As the global energy transition accelerates, the large-scale deployment of renewable energy sources such as wind and solar power imposes higher demands on grid stability. Energy storage technology, particularly electrochemical storage using lithium-ion batteries, has become a critical enabler for balancing supply and demand, improving power quality, and facilitating the integration of intermittent renewables. However, one of the most pressing challenges in large-scale battery energy storage systems (BESS) is thermal management. The heat generated during high-rate charge and discharge cycles, if not efficiently dissipated, leads to accelerated aging, capacity fade, and in extreme cases, thermal runaway. Therefore, designing an effective and energy-efficient cooling system for energy storage cells is of paramount importance.
Conventional cooling approaches for energy storage cells include air cooling and liquid cooling. While air cooling is simple and low-cost, its heat transfer coefficient is limited, making it inadequate for high-power density modules. Liquid cooling, especially indirect liquid cooling via cold plates, offers superior thermal performance and temperature uniformity. However, traditional indirect liquid cooling systems rely on an additional chiller or air conditioning (AC) refrigeration unit to provide a continuous cold source, which significantly increases the system’s overall energy consumption and complexity. Moreover, the cooling liquid in cold plates is usually water-glycol mixture, requiring pumps and complex flow channels. In recent years, direct immersion cooling using dielectric fluids has emerged, but the high cost of fluids and strict pressure containment requirements limit its widespread adoption.
To address these limitations, I propose a novel indirect liquid cooling system for energy storage cells based on mechanical vapor recompression (MVR) and falling film evaporation. This system leverages the high heat transfer efficiency of phase change, uses water (with high specific heat and low cost) as the cooling medium, and integrates an MVR cycle to enhance the temperature difference between the cooling medium and the ambient, thereby enabling efficient heat rejection without a conventional vapor-compression chiller. In this paper, I present the detailed process design of this MVR falling film-based indirect liquid cooling system, simulate its thermodynamic performance using Aspen Plus, and compare it with a conventional air conditioning refrigeration system under identical cooling load (210 kW). The effects of key parameters such as cooling temperature and compression temperature rise on system performance are analyzed. Furthermore, an economic evaluation based on annualized cost and payback period is conducted to assess its feasibility.
1. Introduction
Thermal management of energy storage cells is a critical aspect of modern battery energy storage systems. The heat generated inside the cell arises from ohmic losses, electrochemical reactions, and entropic effects. In large-scale battery packs, the accumulation of heat can lead to a significant temperature rise, non-uniform temperature distribution, and eventual system failure. Effective cooling not only extends the cycle life of energy storage cells but also ensures safe operation under high C-rates.
Among various cooling strategies, indirect liquid cooling via cold plates is widely adopted in both electric vehicles and stationary storage applications. In such systems, a liquid coolant (usually water-glycol) circulates through channels embedded in cold plates that are in thermal contact with the battery modules. The heat is then rejected to the ambient via a refrigeration loop, typically a chiller or a heat pump. The efficiency of the refrigeration loop directly affects the overall energy consumption of the cooling system. For a typical 210 kW cooling load, a first-tier energy-efficient air conditioning system consumes about 62 kW of electrical power (assuming a COP of 3.3). This represents a substantial parasitic load, especially in large-scale storage plants where dozens of such units are installed.
MVR technology has been successfully applied in thermal desalination, wastewater treatment, and food processing industries. The principle involves compressing the vapor generated from evaporation to increase its saturation temperature and pressure, and then using the compressed vapor to drive further evaporation, thereby significantly reducing external energy input. By applying MVR to the cooling of energy storage cells, I can create a self-contained cycle where the water (coolant) evaporates on the surface of the battery module, the vapor is compressed, and then condensed in a heat exchanger that rejects heat to the ambient. Because the compressor only handles the vapor produced by the battery’s heat load—which is a relatively small mass flow rate due to the high latent heat of water—the compression work is much lower than the work required by a conventional vapor-compression cycle that must lift heat from a low-temperature evaporator to a high-temperature condenser against a large temperature lift.
In this work, I design an integrated system combining a falling film evaporator directly attached to the energy storage cells, an MVR compressor, a condenser (cooler), a liquid-vapor separator, and a recirculation pump. The system operates under vacuum to facilitate evaporation at moderate temperatures (around 30 °C). I simulate the thermodynamic cycle using Aspen Plus with the IAPWS-95 model for water/steam properties. The simulation results are then compared to a baseline refrigeration system with a cooling capacity of 210 kW. Additionally, sensitivity analysis on cooling temperature and compressor temperature rise is carried out. Finally, an economic assessment reveals that the MVR-based system can achieve a payback period of about two months due to its drastically reduced operating energy cost.
