Recent Advances in Adiabatic Compressed Air Energy Storage Systems with Thermal Energy Storage: Classification and Control Strategies

Abstract Driven by accelerating global electrification and the increasing share of renewable energy, adiabatic compressed air energy storage (ACAES) has emerged as a promising long-duration energy storage solution for grid-scale applications. The successful development of 300 MW class demonstration projects further indicates the growing commercial potential of ACAES with round-trip efficiencies exceeding 70%. This paper provides a comprehensive review of the system classification, thermal energy storage (TES), compressed air reservoirs (CARs), and cutting-edge control strategies of the ACAES systems. Distinct from previous reviews focused on individual technologies, this review establishes a multilevel, design-oriented perspective spanning component selection, system configuration, operational control, and grid-level dispatch, thereby integrating fragmented technological knowledge to systematic guidance for the design, optimization, and operation of ACAES systems. First, the characteristics and performances of low-temperature (LT-ACAES), medium-temperature (MT-ACAES), and high-temperature ACAESs (HT-ACAES) are systematically compared. Although HT-ACAES can achieve the highest round-trip efficiency, up to 80.2%, its commercialization is constrained by high material costs and long-term reliability challenges, whereas LT-ACAES is limited by thermodynamic performance limits. MT-ACAES therefore represents a favorable compromise among efficiency, cost, and technological maturity, making it a promising candidate for near-term deployment. Current studies further demonstrate that overall ACAES performance is governed strongly by system-level configuration, highlighting the importance of multistage compression and expansion, optimized heat-exchanger networks, advanced TES configurations, and reduced throttling losses. For TES, the selection of liquid and solid media is strongly temperature dependent. Water is preferable for low-temperature applications, thermal oil is more suitable for medium-temperature systems, and molten salts become advantageous at higher temperatures, while low-cost natural rocks and concrete are attractive for large-scale sensible heat storage, and metal oxides show greater potential under ultra-high-temperature conditions. For CARs, salt caverns remain the preferred choice for large-scale and long-duration applications where suitable geology exists, whereas lined rock caverns, abandoned mines, depleted gas reservoirs, abandoned oil wells, and above-ground pressure vessels provide alternatives for different geological and deployment conditions. Reservoir performance is further governed by coupled thermodynamic, mechanical, seepage, and interfacial degradation processes, which jointly determine the airtightness, structural integrity, and long-term reliability. Control strategies are subsequently reviewed across charging and discharging regulation, mode switching, and system dispatch. Their development has progressed from single-objective and rule-based control toward coordinated multiobjective regulation, transient disturbance suppression, model predictive control, machine learning-assisted optimization, and security-constrained dispatch, enabling simultaneous improvements in round-trip efficiency, dynamic response, operational safety, and grid-support capability. Finally, key research gaps are identified in high-temperature material durability, long-term multiphysics reservoir modeling, integrated turbomachinery-TES dynamics, multitime scale dispatch-control coordination, and fault-tolerant operation under extreme grid disturbances. By integrating thermodynamic design, TES material selection, CAR behavior, and advanced control strategies, this review provides systematic guidance for the design, optimization, and reliable deployment of next-generation ACAES systems.