In recent years, the global push toward carbon neutrality has accelerated the adoption of renewable energy sources, leading to a significant increase in the deployment of battery energy storage systems (BESS). These systems, particularly containerized lithium battery energy storage systems, offer high power density, long lifespan, and reliability, making them ideal for grid stabilization and energy management. However, the rapid expansion of BESS installations has raised concerns about fire safety, as incidents involving thermal runaway and fires have resulted in substantial economic losses, injuries, and even fatalities. For instance, major accidents in Beijing and Australia underscore the critical need for effective fire protection strategies. As a researcher focused on enhancing the safety of energy infrastructure, I aim to evaluate the effectiveness of various fire protection facilities in containerized BESS through numerical simulation and analysis. This study employs PyroSim software to model fire scenarios triggered by lithium iron phosphate battery thermal runaway, examining parameters such as smoke concentration, visibility, and temperature under different fire suppression systems. The findings provide valuable insights for optimizing fire safety designs in electrochemical energy storage stations, ultimately contributing to the reliable operation of battery energy storage systems.
The fire hazards associated with containerized battery energy storage systems primarily stem from the inherent risks of lithium-ion batteries, electrical components, and operational factors. A BESS typically comprises battery packs, power cables, and control modules housed in a confined container, which can exacerbate fire spread if not properly managed. Key fire risk sources include:
- Equipment failures: Short circuits in battery cells or malfunctions in the battery management system can lead to uncontrolled energy release, initiating fires.
- Operational errors: Incorrect settings or mishandling during charging/discharging processes may cause overcurrent or overheating, triggering thermal runaway.
- External environmental factors: Inadequate ventilation, direct sunlight, or high ambient temperatures can elevate internal temperatures, increasing the likelihood of fire incidents.
Thermal runaway in batteries is a self-sustaining exothermic reaction that occurs due to internal short circuits, overcharging, mechanical damage, or excessive heat. The process involves the breakdown of battery components, releasing flammable gases and heat, which can propagate to adjacent cells. The mechanism can be described by the following equation representing the energy release during thermal runaway: $$ Q_{\text{release}} = \sum m_i \cdot c_{p,i} \cdot \Delta T + \Delta H_{\text{reaction}} $$ where $m_i$ is the mass of component $i$, $c_{p,i}$ is the specific heat capacity, $\Delta T$ is the temperature rise, and $\Delta H_{\text{reaction}}$ is the enthalpy change from chemical reactions. In a containerized BESS, the high energy density and compact layout amplify these risks, necessitating robust fire protection measures.
Fire protection facilities in a battery energy storage system are essential for early detection, suppression, and containment of fires. They typically include automatic alarm systems and various灭火 systems. The alarm system consists of smoke detectors, heat sensors, and controllers that trigger alerts and initiate countermeasures like power shutdown and ventilation closure.灭火 systems are categorized as follows:
- Water-based systems: These include open, closed, and pre-action sprinklers that cool the fire source and suppress flames through water mist. However, they may lead to incomplete combustion and increased toxic gas production.
- Gas-based systems: Utilizing agents like HFC-227ea (heptafluoropropane), these systems extinguish fires by reducing oxygen concentration and inhibiting chemical chain reactions. They are efficient and leave no residue, making them suitable for electrical fires.
- Combined gas-water systems: Integrating the benefits of both, these systems provide rapid fire suppression and cooling, minimizing re-ignition risks.
The effectiveness of these systems depends on factors such as deployment timing, agent distribution, and environmental conditions within the BESS container.
To assess the fire protection facilities, I developed a numerical model based on a real-world containerized BESS from a photovoltaic project. The system comprised lithium iron phosphate batteries with a total capacity of 2 MW/4 MWh per container. The container dimensions were 8.500 m in length, 2.800 m in width, and 2.896 m in height, and the simulation domain was set to 10 m × 4 m × 3 m to account for boundary effects. The battery model simplified individual cells into a homogeneous material with thermodynamic properties, as summarized in Table 1.
