Climate change and traditional fossil fuel shortage have spurred the development of sustainable energy (Gallo et al., 2016; Hoag, 2011), particularly renewable energy. However, limited availability of renewable energy should be carefully handled to ensure stability of the power system (Bazdar et al., 2022; Zhang et al., 2021). Underground gas storage can overcome the intermittence challenges of renewable energy sources such as wind and solar power (Edlmann, 2024; Krevor et al., 2023), thus being a critical part of energy transition. For example, surplus electricity during low demand can be converted into compressed air energy (or hydrogen energy) and subsequently stored to cope with the demand during shortages (Masoudi et al., 2024). In the context of the ongoing Paris Agreement and longer-term climate mitigation targets, vital and reliable technologies using safe underground gas storage may play an important role in leading international collaboration and to capitalize on emerging market opportunities. The replacement of fossil fuel with renewable energy can contribute toward a low-carbon economy and also improve energy security (Al-Yaseri et al., 2023). Moreover, growing supplies of natural gas are also increasing the demand for underground storage, whereas insufficient storage facilities can prevent stable supply of natural gas (Yang et al., 2023). Hence, corresponding underground storage technologies and facilities are critical for a feasible energy security strategy (Sambo et al., 2022). The deployment of underground gas storage facilities may promote upgradation of energy storage systems, thus promoting substantial development across the energy landscape and ultimately contributing to a more sustainable future (Navaid et al., 2023).

The tightness and stability of sealing in an underground facility are two critical issues to achieve large-scale gas storage. Previous studies have focused on gas storage in salt caverns (Vandeginste et al., 2023), depleted hydrocarbon reservoirs (Liu, Zhu, et al., 2024), and aquifer reservoirs (Mouli-Castillo et al., 2019). Salt caverns are ideal gas storage reservoirs due to their very low permeability and self-healing properties, whereas cavern deformation (creep characteristic of salt rock) can adversely affect the mechanical integrity of salt caverns (Bérest & Brouard, 2003). Depleted hydrocarbon reservoirs can offer large-scale gas storage capacity, but recent studies have reported on the risk of a reaction between hydrogen and rock minerals and the resulting hydrogen loss (e.g., carbonate minerals) (Zeng et al., 2022). Aquifer reservoirs represent potentially alternative options for achieving large-scale gas storage, since they are more geographically available than depleted hydrocarbon reservoirs (Guo et al., 2021). However, whether an aquifer reservoir is suitable for hydrogen storage or not should be further evaluated because of the reactivity of hydrogen with aquifer abiotic and biotic components (Gao et al., 2024; Truche et al., 2010). Almost one million abandoned mines worldwide have the potential to provide abundant underground space for large-scale gas storage (Lin et al., 2023; McLean et al., 2020). One of the options includes lined rock caverns (LRCs). LRCs have been proven to be an effective option for gas storage since successful storage of LRCs natural gas in Sweden for 20 years (Marnate & Grönkvist, 2024). LRCs are initially designed for gas storage in hard rock strata, where the steel lining is used to seal high-pressure gas, concrete is applied to transfer internal pressure to the rock, and the sliding layer (e.g., bitumen) can be used to reduce the friction between the steel lining and concrete (Damasceno et al., 2023). If the LRC composite structures (including steel lining, concrete, and rock mass) can withstand the internal gas pressure simultaneously, the requirements for the strength of the surrounding rock can be lowered. This can enable conversion of abandoned mine roadways into LRCs for gas storage, thus repurposing legacy mines and promoting a shift in the mining industry from carbon-intensive to a clean alternative (Chu et al., 2024). However, the load-sharing mechanisms of composite structures of rock caverns and selection of the sealing material in different gas scenarios remain unclear. An experimental facility that can simulate the mechanical responses of various structures and test the gas-tight performance of the sealing material will be useful, thus enabling adaptation of LRC designs to abandoned mines.

