Amidst the accelerated global transition towards renewable energy, lithium-ion battery energy storage systems (BESS) have emerged as a critical pillar for providing flexibility and stability to power grids, owing to their high energy density, rapid response capabilities, and modular deployment advantages. However, in practical engineering applications, the actual energy efficiency of energy storage battery cabinets often falls short of theoretical expectations, posing a significant bottleneck to their economic viability and large-scale adoption. Studies indicate that lifecycle efficiency losses for battery energy storage systems can reach 10%–20%, with factors such as battery charge/discharge voltage limits, thermal management strategies, and ambient temperature being particularly impactful.
This study focuses on a commercially deployed 1500V outdoor liquid-cooled battery energy storage system cabinet. Through systematic experimentation under simulated operational conditions, it investigates the synergistic effects of key operational parameters on system energy efficiency. The primary objective is to develop a dynamic, parameter-calibrated optimization strategy to enhance the performance and reliability of large-scale battery energy storage systems.
1. Experimental Methodology
The test subject was a 372.736 kWh battery cabinet employing a 1P416S configuration with 280Ah Lithium Iron Phosphate (LFP) cells. The nominal cell voltage was 3.2V. The main test equipment included a Power Conversion System (PCS), a walk-in environmental chamber, a dry-type transformer, a soft-start cabinet, and a liquid chiller unit.
1.1 Initial Charge/Discharge Energy Test Procedure
The battery energy storage system cabinet was placed inside the environmental chamber and connected to the test circuit. The procedure for measuring initial energy efficiency was as follows:
- Set the chamber to the target ambient temperature (Ta) and soak the cabinet for 5 hours.
- Charge at a constant power of Prc (186.368 kW) until any cell reaches the upper cut-off voltage (U1). Rest for 10 minutes.
- Discharge at a constant power of Prd (186.368 kW) until any cell reaches the lower cut-off voltage (U2). Rest for 10 minutes.
- Repeat steps 2 and 3 to complete a full charge-discharge cycle from a stable state. The energy from the second charge (Echarge) and discharge (Edischarge) was recorded.
- The energy efficiency (η) was calculated using the DC-side power measurements, excluding auxiliary power consumption:
$$η = \frac{E_{discharge}}{E_{charge}} \times 100\%$$
1.2 DC Internal Resistance Test Procedure
Following the initial energy test, the DC internal resistance of the battery energy storage system was measured at 50% State of Charge (SOC):
- Charge/discharge the cabinet to 50% SOC.
- After a rest period, record the open-circuit voltage (UOCV).
- Apply a constant current pulse (Ipulse = 280A) for 10 seconds and record the voltage at the 10th second (Upulse).
- The DC internal resistance (RDC) is calculated as:
$$R_{DC} = \frac{|U_{pulse} – U_{OCV}|}{I_{pulse}}$$
Separate measurements were taken for charge and discharge pulses.
1.3 Thermal Management Strategy
The liquid cooling system operated based on predefined logic tied to battery temperature readings (Tmax, Tavg, Tmin). The strategy for the heating mode, which is critical for low-temperature operation, involved setting different water outlet temperatures (T1, T2, T3) based on Tmin thresholds, as detailed in Table 1.
| Mode | Activation Logic | Deactivation Logic | Command to Chiller |
|---|---|---|---|
| Heating | Tmin < 14°C | Tmin ≥ 21°C | Heat, outlet temp = T1 |
| Heating | 14°C ≤ Tmin < 17°C | Tmin ≥ 21°C | Heat, outlet temp = T2 |
| Heating | 17°C ≤ Tmin < 20°C | Tmin ≥ 21°C | Heat, outlet temp = T3 |
| Cooling | Tmax ≥ 25°C & Tavg ≥ 24°C | Tmax < 22°C & Tavg < 21°C | Cool, outlet temp = 30°C |
| Self-circulation | Meet deactivation condition & ΔT ≥ 5°C | – | Pump only |
| Standby | None of the above conditions met | – | Off |
2. Results and Discussion
2.1 Impact of Ambient Temperature on Energy Efficiency
Tests were conducted with the cooling system off across a wide ambient temperature (Ta) range from 5°C to 60°C, with a voltage window of 2.8V–3.6V. The results, depicted graphically, revealed a strong correlation between ambient temperature and the energy efficiency of the battery energy storage system.
In the range of 5°C to 50°C, the energy efficiency increased monotonically with temperature, albeit at a diminishing rate. This is attributed to enhanced electrochemical reaction kinetics and reduced ionic/charge transfer resistance at elevated temperatures, leading to lower polarization losses. However, beyond 50°C, a reversal was observed, with efficiency declining as temperature increased further. This decline is linked to accelerated parasitic side reactions, including solid-electrolyte interphase (SEI) decomposition and electrolyte degradation, which increase irreversible capacity loss and internal resistance.
The relationship between ambient temperature (Ta in °C, denoted as X) and energy efficiency (η in %, denoted as Y) for the tested battery energy storage system cabinet was empirically fitted to a 6th-order polynomial:
$$Y = 0.86129 + 0.00779X – 2.91487 \times 10^{-4}X^2 + 8.03825 \times 10^{-6}X^3 – 1.63113 \times 10^{-7}X^4 + 2.12115 \times 10^{-9}X^5 – 1.25571 \times 10^{-11}X^6$$
with a coefficient of determination R² = 0.99606. This model captures the non-linear peak efficiency occurring around 50°C for this specific cell chemistry.
