Micro-scale Co-generation Near-isothermal-Adiabatic Compressed Air Energy Storage
Lead Research Organisation:
The University of Manchester
Department Name: Mechanical & Aerospace Engineering
Abstract
Decarbonisation of the UK's energy system will require substantial action at a regional and local level. Therefore, the UK's energy system is growing rapidly to a more decentralised model by 2050 with a great level of small-scale electricity and heat generation at the distribution level, where wind and solar renewable energies will play a large role. However, the intermittent nature of these renewable sources presents a great challenge in energy generation and load balance maintenance to ensure stability and reliability of the power network. This highlights the need for electricity storage technologies as they provide flexibility to store excess electricity for times when it is in demand. The majority of recent installations deploy fast response electricity storage systems (e.g. batteries) with short-duration electricity storage (minutes-days) and short-discharge duration of up to 4 hours. However, technologies with long-duration electricity storage (days-weeks) and medium-duration discharge of over 4 hours, with negligible capacity and efficiency degradation are required to ensure power supply security in all weather conditions (e.g. wind or solar energies are not available for several days).
There are several possible technologies for long-duration energy storage, e.g., pumped-hydro storage, liquid air energy storage and compressed air energy storage (CAES). Among them, adiabatic CAES systems (ACAES) has the lowest installed energy capital costs (2-50$/kWh) for a wide range of storage applications from micro scale (few kW) to large scale (few MW). In conventional ACAES systems, the electricity is used to compress air in compressors, generating high levels of heat during the process. The heat of the compressed air is removed at the outlets of the compressors and stored in a thermal energy storage (TES) unit, while the cool compressed air is stored in a cavern at depths of hundreds of metres. To discharge the energy on demand, the cool compressed air heats up in the TES before expansion in turbines to generate electricity.
Despite its promising features for decarbonising the electricity power system, there are major challenges which hinder further development of ACAES systems, including (1) limitations on the underground geology, (2) low roundtrip efficiency and (3) thermal and structural challenges on the TES unit because of high-temperature air at the outlets of the compressors. This proposal aims to address these major challenges through development of an affordable micro-scale co-generation near-isothermal and adiabatic CAES system with overground air storage vessels (micro-Ni-ACAES). The system utilizes near-isothermal and high-efficiency compressor/expander devices, TES and heat exchanger units based on an innovative composite phase change material and air storage vessels. The project will perform a fundamental experimental and modelling analyses to gain deep insight into the flow and thermal fields in the near-isothermal compressors/expanders as well as charging and discharging kinetics of the TES unit. Both isochoric and isobaric storage processes will be analysed. These fundamental studies will lead to efficient designs of the micro-Ni-ACAES system components and further support the development of a thermodynamics-based design tool. The design tool will be used to identify the system's optimum operating condition and control strategy for steady-state and dynamic operations of the system. Additionally, the project will include a techno-economic and environmental impact assessment in order to evaluate the economic viability of the system, as well as CO2 abatement and fossil fuel savings over the system's lifetime. The proposed high efficiency co-generation micro-Ni-ACAES systems are believed to be the future of the CAES technology, eventually culminating in decentralised microgrid power network in application to district energy network or commercial sectors (e.g. business parks).
There are several possible technologies for long-duration energy storage, e.g., pumped-hydro storage, liquid air energy storage and compressed air energy storage (CAES). Among them, adiabatic CAES systems (ACAES) has the lowest installed energy capital costs (2-50$/kWh) for a wide range of storage applications from micro scale (few kW) to large scale (few MW). In conventional ACAES systems, the electricity is used to compress air in compressors, generating high levels of heat during the process. The heat of the compressed air is removed at the outlets of the compressors and stored in a thermal energy storage (TES) unit, while the cool compressed air is stored in a cavern at depths of hundreds of metres. To discharge the energy on demand, the cool compressed air heats up in the TES before expansion in turbines to generate electricity.
Despite its promising features for decarbonising the electricity power system, there are major challenges which hinder further development of ACAES systems, including (1) limitations on the underground geology, (2) low roundtrip efficiency and (3) thermal and structural challenges on the TES unit because of high-temperature air at the outlets of the compressors. This proposal aims to address these major challenges through development of an affordable micro-scale co-generation near-isothermal and adiabatic CAES system with overground air storage vessels (micro-Ni-ACAES). The system utilizes near-isothermal and high-efficiency compressor/expander devices, TES and heat exchanger units based on an innovative composite phase change material and air storage vessels. The project will perform a fundamental experimental and modelling analyses to gain deep insight into the flow and thermal fields in the near-isothermal compressors/expanders as well as charging and discharging kinetics of the TES unit. Both isochoric and isobaric storage processes will be analysed. These fundamental studies will lead to efficient designs of the micro-Ni-ACAES system components and further support the development of a thermodynamics-based design tool. The design tool will be used to identify the system's optimum operating condition and control strategy for steady-state and dynamic operations of the system. Additionally, the project will include a techno-economic and environmental impact assessment in order to evaluate the economic viability of the system, as well as CO2 abatement and fossil fuel savings over the system's lifetime. The proposed high efficiency co-generation micro-Ni-ACAES systems are believed to be the future of the CAES technology, eventually culminating in decentralised microgrid power network in application to district energy network or commercial sectors (e.g. business parks).