Compressed air energy storage (CAES) in underground LRCs has significant advantages in new energy storage. Table 1 demonstrates the history of CAES development, and the achievements of China in recent years in the construction of underground caverns for CAES are described in Section 3. CAES has the advantages of long power generation time, large-scale energy storage, short construction period, flexible site selection, long operation cycle, and environmentally friendly operations (King et al., 2021). Gas storage in a rock cavern usually consists of four stages: gas charge, storage, discharge, and subsequent storage stages (Geng et al., 2024). Caverns are subjected to alternating changes in high internal air pressure and temperature (Jiang et al., 2020; Masoudi et al., 2024), and these changes impact the structural safety and stability of rock caverns. Hence, high requirements are imposed on composite structures and sealing material under multifield coupling conditions (Liang et al., 2025). In March 2022, the Jiangsu Provincial Government (China) approved the construction of a deep underground research facility for fluid matter transport research (DEFINES). Yunlong Lake Laboratory of Deep Underground Science and Engineering (Yulong Lake Laboratory) is now undertaking the construction of this major scientific research facility (e.g., Figure 1) and Professor Xiaozhao Li is the principal investigator of this project. In this major scientific research facility, compressed air and hydrogen storage experimental facilities for sustainable energy storage technologies (CAPABLE) are one of the key components. The research facility at CAPABLE can be used to investigate the mechanical performance of various structures, installation technology of various key structures, mechanical properties in a rock-concrete lining-sealing layer system, long-term performance of the sealing layer, early warning systems in rock caverns, and so on. Many potential partners can make use of this experimental facility for their own projects to examine composite structure selection, to advance sealing material development, and to develop and design a monitoring and early warning system in deep underground environments.

           

2 CRITICAL SCIENTIFIC AND TECHNOLOGICAL ISSUES AND INTERNATIONAL DEVELOPMENTS

3 CURRENT STATUS OF POLICIES, RESEARCH AND DEVELOPMENT, AND ENGINEERING APPLICATIONS IN CHINA

In January 2022, the development of new energy storage during the 14th 5-year plan period was initiated in China. This can promote large-scale, industrialized, and market-oriented development of new energy storage. In March 2022, the National Development and Reform Commission and the National Energy Administration of China jointly issued the medium- and long-term plan for the development of hydrogen energy industry (2021–2035). In November 2024, the Ministry of Industry and Information Technology of China released a plan for the high-quality development of the new energy storage manufacture industry (draft for soliciting opinions). China has already promulgated the operation and maintenance regulations for CAES power stations. The National Energy Administration has issued three energy industry codes, that is, the design code for CAES power stations, the design code for underground high-pressure gas storage reservoirs of CAES power stations, and compilation regulations for feasibility study reports of CAES power stations. These regulations and codes may provide valuable guidance in engineering CAES power stations.

In 2014, China completed the construction of a 0.5 MW nonfuel-consuming CAES demonstration project in Wuhu (Mei et al., 2015). In 2017, Tsinghua University collaborated with Qinghai University to establish a 100 kW solar–thermal-assisted CAES power station in Xining, Qinghai Province (Liang et al., 2024). Subsequently, a 10 MW CAES power station in Bijie (Guizhou province, China) and a 10 MW CAES power station in Feicheng (Shandong province, China) were successfully established in 2021. In 2022, the commissioning stage for a 100 MW CAES project in Zhangbei, Hebei Province, China, was initiated. The demonstrative operation of a 60 MW CAES power station in Jintan (Jiangsu province, China) commenced in 2022 (Figure 2a) (Li, Ma, et al., 2023). These projects indicate that China has achieved significant progress in the development and application of underground energy storage technologies. The deployment of these CAES power stations can contribute to the achievement of China's carbon target.

           

In the past 2 years, the designed power output of the advanced CAES system in China has improved from 100 to 300 MW through diversified efforts. In 2024, the 300 MW salt cavern advanced adiabatic CAES project in Yingcheng (Hubei Province) (Figure 2b) and the 300 MW CAES demonstration project in Feicheng (Shandong province) (Figure 2c) were officially connected to the grid, and the overall turnaround storage efficiency of CAES plants achieved was about 70%. By December 2023, 11 CAES demonstration projects were announced by the National Energy Administration (Figure 3), among which eight projects were artificially constructed in LRCs (Zhu et al., 2024). In addition, several LRCs for CAES power stations are under construction. These data indicate that LRCs represent a promising option for underground gas storage due to their excellent sealing properties and capability. However, stability control and sealing enhancement of LRCs still pose major challenges; hence, substantial efforts should be made in terms of research before implementing large-scale underground gas storage.

