The global imperative to achieve carbon peak and neutrality goals has catalyzed the rapid development of green and intelligent vessels. Within this transformation, lithium-ion battery systems, prized for their high energy density, power density, and fast-charging capabilities, have emerged as the preferred power source for propulsion and onboard auxiliary systems. However, the inherent electro-thermal coupling within lithium-ion cells presents a significant challenge. During charge and discharge cycles, heat is generated due to internal resistance, electrochemical polarization, and the entropy changes of the underlying reactions. This heat generation, if not managed effectively, can lead to accelerated degradation, reduced performance, and in extreme cases, thermal runaway—a critical safety hazard.
In practical marine applications, EV battery pack systems often operate in harsh and variable environmental conditions, exposing potential shortcomings in thermal management design. Inadequate heat dissipation, particularly under high ambient temperatures or high-load operations, can compromise the safety, reliability, and longevity of the EV battery pack, thereby diminishing the overall performance and economic viability of the vessel. To address this and deliver standardized, series-produced, and reliable marine EV battery pack products, it is essential to conduct focused research on thermal management strategies. This study centers on a marine-grade lithium iron phosphate (LFP) EV battery pack configured within a containerized mobile power supply. We propose and execute a series of temperature rise cycle tests based on a liquid-cooling system design. By systematically varying environmental temperature, coolant temperature, and current rate (C-rate), we evaluate the thermal performance of the EV battery pack. The findings provide a critical empirical basis for thermal management system design and validation, ultimately enhancing the reliability of marine EV battery pack products.
The core of the thermal challenge lies in the heat generation profile of the battery cell. The total heat generation rate ($\dot{Q}_{total}$) within a single cell can be modeled using Bernardi’s general energy balance equation:
$$
\dot{Q}_{total} = I(V_{ocv} – V_t) + I T \frac{\partial V_{ocv}}{\partial T}
$$
Where $I$ is the current (positive for discharge), $V_{ocv}$ is the open-circuit voltage, $V_t$ is the terminal voltage, and $T$ is the absolute temperature. The first term, $I(V_{ocv} – V_t)$, represents the irreversible Joule heating due to internal resistance and polarization. The second term, $I T (\partial V_{ocv}/\partial T)$, represents the reversible heat associated with the entropy change of the cell reaction. For LFP chemistry, the entropy coefficient ($\partial V_{ocv}/\partial T$) is relatively small, making irreversible heat the dominant source, especially at high C-rates.
For an entire EV battery pack comprising $N_s$ cells in series and $N_p$ cells in parallel, the total pack-level heat generation ($Q_{pack}$) under a given pack current $I_{pack}$ can be approximated as:
$$
Q_{pack} \approx N_s N_p \cdot I_{cell}^2 R_{int,cell}
$$
where $I_{cell} = I_{pack} / N_p$ and $R_{int,cell}$ is the internal resistance of an individual cell. This simple model underscores that heat generation scales quadratically with current, making thermal management crucial for high-power operations.
Test Platform and Methodology
The test object is a marine containerized mobile power unit housing an LFP EV battery pack with a nominal capacity of 280 Ah and a nominal voltage of 57.6 V. The thermal management system employs a two-stage cooling strategy. The primary heat transfer path is: Cell → Immersion Dielectric Fluid → Steel Inner Shell of Water Jacket → Circulating Coolant Water. An aluminum extruded cold plate is positioned at the bottom of the battery box. This design philosophy prioritizes safety and engineering simplicity. Using a high-boiling-point dielectric immersion fluid coupled with a separate, isolated water cooling circuit mitigates the risks associated with direct water cooling (e.g., leakage, control of high-temperature/pressure water) while effectively transferring heat.
Extensive simulation was conducted to optimize the flow channel design within the cold plate. The goal was to ensure higher flow velocity in the central channels compared to the side channels and to guarantee active flow in every channel, thereby promoting uniform cooling across the entire EV battery pack base.
The test setup was designed to replicate the extreme environmental conditions specified by marine classification societies. The core test equipment included:
- A high-performance climate chamber capable of achieving temperatures from -20°C to +55°C.
- A 250 kW battery cycler for controlled charge and discharge profiles.
- A recirculating chiller unit to supply temperature-controlled coolant to the EV battery pack‘s cold plate.
- A comprehensive Battery Management System (BMS) for real-time monitoring of cell voltages, temperatures, state-of-charge (SOC), and system alarms.
