[Objective] Configuring hydrogen energy storage and waste heat recovery in data centers directly powered by green electricity constitutes an effective approach to mitigate renewable energy intermittency and improve overall energy efficiency. This study aims to verify the technical and economic feasibility of this strategy in enhancing the operational performance of data center energy systems and reducing operating costs. [Methods] Accordingly

a heat pump-assisted organic Rankine cycle is proposed to recover data center waste heat for power generation. At the same time

hydrogen energy storage is used to maintain power supply-demand balance amid fluctuations in renewable energy. Subsequently

an integrated energy-exergy-economic model is developed

and system performance is evaluated at monthly and annual time scales using 2023 meteorological data from Guiyang

China. Furthermore

a sensitivity analysis is conducted to investigate the impacts of varying server utilization rates and six low global warming potential(GWP) working fluids on overall system performance. [Results] The results demonstrate that the system achieves energy and exergy efficiencies of 57.12% and 55.36%

respectively

with a data center energy usage effectiveness of 1.3 and an energy reuse effectiveness of 0.74. Economic analysis indicates a dynamic payback period of 6 years and a net present value of 1.749 6 million yuan. Notably

at a server utilization rate of 90%

the low-GWP R161/Pentane working fluid pair outperforms the high-GWP R134 a/R245 fa combination. [Conclusion] The synergistic integration of waste heat recovery and hydrogen energy storage effectively improves techno-economic performance

providing a viable technical pathway for data center operators to invest in and deploy supporting energy systems for direct green electricity supply.