           

4 CAPABLE AT YUNLONG LAKE LABORATORY

4.1 Outline of Yunlong Lake Laboratory

Yunlong Lake Laboratory of Deep Underground Science and Engineering was established in October 2021 (Figure 4). It is a major scientific and technological innovation platform established by the Xuzhou Municipal government and relies on advantageous innovations in China University of Mining and Technology and several large state-owned enterprises. It has now been approved as a Jiangsu Provincial Laboratory. Yunlong Lake Lab is a research institute with one of its main interests in the use of underground space, and it is now undertaking construction of this major scientific research facility project, that is, a deep underground research facility on fluid matter transport (DEFINES). CAPABLE is a major part of DEFINES.

               

4.2 Several testing caverns in CAPABLE

CAPABLE is built at an in situ experimental base of Yunlong Lake Laboratory, Xuzhou, China. CAPABLE includes gas storage caverns, compressor systems, high-pressure gas transmission pipelines, monitoring systems, and so on. In the context of large-scale underground energy storage in China and the pursuit of carbon neutrality goals, the experimental facility (comprised of several caverns) is expected to be the world's first high-pressure energy storage cavern capable of investigating different support structures and sealing structures under various stratum conditions. CAPABLE mainly consists of three gas storage testing caverns and one compressor equipment cavern (Figure 5).

               

The No. 1 testing cavern is located on the northwest side; the internal diameter of the cavern is 3 m and its length is 15 m. It is used to study the design of composite lining structures to simulate the influence of different rock lithologies. It is noteworthy that LRC was initially designed for gas storage in hard rock, and the mechanical response of composite structures in soft rock has been less studied. This inhibits the extension of the LRC design to abandoned mines. The No. 1 testing cavern provides a scenario to test the mechanical response of a rock cavern using simulated soft rock, thus facilitating the usage of legacy mines for gas storage (Figure 6). The No. 2 testing cavern is located on the northeast side; the internal diameter of the cavern is 3 m and its length is 10 m. It is used to investigate the design of prestressed composite structures. The No. 3 testing cavern is located on the southeast side; the internal diameter of the cavern is 3 m and its length is 10 m. It is used to study the design of LRCs (mainly composed of a concrete lining, a sliding layer, and a steel lining). The surrounding rock of the gas storage caverns is mainly moderately weathered limestone of the Zhangxia Formation with a cryptocrystalline to microcrystalline structure. The average saturated uniaxial compressive strength is about 55 MPa (relatively hard rock), and most of the rock masses are relatively intact, with low water permeability.

               

In addition, the sealing performance of various newly invented sealing material can be validated using gas pressure sensors installed inside the LRC, and the porosity and permeability can be subsequently tested by cutting a small piece from the LRC. Pressure cells embedded in concrete and rock mass will be used to monitor their mechanical responses. A surface-mounted vibrating wire strain gauge will be applied on a steel plate surface to test the stress of the steel plate. Radial stress of different structural components will be tested and subsequently used to analyze the load transfer mechanisms in the LRC. The experimental scheme and steps are as follows:

Step 1: the gas pressure increases to 5 MPa; the plastic strain and irreversible strain that inevitable occur in the LRC are determined; the data collection method is described in Section 4.4.

Step 2: the long-term performance in terms of gas tightness and the stability of the LRC are investigated within 5 MPa, and the experimental duration can be last several days.

Step 3: the gas pressure increases up to 7 MPa, and the gas tightness of the sealing material and more complicated load transfer and crack propagation in the LRC can be analyzed.

4.3 Investigation of structural components of a LRC

• 1.

