In modern naval operations, the demand for reliable and efficient energy systems has grown significantly due to the deployment of advanced weaponry and high-intensity maritime scenarios. The battery energy storage system (BESS) plays a critical role in ensuring stable power supply for ships, particularly in hybrid electric propulsion and auxiliary systems. However, the safety and reliability of lithium-ion batteries in such environments are paramount, as thermal runaway events can lead to catastrophic failures, including fires and explosions. This study focuses on simulating thermal runaway in a BESS tailored for naval platforms, comparing the effectiveness of immersion cooling and liquid cold plate cooling methods. By integrating simulation models in Fluent and Simulink, we analyze temperature dynamics under real ship operating conditions, emphasizing the importance of thermal management in enhancing the safety and performance of the BESS.
The BESS for naval applications typically consists of multiple battery packs organized in clusters, each containing numerous cells. These systems are integrated with components like the battery management system (BMS), cooling systems, and power distribution networks to meet the dynamic power demands of ships. In high-stress environments, such as during maneuvering or combat, the BESS must handle rapid power fluctuations, which can exacerbate thermal issues. Thermal runaway in lithium-ion batteries is often triggered by mechanical, electrical, or thermal abuse, leading to internal short circuits and uncontrolled temperature rise. For instance, mechanical impacts from collisions or electrical faults like overcharging can initiate this process, posing severe risks in confined ship compartments. Therefore, developing robust thermal management strategies is essential to mitigate these hazards and ensure the operational integrity of the BESS.
To address this, we developed a co-simulation framework using ANSYS Fluent for the battery thermal model and MATLAB/Simulink for the cooling system dynamics. This approach allows for real-time data exchange, simulating the fluid flow and heat transfer in the BESS under various scenarios. The immersion cooling method, which submerges cells in a dielectric fluid like hydrocarbon oil, is compared against traditional liquid cold plate cooling. Our simulations incorporate actual ship power profiles, including normal cruising, docking, and acceleration phases, to evaluate the BESS’s response to thermal runaway events. The results demonstrate that immersion cooling significantly outperforms liquid cold plate cooling in controlling temperature spikes and preventing thermal propagation, thereby enhancing the safety of the BESS on naval platforms.
Fundamentals of Marine Battery Energy Storage Systems
The marine BESS is designed to provide high-density energy storage for ships, supporting functions such as propulsion, weapon systems, and auxiliary loads. A typical BESS architecture includes battery cells arranged in modules, packs, and clusters, connected in series or parallel to achieve desired voltage and current levels. The BMS monitors parameters like state of charge, temperature, and health, while the cooling system maintains optimal operating conditions. In naval contexts, the BESS must withstand harsh environmental conditions, including vibrations, shocks, and temperature variations, making thermal management a critical aspect of design.
Thermal runaway in lithium-ion batteries involves a chain of exothermic reactions that can lead to cell failure and fire. The process begins with internal short circuits caused by factors like dendrite growth, separator damage, or external heat. Key reactions include the decomposition of the solid electrolyte interface (SEI), anode and cathode reactions, and electrolyte breakdown, releasing heat and flammable gases. The heat generation rate can be modeled using the Arrhenius equation, where the reaction kinetics depend on temperature. For a BESS, understanding these mechanisms is vital for designing effective cooling solutions. The general energy balance for a battery cell can be expressed as:
$$ \rho c_p \frac{\partial T}{\partial t} = \nabla \cdot (k \nabla T) + \dot{q}_{gen} $$
where \( \rho \) is the density, \( c_p \) is the specific heat capacity, \( k \) is the thermal conductivity, \( T \) is the temperature, and \( \dot{q}_{gen} \) is the heat generation rate. In a BESS, this equation must be solved for multiple cells, considering interactions through conduction, convection, and radiation. The heat generation in a cell during operation includes irreversible heat from internal resistance and reversible heat from entropy changes, given by:
$$ \dot{q}_{gen} = I (V_{oc} – V) – I T \frac{\partial V_{oc}}{\partial T} $$
where \( I \) is the current, \( V_{oc} \) is the open-circuit voltage, and \( V \) is the terminal voltage. For a BESS under thermal abuse, additional heat from side reactions accelerates temperature rise, leading to runaway.