Organisations
- The University of Manchester (Lead Research Organisation)
- IMPERIAL COLLEGE LONDON (Collaboration)
- UNIVERSITY OF BIRMINGHAM (Collaboration)
- EDF Energy Plc (UK) (Project Partner)
- EPSRC Supergen Storage Network + (Project Partner)
- Hydrogennovate Ltd (Project Partner)
- Energy Systems Catapult (Project Partner)
- University of Minnesota (Project Partner)
- University of Nottingham (Project Partner)
- Aavid (Project Partner)
- Chinese Academy of Sciences (Project Partner)
- Siemens plc (UK) (Project Partner)
- Arup Group (Project Partner)
- FutureBay (Project Partner)
- Static Hydro Energy Ltd (Project Partner)
- ProDesign GmbH (Project Partner)
- IPT Technology Ltd (Project Partner)
- EGB Engineering (Project Partner)
- TNEI Group (Project Partner)
- Storelectric (Project Partner)
- FeTu Ltd (Project Partner)
- National Grid plc (Project Partner)
Publications

Haney L
(2025)
Modelling of near isothermal liquid piston gas compressor employing porous media for compressed air energy storage systems
in International Journal of Thermal Sciences

Jalili D
(2024)
Transfer learning through physics-informed neural networks for bubble growth in superheated liquid domains
in International Journal of Heat and Mass Transfer

Jalili D
(2025)
Physics-informed neural networks for two-phase film boiling heat transfer
in International Journal of Heat and Mass Transfer

Jang S
(2024)
Hidden field discovery of turbulent flow over porous media using physics-informed neural networks
in Physics of Fluids
Title | Computational fluid dynamic model on the compression process with porous media insert |
Description | Computational fluid dynamic (CFD) simulations are conducted to establish how each porous media insert performs against the baseline case (no porous media) in the compression process. The liquid piston is water, and the working gas to be compressed is air. The porous media cases examined are parallel plates, interrupted plates (linear and tilted) and a 'foam-like' lattice structure, all made from aluminium. The model could help investigate the complex features within the compressible, multiphase turbulent flow and use conjugate heat transfer to determine the heat transport in both fluid and solid phases. The turbulent flow is modelled using unsteady Reynolds-averaged Navier Stokes (uRANS) equations. For the multiphase problem, the volume of fluid (VOF) method can define the position of two immiscible fluids by tracking the volume fraction of both fluids and solving one set of equations. The Eulerian VOF method is applied in Ansys FLUENT to track the interface between the water and air at each time step. |
Type Of Material | Improvements to research infrastructure |
Year Produced | 2025 |
Provided To Others? | Yes |
Impact | The model could help to achieve high-fidelity, three-dimensional, unsteady Reynolds-averaged Navier-Stokes (uRANS) simulations to better understand the heat transfer and fluid physics involved in the liquid-piston-driven compression process. Furthermore, the uRANS is coupled with conjugate heat transfer to study using porous media inserts to manage the temperature increase. We simulate the entire cylinder/porous media arrangement using the volume of fluid (VOF) method to analyse the turbulent, multiphase physics and the fluid-structure interaction, providing a greater understanding of this process. It also investigates how porous media inserts perform against the no insert (baseline) cases in producing a near-isothermal process. The porous mediums used are parallel plates, interrupted plates, and metal foam, all produced from aluminium. |
Description | Collaboration with Imperial College London on CFD analysis |
Organisation | Imperial College London |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | The contribution is to study experimentally WPAC/Es (Water Piston Air Compressor/Expander) with porous inserts to develop near-isothermal and high-efficiency compressors and expanders, and collect data for validation of the solver. Based on the YM's preliminary experimental works on WPAC/E, a testbed will be developed to meet the requirements. Tests will be performed for air compression and expansion in LP WPAC/E from p0 = 1 bar to popt and in HP WPAC/E from popt to final pressure of pF = [30-200 bar] The optimum pressure ratio (popt/po) maximising the µ-Ni-ACAES system's efficiency will be identified by the design tool. The tests will begin with the analysis of geometrically defined porous media of parallel plates and then interrupted plates. The tests will then be extended to more complex geometry porous media made of open-cell metal foams, which have higher surface area than the plate cases. Optimum porous medium configurations identified will be tested. Experiments will be conducted for aluminium plates with different thicknesses in the range of t/D = [0.01-0.05], heights of h/H = [0.2-1] and spacing of s/D = [0.05-0.2]. H and D are the height and diameter of the WPAC/E cylinder, respectively. The parallel and interrupted plates will be made using laser-processing facilities at UoM. Aluminium open-cell foams with a pores-per-inch (PPI) range of 5-20 will be tested. Water flow rate in the range of ? = [350-2500 cm3s-1] will be examined to achieve a power density range of ?PW = [100-800 kW/m3]. In total, measurements for 20-25 cases will be performed to provide enough data for WPAC/Es performance analysis and validation of the solver. |
