and storage. These hubs encompass diverse elements such as integrated heat, electricity, and cooling systems, renewable energy sources, batteries, and thermal energy storage units (TES). This study focuses on a residential EH model that receives inputs in the form of electricity, gas, and solar energy to satisfy the household's electricity, heat, and cooling requirements. To enhance the operational flexibility of this energy hub, the study employs a Demand Response Program (DRP), which facilitates load shifting, load curtailment, and the modeling of flexible heating loads. The primary objective is to develop an efficient heat and power management system that optimally schedules household devices, generation equipment, and storage units. In pursuit of this goal, Long Short-Term Memory (LSTM) techniques have been selected for renewable energy, market price, and load forecasting, enhancing the system's realism and predictive accuracy. The Wild Goats algorithm is employed to solve an optimization problem for three distinct scenarios, aiming to minimize overall power costs while maintaining user comfort levels with regard to temperature and hot water availability. Furthermore, this study incorporates an environmentally conscious approach by considering the reduction of CO2, SO2, and NO pollution originating from consumers in a multi-objective optimization framework. The findings reveal that the implementation of DRP, intelligent Plug-in Hybrid Electric Vehicle (PHEV) management, TES technologies, and LSTM-based forecasting significantly contribute to reducing power costs within the proposed EH scheme. This research underscores the potential of data-driven strategies in advancing energy transition initiatives and fostering sustainable urban energy hubs while acknowledging the role of advanced forecasting techniques in enhancing system realism and efficiency.