Table 1 summarizes the key parameters of a typical lithium iron phosphate (LiFePO4) cell used in naval BESS applications, based on the LF280K model. These parameters are essential for accurate simulation of thermal behavior.
| Parameter | Value |
|---|---|
| Nominal Voltage | 3.2 V |
| Nominal Capacity | 280 Ah |
| Dimensions (Height × Width × Thickness) | 207.2 mm × 173.7 mm × 71.7 mm |
| Density | 2715 kg/m³ |
| Specific Heat Capacity | 956.6 J/kg·K |
| Thermal Conductivity (x, y, z directions) | 1.176, 16.128, 1.176 W/m·K |
The cooling system in a BESS is designed to remove heat from the cells, maintaining temperatures within a safe range. For immersion cooling, the dielectric fluid directly contacts the cells, enhancing heat transfer through convection. In contrast, liquid cold plate cooling uses plates attached to the cell surfaces, relying on conduction. The effectiveness of these methods depends on factors like flow rate, fluid properties, and system design. The hydrocarbon oil used in immersion cooling has properties such as a density of 810 kg/m³, specific heat capacity of 2100 J/kg·K, thermal conductivity of 0.14 W/m·K, and a flash point of 190°C. These properties influence the cooling performance in the BESS during thermal events.
Simulation Model for Immersion Cooling BESS
We developed a co-simulation model to study the thermal behavior of a BESS under immersion cooling. The model integrates Fluent for computational fluid dynamics (CFD) of the battery pack and Simulink for the cooling system dynamics. This approach enables real-time simulation of heat transfer and fluid flow, capturing the interactions between the BESS components. The battery pack consists of multiple LiFePO4 cells arranged in a module, submerged in hydrocarbon oil. The CFD model solves the governing equations for fluid flow and heat transfer, including the continuity, momentum, and energy equations.
The continuity equation ensures mass conservation:
$$ \frac{\partial \rho}{\partial t} + \nabla \cdot (\rho \mathbf{u}) = 0 $$
where \( \rho \) is the fluid density and \( \mathbf{u} \) is the velocity vector. The momentum equation accounts for fluid motion:
$$ \frac{\partial (\rho \mathbf{u})}{\partial t} + \nabla \cdot (\rho \mathbf{u} \mathbf{u}) = -\nabla p + \nabla \cdot \boldsymbol{\tau} + \mathbf{f} $$
where \( p \) is pressure, \( \boldsymbol{\tau} \) is the viscous stress tensor, and \( \mathbf{f} \) represents body forces. The energy equation models heat transfer:
$$ \frac{\partial (\rho c_p T)}{\partial t} + \nabla \cdot (\rho c_p \mathbf{u} T) = \nabla \cdot (k \nabla T) + \dot{q}_{gen} $$
In the BESS context, \( \dot{q}_{gen} \) includes heat from electrochemical reactions and internal resistance. For thermal runaway simulation, we incorporate additional heat sources from side reactions, modeled using empirical correlations based on cell temperature.
The cooling system in Simulink simulates the hydrocarbon oil circulation, including heat exchange with a freshwater loop. The dynamic equations for the cooling liquid side are:
$$ \frac{dt_{co}}{dt} = \frac{1}{W_{cc}} \left[ m_c C_c (t_{ci} – t_{co}) – K_q A_q \Delta T_q \right] $$
where \( t_{co} \) and \( t_{ci} \) are the outlet and inlet temperatures of the coolant, \( m_c \) is the mass flow rate, \( C_c \) is the specific heat capacity, \( K_q \) is the overall heat transfer coefficient, \( A_q \) is the heat exchange area, and \( \Delta T_q \) is the log mean temperature difference. The total heat capacity \( W_{cc} \) is given by:
$$ W_{cc} = M_c C_c + M_q C_q $$
with \( M_c \) and \( M_q \) being the masses of the coolant and heat exchanger, respectively. Similarly, the freshwater side equations balance the heat exchange. This model allows us to simulate the cooling performance under varying operating conditions of the BESS.
The battery pack geometry was modeled in ANSYS SpaceClaim, and meshing was performed in Fluent Meshing with a grid quality above 0.5. The simulation parameters include an initial coolant temperature of 25°C and a flow velocity of 2 m/s. The BESS operates under real ship power profiles, which include periods of steady cruising, docking with high power fluctuations, and acceleration phases. For example, during docking, power demand varies rapidly between 0 and 100 kW, while during acceleration, it spikes to peak levels. These profiles are used to define boundary conditions in the simulation, ensuring realistic scenarios for the BESS.