Collaborator Contribution | The contribution is to perform computational fluid dynamics (CFD) analyses to shed light into the local flow and thermal fields in WPAC/Es with porous medium, which is extremely challenging to achieve experimentally. The contribution involves: (i) validating the CFD model against the experiments for calculating the flow properties (e.g., pressure and temperature) and WPAC/Es efficiencies; (ii) developing empirical correlation for the heat transfer coefficient describing heat exchange between the air/water and porous medium required for the thermodynamic design tool. There are some empirical correlations for this coefficient in the literature, obtained mainly for fully developed and incompressible flows, where unsteady heat transfer and associated losses are neglected - conditions which are not satisfied in WPAC/Es. Thus, deployment of these correlations in thermodynamic models leads to very different predictions of temperature field in WPAC/Es; (iii) identifying optimum property of the porous medium to be tested for the development of near-isothermal WPAC/Es. The fluid flow inside the WPAC/E cylinder involves unsteady compressible multiphase flow with transition of air flow regime from laminar to turbulent, conjugate heat transfer and existence of small-scale vortices due to Rayleigh-Taylor instabilities. Direct numerical simulation of such a complex flow system is computationally prohibitive. Thus, the transient flow and thermal fields in the system will be solved using a large eddy simulation (LES) approach in an open-source software, OpenFOAM. The volume of fluid method will be used to capture the interface between the air and water. Possible water piston evaporation, and condensation of water vapour will be studied. For the open-cell foams, the samples geometry will be obtained using the µ-CT scanning technique, followed by image processing in an open-source package Fiji. Analysis will be performed for different porous medium. In total, 30-35 cases will be simulated to establish an empirical correlation for the heat transfer coefficient. Unstructured meshes of about 20-25 million cells will be generated in OpenFOAM. The simulations will be performed on the ICL's HPC cluster deploying 350+ cores. The optimum porous properties identified will be tested. By post-processing the LES data, a correlation model for the local heat transfer coefficient (h) will be determined, which is required for the design tool. |
Impact | The project is at the initial stage. At this stage, fundamental fluids work including some small-scale capillary-type flows and bubble formation and transport through porous media, and experimental tests on the micro-scale experimental setup have been achieved. Moreover, challenges on the design and manufacture of a full-scale (100+ bar) rig, achieving an isothermal process, establishing a heat-transfer model are discussed. |
Start Year | 2024 |
Description | Collaboration with University of Birmingham on TES |
Organisation | University of Birmingham |
Country | United Kingdom |
Sector | Academic/University |
PI Contribution | The contribution is to study experimentally WPAC/Es (Water Piston Air Compressor/Expander) with porous inserts to develop near-isothermal and high-efficiency compressors and expanders, and collect data for validation of the solver. Based on the YM's preliminary experimental works on WPAC/E, a testbed will be developed to meet the requirements. Tests will be performed for air compression and expansion in LP WPAC/E from p0 = 1 bar to popt and in HP WPAC/E from popt to final pressure of pF = [30-200 bar] The optimum pressure ratio (popt/po) maximising the µ-Ni-ACAES system's efficiency will be identified by the design tool. The tests will begin with the analysis of geometrically defined porous media of parallel plates and then interrupted plates. The tests will then be extended to more complex geometry porous media made of open-cell metal foams, which have higher surface area than the plate cases. Optimum porous medium configurations identified will be tested. Experiments will be conducted for aluminium plates with different thicknesses in the range of t/D = [0.01-0.05], heights of h/H = [0.2-1] and spacing of s/D = [0.05-0.2]. H and D are the height and diameter of the WPAC/E cylinder, respectively. The parallel and interrupted plates will be made using laser-processing facilities at UoM. Aluminium open-cell foams with a pores-per-inch (PPI) range of 5-20 will be tested. Water flow rate in the range of ? = [350-2500 cm3s-1] will be examined to achieve a power density range of ?PW = [100-800 kW/m3]. In total, measurements for 20-25 cases will be performed to provide enough data for WPAC/Es performance analysis and validation of the solver. |