  Sealing layer

  CAPABLE in Yunlong Lake laboratory provides high-pressure and high-temperature conditions for the investigation of sealing and mechanical behaviors of the sealing material. The gas pressure sensors installed inside the LRC can be applied to test gas pressure variations, thus facilitating the evaluation of gas tightness of the sealing material. Moreover, gas leakage can be monitored and identified using a distributed fiberoptic sensing approach. We use the same sensors to investigate the thermal and mechanical responses of the structural components during the storage of different gases. It is noteworthy that the same injection rate may result in different thermal and mechanical responses, since air and hydrogen have different thermodynamic properties. After the experiment, the sealing materials can be drilled to subsequently measure the variation of gas permeability, microstructure, and mechanical parameters (such as the elastic modulus). As an open collaborative platform, future work will indeed investigate the sealing performance of various sealing materials invented by engineers and researchers in different fields. We believe that partial experimental data will be obtained and reported in the near future. CAPABLE at Yunlong Lake Laboratory can focus on material development and installation issues related to the sealing layer.

• 2.

  Sliding layer

  Researchers in Yunlong Lake Laboratory have invented a bitumen-based sliding layer material (Figure 7) and evaluate its effect on the steel lining by sandwiching the sliding layer between concrete and the steel lining. The manufacturing method of sliding layers is similar to that of a bitumen waterproofing membrane, and the sliding layer is mainly prepared using a roller compactor under controlled temperatures (100–150°C). Moreover, a polyethylene (PE) film is placed on the bitumen surface as a covering to prevent adhesion, and an automatic coiler is applied to roll the produced sliding layers. CAPABLE will be applied to quantitatively evaluate to what extent the adverse effects of shear resistance on the steel lining and concrete can be mitigated.

• 3.

  Reinforced concrete lining

  Each testing cavern is equipped with a sliding layer between the steel lining and concrete, which can transfer radial force and reduce the friction between the steel lining and concrete (Figure 8), thus mitigating the cracking of concrete to a certain extent. Furthermore, the maximum allowable cracking criterion is used to design the concrete structure, and the allowable width of reinforced concrete is 0.2 mm (the maximum reinforcement ratio is less than 2.5%). In particular, the No. 2 testing cavern was designed using a prestressed concrete structure. Pre-stressing concrete has the potential to mitigate the cracking of concrete, since it can reduce the adverse effect of tensile stress. Moreover, further research is still needed to investigate the cracking of the concrete lining and to determine the load shared by composite structures (e.g., surrounding rock, concrete lining, and sealing layer).

• 4.

  Concrete plug

  The CAES and hydrogen storage experimental platform at Yunlong Lake Laboratory can enable further research on the shear resistance performance and leakage mitigation of concrete plugs considering various stratum characteristics and working conditions (Figure 9). Furthermore, the mechanical performance of the concrete plug under long-term cyclic loading and chemical degradation can be investigated, thus enabling the design of a concrete plug.

               

               

               

4.4 Monitoring system at CAPABLE

The experimental facility has an advanced integrated monitoring system. This system comprises four functionally distinct modules: real-time monitoring system, a data recording and analysis system, a remote monitoring and equipment management system, and a sophisticated user rights management system for secure access control. It can realize real-time monitoring of the internal variables of caverns, including the stress and deformation of the sealing structure and the surrounding rock, cracking of the concrete lining, and the potential leakage risk. The experimental facility has 18 monitoring items and more than 300 sensors have been installed. For example, a strain field optical fiber system and a temperature optical fiber system are used to investigate the deformation of the steel lining and temperature variations of compressed gas, and optical cables are arranged on the inner surface of the steel lining in axial and circumferential directions. Circumferential strain monitoring sections are spaced at 0.5 m intervals, with eight axially oriented fiberoptic sensing cables deployed in each testing cavern. Given that gas leakage will lead to temperature decrease, the installed temperature optical fiber system can locate the source of gas leakage. In addition, circumferential cracks of reinforced concrete are monitored by arranging high-density optical fiber cables, monitoring sections are arranged at intervals of 3 m, and the optical cables are bonded onto the steel bar. Concrete cracking exceeding a certain extent will result in an early warning of risk. The compressor system adopts the two-stage compression scheme with a maximum loading capacity of 8 MPa, and provides remote startup and termination of applications. CAPABLE can accurately simulate the real physical–mechanical response of a LRC during the storage of compressed air and hydrogen. It provides a scientific and experimental facility for newly emerging technologies and engineering tests, promoting the construction and deployment of large-capacity, ultra-long-duration, high-efficiency, and high-safety LRCs, thus accelerating the development of the energy storage industry. Potential failure of these embedded sensors may occur in the LRC, and may thus influence the data collection and analysis. Hence, three monitoring sections are arranged in each testing cavern, and the number of sensors is increased to 1.5 times the required sensors considering the failure rate of the sensors. A data integration, display, and early warning system has been developed for CAPABLE. The monitoring results are visualized through an interactive 3D virtual model of LRCs. The specific functions are as follows:

