Application of an AI-based optimal control framework in smart buildings using borehole thermal energy storage combined with wastewater heat recovery

This article introduces a cutting-edge energy system to meet the residential building's high energy demands while lowering emissions and related operating expenses. The central concept revolves around rule-based control strategies designed to exploit wastewater's heat, accounting for a substantial amount of the total heating demand in residential buildings. The system is also integrated with heat pumps to recover the radiators' return water energy to preheat the ventilation air passively. Naturally-driven borehole thermal energy storage is added for post-cooling the ventilation air to incorporate higher renewable shares in building energy systems while lowering CO₂ emission. TRNSYS and MATLAB software are used to design, control, optimize, and assess the system's performance from technological, environmental, and economic perspectives with the aid of artificial intelligence. According to the results, the proposed smart system is superior due to higher efficiency and lower energy cost while saving CO₂ emissions compared to the conventional ventilation system. The results further show that the suggested smart integration effectively meets needs and reduces dependency on the centralized energy network by independently recovering or producing thermal energy through the developed control framework. According to the artificial neural network-assisted optimization outcomes, energy cost, total cost, and CO₂ index are reduced by about 41.5 USD.MWh⁻¹, 10,306 USD, and 1.7 kg.MWh⁻¹, respectively. Furthermore, there is an annual extra 3 kWh of improved energy generation thanks to the optimal storage/usage of energy. The results further show that strategic optimization techniques like reducing mass flow rate and borehole depth perform well to maximize efficiency while reducing environmental impact and saving significant money.