Large-scale long-distance data center waste heat district heating system: Exergy analysis, system performance, and application

With the rapid expansion of artificial intelligence, data centers have become major energy consumers, yet their massive low-grade waste heat remains largely unexploited due to spatial distances and temporal load mismatches. This paper proposes a large-scale, long-distance district heating system integrated with thermal energy storage (TES) and multi-stage series heat pumps. First, a thermodynamic model based on exergy analysis is developed to optimize network parameters. Results indicate that a dual-side heat pump configuration is optimal, enabling a 50 K temperature difference that drastically reduces transmission costs. Second, dynamic simulations calibrated with local weather data identify an optimal TES capacity of 5.38 % of the total heating load, effectively balancing seasonal peak-shifting requirements with capital investment. Third, economic analysis demonstrates that for a 100 MW heat source, the viable transmission distance extends to 24.4 km. Finally, the system's performance is validated through a 60,000 m2 retrofitting demonstration. Operational data from the 2024–2025 season confirm a waste heat utilization rate of 69.61 %, a comprehensive system COP of 3.46, and a static payback period of 2.57 years, proving the proposed system is a commercially competitive low-carbon solution.