2. System Design and Working Principle
The proposed MVR falling film evaporative indirect liquid cooling system consists of two main subsystems: the falling film-based battery cooling module and the MVR heat pump module. Figure 1 (the image provided below) illustrates the schematic of the entire system.
2.1 Falling Film Battery Cooling Module
In each battery unit, a thin metal plate (stainless steel, aluminum, or aluminum alloy) is attached to the surface of the energy storage cell using thermal grease. This plate is shaped to form a closed cavity, referred to as the vacuum falling film chamber. A liquid distributor is installed at the top of the chamber, which evenly spreads a thin film of water (the coolant) over the inner surfaces of the plate. Under vacuum (e.g., 4.2 kPa absolute pressure, corresponding to a saturation temperature of 30 °C), the water film absorbs the heat conducted from the battery cell and evaporates. The vapor is drawn from the bottom of the chamber via a pipe to the vapor-liquid separator. The liquid water that is not evaporated is collected and recirculated by a pump back to the distributor. Because the cavity is under vacuum, the liquid film evaporates at a low temperature, enabling effective cooling even when the ambient temperature is relatively high.
2.2 MVR Heat Pump Subsystem
The vapor from the separator enters a mechanical compressor (e.g., a roots blower or screw compressor) where it is compressed to a higher pressure and temperature. For instance, a temperature rise of 15 K raises the saturation temperature to about 45 °C, increasing the pressure correspondingly. The superheated vapor then flows into a condenser (cooler), where it releases its latent heat to a cooling medium (e.g., cooling water or ambient air) and condenses back to liquid water. The condensed liquid passes through a pressure-reducing valve, which lowers its pressure to the low vacuum level in the falling film chamber. The low-pressure liquid is then pumped back to the distributor to repeat the cycle. The MVR compressor is the only major power-consuming component in the heat pump loop; the pump power is negligible in comparison.
2.3 Advantages Over Conventional Cooling
The key innovation lies in replacing the conventional vapor-compression refrigeration cycle (which requires a refrigerant and a large temperature lift) with an MVR cycle that only needs to lift the vapor temperature by a few degrees (e.g., 10–15 K) to reject heat to the ambient. Since the compressor handles only the latent heat of vaporization of water, and at relatively low pressure ratios, the power consumption is dramatically lower than that of a chiller of the same capacity. Moreover, the system uses water as the only working fluid, which is non-toxic, inexpensive, and has excellent thermal properties. The direct integration of the falling film evaporator with the battery surface eliminates the need for cold plates with complex channels and reduces the overall thermal resistance.
3. Modeling and Simulation
I established a steady-state thermodynamic model of the MVR falling film indirect liquid cooling system using Aspen Plus (version 12.0). The IAPWS-95 property method was selected for water/steam because of its high accuracy in the vacuum range. The model comprises four main unit operation blocks:
- EVAP: Represents the falling film evaporation module. It is modeled as a flash evaporator operating at constant pressure (downstream of the pressure-reducing valve). The inlet is saturated liquid at the chamber pressure, and the outlet is saturated vapor. The heat absorbed equals the cooling load transferred from the energy storage cells.
- COMP: A compressor block that compresses the saturated vapor to a higher pressure corresponding to the desired temperature lift. The compressor isentropic efficiency is assumed to be 75%.
- COND: A condenser block that condenses the superheated vapor to saturated liquid at the high pressure. The coolant side is not explicitly modeled; the heat rejected is assumed to be dissipated to ambient (temperature ~35 °C) via a cooling tower or water loop.
- VALVE: A throttling valve that reduces the pressure of the condensate back to the low pressure, with a negligible enthalpy drop.
The model further includes a liquid-vapor separator (not explicitly needed in Aspen Plus because the EVAP outputs only vapor). The recirculation pump is assumed to have a small power (≤0.5 kW) and is ignored in the compressor power comparison. The governing equations are as follows:
Compressor power consumption:
$$ W = \frac{m_v \Delta H_{id,v}}{\eta_c} $$
where mv is the vapor mass flow rate (kg/s), ΔHid,v is the isentropic enthalpy rise (kJ/kg), and ηc is the compressor efficiency (taken as 0.75).