Collaborator Contribution | UoB mainly contributes to the study on the TES. The dynamic modelling of the TES will be based on a transient analysis of cPCM, adopting the enthalpy-porosity approach and by solving mass, momentum and energy conservation equations in OpenFOAM. Deployment of aluminum fins in the cPCM side for enhancing the melting and solidification rates, will be investigated. Plate HXs from Swep will be adopted. Such HXs have been used in the world's first liquid air energy storage plant, which has shown to give the level of technical features that match the requirements in this project. HXs will be modelled employing the CFD solver used for the TES unit. The solver will be validated against the experimental data. The validated modelling tool will then be used to optimise the design of the TES and HXs units for charging and discharging processes under different conditions for measured cPCMs properties, TES and HXs geometries, and CAES system operating conditions. This will inform the project for building the TES and HXs units. UOB also contributes to cPCM formulation specifically tailored for a phase change temperature up to ~200oC according to the µ-Ni-ACAES system requirements, cPCM modules manufacture and characterisation, and TES and HX units assembly. Optimum designs of the TES and HX units will be provided by the modelling. Although other options will be considered, ternary eutectic nitrate salt formulation will be used for matching the TES store at ~200°C; MgO will be used as a skeleton due to proven chemical and physical compatibility at the temperature range, cost-effectiveness, non-toxic and high surface energy towards the cPCM; and graphite will be used as the heat transfer enhancement material, due to cost-effectiveness, high thermal conductivity and chemical stability at the required temperature range. The fabrication of the cPCM plates will involve milling, mixing, granulation, shaping (tableting, extrusion, cold or hot press) and light-sintering. The raw materials and the cPCMs will be fully characterised for their physical and thermal properties, and evolution of microstructures during repeated heating-cooling cycles. This will be achieved using various characterisation devices (e.g., differential scanning calorimetry, Fourier-transform infrared spectroscopy, X-ray diffraction, laser flash thermal conductivity meters and X-Ray Micro Tomography). An existing experimental system at UoB will be modified to meet the requirements. |
Impact | The project is at the initial stage. At this stage, material selection, establishment of the model for the TES, and experimental tests on the micro-scale experimental setup have been achieved. Moreover, further plans on the design of the TES and the whole rig are discussed. |
Start Year | 2024 |
Description | Meeting on the project progress |
Form Of Engagement Activity | A formal working group, expert panel or dialogue |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Study participants or study members |
Results and Impact | Discussion of the meeting Lee Haney - UoM: 1. Introduced the compressed air energy storage (CAES) system and the micro-scale experimental setup built at UoM. Discussed further challenges with design and manufacture of full-scale (100+ bar) rig. 2. CFD modelling results shown for compression and expansion, showing temperature plots and discussing challenges faced when trying to achieve an isothermal process, mainly during expansion stage. 3. Regulating pump speeds to control compression/expansion times was mentioned. Was discussed that if we can control the ramp time to slow down the initial drop in pressure during expansion, as during a fast compression 60/70% of the pressure drop is in the first second, this could potentially control the initial temperature drop. Xingchao Han - UoB: 1. Introduced the project tasks and detailed plans led by the University of Birmingham. 2. Presented the work completed in terms of material selection and models. Using parameters provided by Lee for final temperature (440 K), a PCM (KNO2-NaNO3) was chosen with a melting point of 149°C. 3. Further discussions including domain setup, numerical simulations, material preparation, characterization, and manufacture. Discussed results on using tube and fins method of heat transfer to the PCM. Two different arrangements are applied with either 8 or 12 fins used. Sapin, Paul M C - ICL: 1. Paul's presentation discussed the reduced order modelling they has performed at Imperial. Also, regarding the Computational fluid dynamics heat transfer enhancement with the porous media. He showed some fundamental fluids work including some small-scale capillary-type flows and bubble formation and transport through porous media. 2. A VOF-based conjugate heat transfer model was discussed. Talked about which is the best way to implement a CHT model and heat transfer correlation in OpenFOAM. The use of COMSOL or FLUENT was also discussed regarding potential advantages for each. Action Plan Agreed • Next meeting to be arranged in approximately 6 months at Imperial. • Online meeting to be arranged in the interim. UoM - Lee meeting with Paul to further discuss system modelling. - Also, to meet with UoB discuss further working conditions to for LPGC/TES interaction. ICL - Meet with Lee to discuss heat transfer methods in porous media. - Continue with modelling work already produced. UoB - Birmingham to finalise the charging process (air leaving LPGC) and to investigate discharging process (air entering LPGC), in which solidification of PCM needs to be modelled. - Further optimisation of heat transfer structure to be completed. |
Year(s) Of Engagement Activity | 2025 |