Concurrently, the DC internal resistance measurements showed a complementary trend. Resistance decreased from 5°C to 50°C but began to increase above 50°C, confirming the detrimental effects of excessive heat on cell health and ohmic losses within the battery energy storage system.
2.2 Impact of Voltage Window on Energy Efficiency
The charge/discharge voltage window is a primary control parameter for any battery energy storage system. Tests were performed at 25°C with different lower cut-off voltages (U2) while keeping the upper limit constant at 3.6V. The key performance metrics are summarized in Table 2.
| Voltage Window | Discharge Energy (kWh) | Energy Efficiency (%) |
|---|---|---|
| 2.85V – 3.6V | 372.8 | 95.85 |
| 2.80V – 3.6V | 373.64 | 95.24 |
| 2.70V – 3.6V | 375.76 | 94.31 |
| 2.60V – 3.6V | 376.4 | 93.36 |
A clear trade-off is observed. Widening the voltage window (by lowering U2) increases the usable discharge energy but at the cost of reduced round-trip efficiency. The efficiency drop becomes more pronounced below approximately 2.8V. This is due to increased mass transfer polarization at low voltages, where Li⁺ intercalation into the FePO₄ cathode becomes increasingly difficult due to rising Li⁺ concentration and inter-ionic repulsion within the solid phase.
For the design target of η ≥ 95% and Edischarge ≥ 372.736 kWh, the 2.8V–3.6V window presents the optimal balance, maximizing energy throughput while maintaining high efficiency for the battery energy storage system.
2.3 Impact of Thermal Management Strategy on Energy Efficiency
The heating mode setpoint (T1) during low-temperature operation significantly influences the initial cell temperature at the start of charge/discharge cycles. Five different thermal management schemes were tested, varying T1 from 24°C to 32°C. The performance data is consolidated in Table 3.
| Scheme | Heating Outlet T1 (°C) | Charge Energy (kWh) | Discharge Energy (kWh) | Energy Efficiency (%) | Max. Cell Temp. (°C) |
|---|---|---|---|---|---|
| 1 | 32 | 390.72 | 372.35 | 95.30 | 42.5 |
| 2 | 30 | 390.41 | 372.04 | 95.29 | 39.8 |
| 3 | 28 | 389.89 | 371.52 | 95.29 | 38.2 |
| 4 | 26 | 389.44 | 370.89 | 95.24 | 36.9 |
| 5 | 24 | 389.08 | 370.47 | 95.23 | 35.7 |
The results demonstrate that a higher initial warming temperature (T1) improves the charge/discharge energy and efficiency, consistent with the Arrhenius law which governs reaction rates. Pre-heating reduces the cell’s internal resistance at the cycle’s beginning, enhancing power capability and reducing losses.
However, this benefit must be weighed against long-term battery health. As shown, a T1 of 32°C led to a maximum cell temperature exceeding 40°C during operation. It is well-established that while LFP cell cycle life generally improves from 0°C to 40°C, prolonged operation above 40°C accelerates degradation mechanisms like SEI growth and active material loss. Therefore, while Scheme 1 (T1=32°C) offered marginally better initial efficiency, Scheme 2 (T1=30°C) achieved nearly identical efficiency (95.29%) while maintaining the maximum cell temperature safely below the 40°C threshold, thereby optimizing the trade-off between immediate performance and long-term durability of the battery energy storage system.
3. Conclusion
This study systematically investigated the influence of ambient temperature, operational voltage window, and thermal management strategy on the energy efficiency of a commercial liquid-cooled battery energy storage system cabinet. The key findings are:
- Ambient Temperature: Energy efficiency exhibits a non-linear relationship with temperature, increasing up to an optimum point (~50°C for this cell chemistry) before declining. The derived polynomial model provides a tool for forecasting efficiency under varying climatic conditions. For reliable operation, the ambient temperature for a battery energy storage system should be managed to not exceed this chemistry-specific peak efficiency temperature.
- Voltage Window: A critical trade-off exists between usable energy and efficiency. For the tested LFP battery energy storage system, a voltage window of 2.8V to 3.6V was identified as the optimal compromise, meeting design targets for both discharge energy (≥372.736 kWh) and round-trip efficiency (≥95%).
- Thermal Management: Active heating during low-temperature operation is essential for maintaining efficiency. The heating setpoint must be carefully calibrated. A setting of T1 = 30°C was found to be optimal, effectively warming the cells to reduce initial resistance and achieve high efficiency (95.29%) while preventing the maximum cell temperature from exceeding 40°C during operation, thus safeguarding long-term cycle life.
In summary, a multi-parameter协同 strategy for a battery energy storage system, employing an ambient temperature limit of ≤50°C, a voltage window of 2.8V–3.6V, and a heating mode setpoint T1 = 30°C, can simultaneously optimize energy efficiency, energy throughput, and battery longevity. This integrated “thermal-electrical” optimization framework provides actionable insights for the design and operation of large-scale battery energy storage systems, enhancing their economic value and supporting the integration of high-penetration renewable energy.