Addressing Data Center Cooling Needs Through the Use of Reservoir Thermal Energy Storage Systems

Abstract

The paper provides a first-order analysis of the potential of using subsurface thermal energy storage systems, more specifically, reservoir thermal energy storage (RTES) for reducing the significant energy use and minimizing water consumption to cool information technology (IT) equipment in data centers. Augmented by an industrial advisory group (IAG) with experts from the data center and subsurface energy storage sectors, a scenario-based method was applied to perform the techno-economic analysis (TEA) for two types of data centers with different sizes at two geographical locations. The TEA was conducted to compare RTES scenarios with conventional cooling scenarios. The main conclusions from the investigation are that all RTES systems studied are technically feasible and thermally sustainable for at least a period of 20 years without major modifications in the systems. Two key factors contribute to an economically feasible and attractive RTES location: The first is a shallow, non-potable water-bearing geological formation with high transmissivity. A thick subsurface formation with high permeability can maximize storage capacity and minimize the number of wells needed; a shallow location can minimize the drilling depth. The second key condition is the potential to use free (i.e., compressorless) or inexpensive cooling for both direct data center cooling and RTES cooling. Future studies should consider using chillers for RTES cooling (in addition to dry coolers) when there is a significant grid value to do so (large difference between peak and off-peak power cost). Additionally, system optimization should be performed for specific sites to maximize the benefit of using RTES for cooling when deployed. Additional benefits, such as resiliency during high heat events, are often not captured in traditional TEA studies and should also be considered.

Keywords: reservoir thermal energy storage, thermal energy storage, data center cooling, techno-economic analysis