Table 2 provides the simulation parameters for the cooling system model, which are critical for analyzing the BESS performance under thermal stress.
| Parameter | Value |
|---|---|
| Coolant Volume Flow Rate | 0.2272 m³/s |
| Freshwater Volume Flow Rate | 0.2272 m³/s |
| Heat Exchanger Coefficient | 5.55 m²·K |
| Heat Exchange Area | 0.1224 m² |
Simulation Validation and Results
We conducted thermal runaway simulations for the BESS under two cooling methods: immersion cooling and liquid cold plate cooling. The simulations were performed for scenarios where a single cell experiences an internal short circuit, mimicking thermal abuse conditions. Two critical time points were selected based on ship power profiles: at 100 seconds during docking with high power fluctuations, and at 165 seconds during acceleration with a rapid power increase. The temperature evolution of the affected cell and surrounding cells was monitored to assess the effectiveness of the cooling systems in the BESS.
For the immersion cooling BESS, the results show a controlled temperature rise. At 100 seconds, the maximum cell temperature reached 111.88°C, with a maximum temperature difference of 86.88°C and a peak rate of temperature increase of 0.24°C/s. Similarly, at 165 seconds, the maximum temperature was 99.65°C, with a difference of 74.65°C and a rate of 0.61°C/s. In contrast, the liquid cold plate cooling BESS exhibited significantly higher temperatures: at 100 seconds, the peak temperature was 219.63°C, with a difference of 194.63°C and a rate of 1.25°C/s; at 165 seconds, it was 188.01°C, with a difference of 163.01°C and the same rate. These results highlight the superior performance of immersion cooling in suppressing thermal runaway in the BESS.
The temperature distribution within the battery pack was visualized using contour plots. In the immersion cooling BESS, the high-temperature region was localized around the faulty cell, with minimal spread to adjacent cells. The maximum temperature remained below the flash point of the hydrocarbon oil (190°C), preventing combustion risks. Conversely, in the liquid cold plate BESS, the high-temperature area was more extensive, indicating poor heat dissipation and a higher risk of thermal propagation. The flow trajectory analysis further supports these findings: immersion cooling ensures uniform fluid flow around the cells, facilitating efficient heat removal, while liquid cold plate cooling has limited contact, leading to hotspots.
To quantify the heat transfer efficiency, we calculated the Nusselt number (Nu) for the cooling methods, which relates convective to conductive heat transfer:
$$ Nu = \frac{h L}{k} $$
where \( h \) is the convective heat transfer coefficient, \( L \) is the characteristic length, and \( k \) is the thermal conductivity. For immersion cooling, the Nu values were higher due to direct fluid-cell contact, enhancing convection. The Reynolds number (Re) for the flow was computed as:
$$ Re = \frac{\rho u L}{\mu} $$
where \( \mu \) is the dynamic viscosity. With a flow velocity of 2 m/s and characteristic length based on cell dimensions, the Re indicated turbulent flow, which improves mixing and heat transfer in the BESS.
Table 3 compares key performance metrics between the two cooling methods for the BESS under thermal runaway conditions.
| Metric | Immersion Cooling | Liquid Cold Plate Cooling |
|---|---|---|
| Maximum Temperature at 100 s | 111.88°C | 219.63°C |
| Maximum Temperature at 165 s | 99.65°C | 188.01°C |
| Peak Temperature Rate | 0.24–0.61°C/s | 1.25°C/s |
| Temperature Uniformity | High | Low |
| Risk of Thermal Propagation | Low | High |
The simulation results demonstrate that immersion cooling effectively mitigates thermal runaway in the BESS by maintaining lower temperatures and reducing the rate of heat accumulation. This is crucial for naval platforms, where space constraints and safety requirements demand reliable thermal management. The co-simulation approach accurately captures the dynamic interactions between the battery pack and cooling system, providing insights for optimizing BESS designs.
Conclusion
In this study, we developed a comprehensive simulation framework to analyze thermal runaway in a lithium battery energy storage system (BESS) for naval platforms. By integrating Fluent and Simulink models, we evaluated the performance of immersion cooling and liquid cold plate cooling under real ship operating conditions. The results clearly show that immersion cooling offers superior thermal management, with lower peak temperatures, slower temperature rise rates, and better heat distribution compared to liquid cold plate cooling. This makes it a promising solution for enhancing the safety and reliability of the BESS in high-demand maritime environments.
However, this study has limitations, such as focusing on a single battery module rather than a full-scale BESS, and not accounting for factors like fluid combustion under extreme temperatures or ship motion effects. Future work will address these aspects by scaling up the simulation to entire battery packs, incorporating more complex abuse scenarios, and evaluating the impact of dynamic ship movements on cooling performance. Overall, this research provides valuable insights for designing robust thermal management systems in naval BESS applications, contributing to improved operational safety and efficiency.