太空数据中心热控技术研究现状与展望OA

Significance The rapid expansion of artificial intelligence,large satellite constellations,and deep-space exploration is reshaping global demand for computing infrastructure. On Earth,the continued scaling of data centers has resulted in a sharp rise in energy consumption and increasingly severe thermal constraints,driven by limitations in power supply and cooling efficiency. In the space domain,observation platforms and interplanetary missions generate a growing volume of raw data;however,their heavy reliance on down-link based processing remains constrained by limited bandwidth and communication latency. These parallel trends have stimulated growing interest in space data centers as a means of deploying computing capabilities directly in orbit or deep space. An early conceptualization of space data centers and their enabling technologies was proposed in late 2011 by researchers at the Chinese Academy of Sciences,accompanied by a patent (CN201110452453. 4). By exploiting near-continuous solar power and the cold environment of space,space data centers offer a potential pathway to reducing the overall energy cost of computation while enabling on-orbit data processing,prioritization,and storage. Their practical realization,however,is fundamentally constrained by thermal management technology. The combination of high power density,distributed heat sources,extended heat transport distances,and microgravity-induced flow instability places thermal management at the core of system design. Rather than serving as an auxiliary function,thermal control directly determines system reliability,mass efficiency,and the extent to which space data centers can be scaled beyond early demonstrators. Progress Thermal control technologies for space data centers can be broadly categorized into passive and active approaches,which together establish baseline thermal balance and provide enhanced heat transport and regulation capabilities. Passive thermal control techniques,including heat pipes,thermal interface materials (TIMs),phase change materials (PCMs),radiators,and thermal control coatings,rely on conduction,radiation,and latent heat buffering to stabilize system temperatures with minimal energy input. Advances in variable-conductance heat pipes and loop heat pipes have improved temperature regulation and long-distance heat transport,while emerging TIMs emphasize reduced contact resistance,radiation tolerance,and long-term stability. PCMs are increasingly integrated with heat spreaders and vapor chambers to buffer cyclic and transient thermal loads,and radiator technologies are evolving toward lightweight,variable-emissivity designs capable of dynamically responding to orbital environments. Active thermal control technologies play an indispensable role as input power and thermal load increase. Mechanically pumped fluid loops and pump-driven two-phase convection systems use circulating working fluids to transport large amounts of heat away from concentrated sources,offering higher heat transport capacity and improved temperature uniformity. Significant progress has been achieved in multi-kilowatt-class systems through improvements in pump reliability,accumulator design,and two-phase flow stability in a microgravity environment. Complementary active components,including heaters,thermoelectric coolers,and thermal switches,enable precise local temperature regulation,low-temperature survival in extreme environments,and adaptive control of thermal pathways. Collectively,these technologies have been validated on platforms such as space stations,planetary probes,and high-power satellites,providing a technical foundation for future space data center deployment. Conclusion and Prospect Current thermal control strategies for space data centers are largely based on the combined use of passive and active approaches and have so far supported systems with power levels on the order of several tens of kilowatts. As computing capacity continues to expand,however,these approaches are approaching their intrinsic limits. At the hundreds-of-kilowatts and megawatt levels,constraints associated with radiative heat rejection,system mass growth,and controllability under variable operating conditions are expected to intensify,placing thermal management at the core of system-level scalability. Further advancement demands integrated thermal architectures that address heat generation,transport,storage,and rejection in a coordinated manner across multiple spatial and temporal scales. Progress in microgravity two-phase heat transfer,compact thermal energy storage,and lightweight radiators with controllable emissivity will be particularly critical,alongside the development of thermal materials that combine ultralow thermal resistance with long-term tolerance to the space environment. Cutting-edge thermal management strategies,such as liquid metal cooling,are expected to play increasingly important roles in addressing the extreme heat flux challenges posed by AI chips. Advances in these directions will be decisive in determining whether space data centers can evolve from early demonstrations into a robust and scalable computing infrastructure for future space missions.