• 1.

  Real-time dynamic display of data in the form of graphs and tables.

• 2.

  Data editing and calculation can be performed; the time-history data can be graphically presented, enabling calculation of rate-of-change for any selected interval, identification of maximum/minimum values, and determination of mean values across any specified time interval.

• 3.

  The system has real-time data transmission capability.

• 4.

  Alerting functionality can be achieved: the system automatically alerts operating staff if monitored parameters approach predefined threshold values (e.g., steel plate strain reaching 80% of yield strain).

• 5.

  The monitoring data can be temporarily saved and cleared.

Yunlong Lake Laboratory is equipped with an advanced research apparatus and cutting-edge testing facilities (including excellent on-site testing sites). It has the world's largest in situ experimental site with an area of 20 000 square meters and a space of 100 000 cubic meters. Nine scientific research platforms related to integrated drilling and the detection equipment development in deep underground spaces have been successfully established. In 2023, the laboratory was authorized by China's Ministry of Natural Resources to build a field station to observe and research the evolution of subsurface environments. This station, comprising the Xuzhou main station, along with Jiawang, Fengpei, and Pizhou sub-stations, has been established to address deficiencies in the observation systems in the field of deep underground research (Figure 10). The Xuzhou main station was built at Yunlong Lake Laboratory to observe multiphysical fields at both surface and deep underground spaces during the operation of high-pressure rock caverns. These stations facilitate the investigation of underground space development and the dynamics of environmental evolution and catastrophic events by conducting long-term monitoring of ground displacement, surface subsidence, and ecological transformation and recovery triggered by underground space development. This can promote development of safe underground spaces and a sustainable environment.

               

5 INTERNATIONAL COLLABORATION

Underground energy storage facilities in other countries are mostly built into hard rock; therefore, multivariate early warning methods should be further studied. Many improvements have been made to CAPABLE, such as use of simulated soft rock to investigate load transfer mechanisms, use of prestressed concrete to resist the internal gas pressure, consideration of implications of a sliding layer to determine its positive effect on the steel lining, and use of advanced monitoring to detect potential gas leakage and failure of structures. These innovative and unique measures can promote the deployment of new technologies and materials in LRCs to reduce the risks of high-pressure gas storage. Yunlong Lake Laboratory is now cooperating with SINTEF (Norway), University of Strathclyde (UK), and University of Queensland (Australia) to conduct underground energy storage-related natural science and engineering research activities (Table 2).

Yunlong Lake Laboratory in China and SINTEF in Norway are jointly investigating hydrogen storage in abandoned mines, since Yunlong Lake Laboratory has excellent sites for field tests, including advanced testing facilities. The fundamental research focuses on the failure mechanism of LRCs and the leakage mechanism of the sealing component of LRCs under high-pressure hydrogen storage. In addition, SINTEF is supporting and assisting with the design of a monitoring system of LRCs, which can also be useful for providing early warning of risks during the operation of underground gas storage systems in abandoned mines. Yunlong Lake Laboratory is collaborating with University of Strathclyde (UK) to study rock fatigue mechanisms in underground hydrogen storage, since cyclic tensile and shear stresses can generate fatigue of rock and lead to mode-I, mode-II, and/or mixed-mode cracks at lower operational pressures. This fatigue can adversely affect the structural integrity of rock caverns. The collaboration between Yunlong Lake Laboratory and University of Queensland focuses on multiphysical field monitoring, multisource heterogeneous information fusion, and multiobjective prediction models for early warning of risk during the operation and maintenance of an underground energy storage system. This aids in risk prevention, intelligent operation, and safe maintenance of underground energy storage systems.