The EV battery pack was placed inside the climate chamber. High-voltage power cables and communication (CAN) lines were routed through sealed ports to connect the pack to the external cycler and monitoring上位机. The cold plate was connected to the chiller via insulated hoses, with coolant flow rate maintained at approximately 5 L/min. The BMS parameters, including temperature alarm thresholds (e.g., Level 1: 45°C, Level 2: 50°C, Level 3: 55°C), were pre-configured to ensure operational safety.
Experimental Protocol: Evaluating Extreme Conditions
The experimental matrix was designed to investigate the combined influence of extreme ambient temperature, coolant temperature, and operational current rate on the thermal behavior of the EV battery pack. Proper selection of coolant temperature is critical; it must ensure safe operation within the cell’s optimal window while avoiding excessive energy consumption by the chiller itself.
The following test conditions were executed:
- High-Temperature Extreme: Ambient temperature set to +55°C. Coolant inlet temperature set to +18°C. Tests performed at discharge/charge rates of 0.5C (140A) and 0.8C (224A).
- Low-Temperature Extreme: Ambient temperature set to -20°C. Coolant inlet temperature set to +30°C (to prevent overcooling and assist in maintaining minimal cell temperature). Tests performed at discharge/charge rates of 0.5C and 0.8C.
For each test, the climate chamber and chiller were stabilized at the target temperatures first. The EV battery pack was then subjected to a full charge-discharge cycle (typically from 100% SOC to 0% SOC and back, or vice-versa, within voltage limits) while the BMS recorded all cell temperatures. Key metrics analyzed were:
- Maximum individual cell temperature ($T_{max}$).
- Minimum individual cell temperature ($T_{min}$).
- Maximum temperature differential within the pack ($\Delta T_{max} = T_{max} – T_{min}$).
- Delivered/absorbed capacity during the cycle.
Results and Analysis
1. Performance under High-Temperature Extreme (+55°C)
The thermal performance of the EV battery pack under the high-temperature stress condition is summarized in the table below. The data clearly shows the impact of increased C-rate on heat generation.
| Test Condition | Max. Cell Temp., $T_{max}$ (°C) | Min. Cell Temp., $T_{min}$ (°C) | Max. Temp. Delta, $\Delta T_{max}$ (°C) | Capacity (Ah) | Thermal Assessment |
|---|---|---|---|---|---|
| Ambient: +55°C Coolant: +18°C C-rate: 0.5C |
42 | 35 | 7 | ~278 | Excellent. $T_{max}$ well below 45°C alarm. |
| Ambient: +55°C Coolant: +18°C C-rate: 0.8C |
48 | 40 | 8 | ~276 | Good. $T_{max}$ below 50°C alarm, managing high-power heat load. |
At a 0.5C rate, the cooling system maintained the hottest cell at 42°C, which is 13°C above the ambient but safely within the primary alarm threshold. The temperature uniformity was acceptable, with a maximum delta of 7°C. When the power demand increased to 0.8C, the heat generation rose significantly, as predicted by the $I^2R$ relationship. Consequently, $T_{max}$ increased to 48°C. Crucially, this temperature remains below the critical 50°C secondary alarm, demonstrating that the liquid cooling system, with 18°C coolant, has sufficient capacity to handle the substantial thermal load even in a +55°C environment. The $\Delta T_{max}$ of 8°C indicates that the optimized flow channel design provided reasonably uniform cooling, preventing the development of severe local hot spots within the EV battery pack.
2. Performance under Low-Temperature Extreme (-20°C)
The low-temperature tests presented a different challenge: preventing cell temperatures from dropping too low (which increases internal resistance and can cause lithium plating) while still managing the heat generated during operation. Using a warmer coolant (+30°C) helped maintain a higher minimum cell temperature.
| Test Condition | Max. Cell Temp., $T_{max}$ (°C) | Min. Cell Temp., $T_{min}$ (°C) | Max. Temp. Delta, $\Delta T_{max}$ (°C) | Capacity (Ah) | Thermal Assessment |
|---|---|---|---|---|---|
| Ambient: -20°C Coolant: +30°C C-rate: 0.5C |
32 | 22 | 10 | ~277 | Good. $T_{min}$ > 20°C, ensuring safe operation. |
| Ambient: -20°C Coolant: +30°C C-rate: 0.8C |
41 | 29 | 12 | ~275 | Acceptable. $T_{max}$ controlled, $T_{min}$ maintained in operational range. |
The results are insightful. At 0.5C, the system successfully kept the coldest cell at 22°C, which is 42°C above the ambient temperature, ensuring the electrochemical kinetics were not severely hampered. The $\Delta T_{max}$ of 10°C is higher than in the high-temperature case, partly due to the larger temperature gradient between the pack core and the cold environment. At the higher 0.8C rate, the increased heat generation naturally raised both $T_{max}$ and $T_{min}$. The $T_{max}$ of 41°C is well-controlled, and the $T_{min}$ of 29°C remains in a favorable operational zone. The capacity delivered in all low-temperature tests showed a minimal decrease of less than 4 Ah compared to nominal, indicating no significant low-temperature capacity fade was induced during these operational cycles. This confirms the thermal management system’s ability to maintain the EV battery pack within a functional temperature window even in severe cold.