Evaporator (cooling load) heat transfer:
$$ Q = m_w \Delta H_{w,e} $$
where mw is the water flow rate (kg/s), and ΔHw,e is the latent heat of vaporization (kJ/kg). Under the assumption that all water entering the evaporator is fully evaporated, mw = mv.
The design conditions for the baseline simulation are listed in Table 1.
| Parameter | Value |
|---|---|
| Cooling load (Q) | 210 kW |
| Cooling temperature (Tevap) | 303 K (30 °C) |
| Evaporator pressure | 4247 Pa (absolute) |
| Water flow rate (fully evaporated) | 311 kg/h |
| Compressor temperature rise (ΔTcomp) | 15 K |
| Compressor isentropic efficiency | 75% |
The simulation results show that at the design point the actual net cooling load delivered is 204.49 kW (slightly less than 210 kW due to minor heat loss and compressor inefficiency). The MVR compressor power consumption is 14.23 kW. The vapor mass flow rate is 0.0864 kg/s (311 kg/h), and the compression ratio is about 1.5.
4. Results and Discussion
To evaluate the energy-saving potential of the proposed system, I compare it with a conventional air conditioning (AC) refrigeration system that provides the same cooling load. The baseline AC system is a typical room-level chilled water unit with a COP of 3.3 (first-tier energy efficiency). For a cooling capacity of 210 kW, its electrical power consumption would be:
$$ P_{AC} = \frac{Q}{COP} = \frac{210}{3.3} \approx 63.6 \text{ kW} $$
However, manufacturers often specify a total power including fans and pumps. In the commercial product used as reference (see Table 2), the total power for 210 kW cooling is 62 kW (COP=3.39). I will adopt 62 kW for comparison.
Table 2 summarizes the comparison at the design condition (Tevap = 303 K).
| Parameter | MVR System | Conventional AC (COP=3.3) |
|---|---|---|
| Cooling load (kW) | 204.5 | 210 |
| Compressor/power consumption (kW) | 14.23 | 61.97 |
| Energy saving (%) | — | 77.04% |
The MVR system consumes only 14.23 kW, which is 77.04% less than the conventional AC system. This dramatic reduction stems from the fact that the MVR compressor only lifts the steam temperature by 15 K, whereas a conventional chiller must lift the refrigerant temperature from ~5 °C (evaporator) to ~45 °C (condenser), a 40 K lift. The power required for compression is proportional to the logarithmic mean temperature difference; a smaller lift directly translates into less work.
4.1 Effect of Cooling Temperature
Figure 2 (not shown, but data are presented) illustrates the influence of the cooling (evaporation) temperature on compressor power for different compression temperature rises. As the cooling temperature increases from 283 K (10 °C) to 313 K (40 °C), the compressor power decreases by about 10.7% for a fixed 12 K temperature rise. This is because higher evaporation temperature leads to lower pressure ratio and smaller specific volume, reducing the volumetric flow rate and thus the power input. For example, at ΔTcomp = 12 K, reducing the cooling temperature from 303 K to 283 K increases the compressor power from about 11.0 kW to 12.3 kW. Therefore, to minimize energy consumption, the cooling system should operate at the highest permissible temperature that still maintains safe battery cell temperature (typically <45 °C).
4.2 Effect of Compression Temperature Rise
As expected, the compressor power rises sharply with increasing temperature lift. For a fixed cooling temperature of 303 K, increasing the compression temperature rise from 10 K to 15 K raises the power from 9.34 kW to 14.23 kW, a 52.3% increase. A larger temperature lift allows the condenser to reject heat to a higher ambient temperature, but at the cost of higher compression work. In practical design, one must balance the ambient conditions and the thermal resistance of the condenser. Typically, a 10–15 K lift is sufficient for moderate climates; a higher lift may be needed only in hot regions.
4.3 Energy Saving Percentage vs. Parameters
The energy saving percentage relative to the baseline AC system is defined as:
$$ \text{Saving} = \frac{P_{AC} – P_{MVR}}{P_{AC}} \times 100\% $$
Using a constant baseline AC power of 61.97 kW, the saving varies from 83.9% (at ΔTcomp = 10 K) down to 77.0% (at ΔTcomp = 15 K). Raising the cooling temperature also slightly improves saving (e.g., from 80.4% at 283 K to 82.5% at 313 K for 12 K lift). This confirms that the MVR falling film system is a highly efficient alternative, especially when the temperature lift can be kept low.