In the context of the Paris Agreement and China's carbon neutralization plan, international cooperation between developed and developing regions is necessary on underground hydrogen storage, since this may also motivate developing countries to consider use of hydrogen storage technology, and this in turn will greatly contribute to climate change mitigation. International cooperation is also vital to share valuable fundamental research and engineering experience, and to promote innovation. It is of great academic value, engineering significance, economic, and social benefits to explore the original technology of hydrogen storage in underground space of abandoned caverns, vigorously promote the utilization of underground space and accelerate the formation of hydrogen energy industry chain and thus reduce the consumption of traditional fossil energy and realize the goal of carbon target as soon as possible. This international collaboration can facilitate utilization of underground spaces in abandoned mines and promote adoption of large-scale gas storage in rock caverns. The deployment of underground gas storage can reduce fossil energy consumption and reduce the emission of greenhouse gases and harmful gases, thus contributing to achievement of the carbon neutrality target and protecting the ecological environment. This can modify energy consumption structures and transform cities running on traditional energy into sustainable low-carbon cities.

6 SUMMARY AND FUTURE PERSPECTIVES

Previous research studies have primarily focused on theoretical derivations and numerical simulations, as the lack of experimental platforms poses significant barriers to experimental validation. Furthermore, the load-sharing mechanism of composite structures and sealing material development in different gas scenarios are less studied. The CAPABLE platform has emerged as a transformative solution, addressing these gaps while advancing sustainable energy development during the global energy transition. It can be applied to simulate the mechanical responses of various structures and test the gas-tight performance of the sealing material in field test scenarios, thus accelerating the adoption of LRC designs to different mines. CAPABLE focuses on key areas for future research, such as long-term performance testing of sealing materials, mechanical responses of various structural components in soft rock, and chemical reaction mechanisms during reactive gas storage (e.g., hydrogen). Installation of advanced sensors inside the LRC and identification of physicochemical properties of these materials after reactive gas storage will facilitate development of these key areas.

CAPABLE's capabilities consist of four key areas: (1) simulation of diverse operating conditions; (2) comprehensive material testing (including sealing and sliding materials); (3) analysis of load transfer and material failure mechanisms; and (4) validation of advanced monitoring systems. Specifically, the platform enables

This experimental platform addresses critical technical challenges in large-scale commercial energy storage in underground LRCs, significantly advancing industry capabilities. CAPABLE serves as a catalyst for transforming technological innovations into practical applications, driving modernization of the energy industry.

As an open collaborative platform, CAPABLE partners with leading global research institutions to:

As a collaborative platform, future work will be conducted to investigate the performance of various structural components invented by engineers and researchers in various fields, and experimental data obtained from CAPABLE will be applied to explore the physiochemical responses of various structural components of LRCs. Through these collaborations, CAPABLE is contributing unique perspectives and innovative solutions of China toward addressing global energy security and climate change challenges, thus fostering international cooperation for a green, low-carbon, and sustainable energy future.

AUTHOR CONTRIBUTIONS

Xiaozhao Li: Conceptualization; project administration; writing—review and editing. Yukun Ji: Investigation; methodology; writing—first draft and review and editing. Kai Zhang: Methodology; investigation. Chengguo Hu: Investigation; writing—review and editing. Jianguo Wang: Supervision; investigation. Lixin He: Investigation; visualization; writing—review and editing. Lihua Hu: Investigation. Bangguo Jia: Methodology, writing—review and editing.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict interest.