3. Thermal Model Correlation and System Efficiency
The experimental data can be used to back-calculate an effective overall heat transfer coefficient ($U$) for the EV battery pack cooling system. In a steady-state simplification, the heat removed by the coolant should equal the heat generated by the pack:
$$
\dot{Q}_{coolant} = \dot{m} c_p (T_{out} – T_{in}) \approx Q_{pack}
$$
Where $\dot{m}$ is the coolant mass flow rate, $c_p$ is the specific heat capacity of the coolant, and $T_{in}$ and $T_{out}$ are the inlet and outlet coolant temperatures. The heat transfer can also be expressed as:
$$
Q_{pack} = U A \Delta T_{lm}
$$
Where $A$ is the effective heat transfer area and $\Delta T_{lm}$ is the log-mean temperature difference between the average cell temperature and the coolant. Combining these, we can assess the cooling system’s effectiveness. For the high-temperature, 0.8C case, the significant temperature rise of the cells (up to 48°C) relative to the 18°C coolant indicates a substantial $\Delta T_{lm}$, driving efficient heat rejection. The fact that temperatures stabilized and did not exceed safety limits confirms that the product $U A$ is adequately sized for the worst-case heat load of this EV battery pack configuration.
Discussion on Design Implications
The successful thermal performance under extreme conditions validates several key design choices for the marine EV battery pack:
- Immersion with Indirect Liquid Cooling: This architecture proved robust. The dielectric fluid ensures good thermal contact with all cell surfaces, promoting heat spreading, while the separate water loop provides reliable, high-capacity heat rejection to the external chiller, avoiding single-point failure risks associated with complex direct cooling.
- Cold Plate Design: The flow-optimized cold plate was effective in establishing a relatively uniform base temperature, as evidenced by the controlled $\Delta T_{max}$ values (mostly under 10°C). This uniformity is critical for balancing cell aging and maximizing pack lifespan.
- Adaptive Coolant Temperature Setpoints: The test protocol highlights the importance of adaptive thermal management logic. Using a warmer coolant (+30°C) in a -20°C environment is essential to prevent excessive cooling, whereas a colder coolant (+18°C) is necessary to achieve sufficient heat flux in a +55°C environment. An intelligent BMS should dynamically adjust chiller setpoints based on ambient temperature and pack load.
- Safety Margin: In all extreme tests, the maximum cell temperature remained below the next-level BMS alarm threshold. This provides a crucial safety margin for transient overloads or temporary cooling system degradation, enhancing the inherent safety and reliability of the EV battery pack product.
Conclusion
This study presented a comprehensive experimental investigation into the extreme environmental temperature rise performance of a marine containerized LFP EV battery pack equipped with an immersion-coupled liquid cooling system. By subjecting the pack to rigorous testing at the classification society limits of -20°C and +55°C ambient temperature, and evaluating its response at different current rates (0.5C and 0.8C), we have rigorously validated the thermal management system’s design and the product’s operational reliability.
The key findings are:
- Under the high-temperature extreme (+55°C), the cooling system maintained cell temperatures within safe limits ($T_{max}$ < 48°C at 0.8C), demonstrating sufficient heat dissipation capacity.
- Under the low-temperature extreme (-20°C), with appropriately raised coolant temperature, the system prevented cell temperatures from falling into a critical low zone ($T_{min}$ > 22°C at 0.5C) while still managing operational heat, with negligible capacity impact.
- Temperature uniformity within the pack was effectively controlled, with maximum differentials typically below 12°C even under the most strenuous conditions, affirming the efficacy of the optimized cold plate flow design.
These results conclusively demonstrate that the implemented two-stage liquid cooling strategy is capable of ensuring the safe, reliable, and efficient operation of the EV battery pack across the full spectrum of mandated marine environmental conditions. This work provides a validated methodological framework and critical performance data to guide the design and qualification of future high-power, high-reliability marine EV battery pack systems, contributing directly to the advancement of green shipping technologies.