5. Economic Analysis
To assess the economic viability, I performed a life-cycle cost analysis using annualized cost (AC) and payback period (PBP) methods. The initial investment of the MVR system includes: compressor ($1200), condenser ($300), liquid-vapor separator ($100), pressure reducing valve ($40), circulation pump ($50), and valves ($30), totaling approximately $1720 (converted to RMB: 17,200 ¥ using an approximate exchange rate of 1 USD = 7 RMB, though the original paper uses RMB). The baseline AC system (a commercial room-level chilled water unit) has an initial cost of about $2420 (24,200 ¥). The following assumptions are made: discount rate i = 10%, system lifetime n = 15 years, residual value S = 20% of initial cost, annual operating hours t = 3000 hours, electricity price = 0.8 ¥/kWh, maintenance cost = 10% of annualized capital cost.
The annualized capital cost (Cacc) is calculated as:
$$ C_{acc} = C_{ccm} \cdot \frac{i(1+i)^n}{(1+i)^n – 1} $$
where Cccm is the initial total investment.
Annual electricity cost:
$$ C_{rec} = C_{ee} \cdot W \cdot t $$
Annualized maintenance cost: Cmc = 0.1 × Cacc.
Annualized residual value: Sa = S × i / ((1+i)n – 1).
Total annualized cost: AC = Cacc + Cmc – Sa + Crec.
The payback period (PBP) is calculated using the discounted cash flow formula:
$$ PBP = \frac{ \ln\left(1 – \frac{C_{ccm}}{B_j} \cdot (i-r)\right) }{ \ln\left(\frac{1+r}{1+i}\right) } $$
where Bj represents the annual savings in the first year, and r = 6% is the discount rate (bank interest rate).
Table 3 summarizes the economic comparison under the same 210 kW cooling load.
| Item | MVR System | Conventional AC |
|---|---|---|
| Initial investment Cccm (¥) | 17,200 | 24,200 |
| Annualized capital Cacc (¥) | 2,262 | 3,182 |
| Annual maintenance Cmc (¥) | 227 | 319 |
| Annual electricity cost Crec (¥) | 34,145 | 148,721 |
| Total annualized cost AC (¥) | 36,525 | 152,069 |
| Payback period PBP (years) | 0.163 | — |
The MVR system’s annual electricity cost is only 34,145 ¥, compared to 148,721 ¥ for the conventional AC system—a saving of 114,576 ¥ per year. Despite a slightly higher initial investment (the MVR system is actually cheaper by 7,000 ¥ due to the elimination of the refrigerant loop heat exchanger), the huge reduction in operating cost leads to a payback period of only 0.163 years (about two months). This remarkable economic benefit makes the MVR falling film indirect liquid cooling system extremely attractive for large-scale energy storage applications where cooling loads are continuous and high.
6. Conclusion
In this study, I have designed, simulated, and analyzed a novel indirect liquid cooling system for energy storage cells based on MVR falling film evaporation. The system uses water as the coolant, operates under vacuum, and integrates an MVR compressor to reject the absorbed heat to the ambient with a small temperature lift. The key findings are as follows:
- Under a cooling load of 210 kW and a cooling temperature of 30 °C, the MVR falling film system consumes only 14.23 kW of compressor power, achieving a 77.04% energy saving compared to a first-tier energy-efficient conventional air conditioning system (COP 3.3).
- Increasing the cooling temperature or decreasing the compression temperature rise further reduces system energy consumption. A 10 K rise in cooling temperature (from 283 K to 313 K) reduces compressor power by about 10.7%; a 5 K reduction in temperature lift (from 15 K to 10 K) reduces power by 34%.
- The economic analysis shows that the MVR system’s total annualized cost is only 36,525 ¥ versus 152,069 ¥ for the conventional system, with a payback period of just 2 months. The system is thus both energy-efficient and highly cost-effective.
This work demonstrates that MVR falling film evaporation technology can be a game-changer for thermal management of energy storage cells, offering substantial energy and economic benefits. Future research should focus on experimental validation of the falling film module’s thermal uniformity and the transient behavior under variable charge/discharge rates. The integration of this system with real-world energy storage plants will further confirm its reliability and scalability.