| Parameter | Electrolyte | Anode | Cathode | Separator |
|---|---|---|---|---|
| Density, $\rho$ (kg/m³) | 2600 | 8500 | 2700 | 492 |
| Specific Heat, $c_p$ (kJ/kg·K) | 1.1 | 0.385 | 0.9 | 1.978 |
| Thermal Conductivity, $\lambda$ (W/m·K) | 21 | 146 | 160 | 0.334 |
| Absorption Coefficient | 0.9 | 0.8 | 0.8 | 0.8 |
The fire load within the container was calculated to quantify the potential energy release. Using the formula for fire load $Q_i$: $$ Q_i = G_i \times H_i = C \times M \times H_i $$ where $C = 2240$ is the number of batteries, $M = 5.42$ kg is the mass per battery, and $H_i = 212.36$ kJ/kg is the heat release per unit mass, the total fire load was determined as $Q_i = 2492.6$ MJ. The fire load density $q$ was then computed as: $$ q = \frac{Q_i}{A} = \frac{2492.6}{23.8} \approx 104.73 \text{ MJ/m}^2 $$ where $A = 23.8$ m² is the area of the battery compartment. This high energy density highlights the severe fire risk in the BESS.
In the simulation, the fire source was modeled as a heat release rate (HRR) of 4.236 MW, based on the combustion of a hypothetical compound $C_{6.3}H_{7.1}O_{2.1}N$ derived from battery decomposition. Four scenarios were established: a control group with no fire protection, water sprinkler灭火, HFC-227ea gas灭火, and a combined gas-water system. Fire suppression systems were activated at an ambient temperature of 150°C, corresponding to the time when the control group first reached this threshold. Sensors for temperature, CO concentration, and visibility were placed at strategic points within the container to monitor changes over a 100-second period, ensuring computational feasibility.
The simulation results revealed significant variations in CO concentration, visibility, and temperature across the different fire protection scenarios. CO concentration, a key indicator of toxic gas production, was analyzed at a specific detection point. As shown in Figure 3 and summarized in Table 2, the control and water sprinkler groups exhibited steadily increasing CO levels due to sustained combustion, whereas the gas and combined systems showed rapid reduction post-activation. The water sprinkler system, while effective in cooling, often resulted in higher CO concentrations due to incomplete combustion, as the water mist suppressed flames but not gas generation. In contrast, the gas灭火 system diluted CO through HFC-227ea injection, lowering overall concentrations.
| Time (s) | No Protection | Water Sprinkler | Gas灭火 | Combined System |
|---|---|---|---|---|
| 20 | 150 | 140 | 130 | 135 |
| 50 | 450 | 500 | 200 | 250 |
| 80 | 600 | 650 | 150 | 180 |
| 100 | 700 | 750 | 100 | 120 |
Visibility, critical for evacuation and firefighting, was assessed through simulation data. The water sprinkler scenario initially reduced visibility due to smoke accumulation, with improvements after 40 seconds, but overall performance was inconsistent. The gas灭火 system restored visibility to approximately 2.5 m by 64 seconds, demonstrating effective smoke control. The combined system achieved similar results faster, by 34 seconds, indicating superior performance. The relationship between visibility $V$ and smoke density can be expressed as: $$ V = \frac{k}{\sigma} $$ where $k$ is a constant and $\sigma$ is the extinction coefficient, which increases with smoke concentration. This underscores the importance of rapid smoke suppression in a BESS environment.
Temperature analysis, as illustrated in Figure 6 and Table 3, highlighted the thermal dynamics under different fire protection measures. The control group reached peak temperatures around 600°C, with widespread heat distribution. The water sprinkler system reduced maximum temperatures to 440°C, containing the fire near the source but with periodic fluctuations. The gas灭火 system achieved the most rapid temperature drop, extinguishing the fire completely by disrupting combustion chains and reducing oxygen levels. The combined system further enhanced cooling, preventing re-ignition and maintaining lower temperatures. The energy balance during fire suppression can be described by: $$ \frac{dT}{dt} = \frac{\dot{Q}_{\text{in}} – \dot{Q}_{\text{out}}}{m c_p} $$ where $\dot{Q}_{\text{in}}$ is the heat input from the fire, $\dot{Q}_{\text{out}}$ is the heat removed by灭火 agents, $m$ is the mass of involved materials, and $c_p$ is the specific heat. This equation explains the temperature trends observed in the simulations.
| Scenario | Point 1 (Near Fire) | Point 2 (Center) | Point 3 (Far) |
|---|---|---|---|
| No Protection | 600 | 550 | 500 |
| Water Sprinkler | 440 | 400 | 350 |
| Gas灭火 | 300 | 250 | 200 |
| Combined System | 280 | 230 | 180 |
Additionally, I investigated the impact of mechanical smoke ventilation on fire control. Contrary to expectations, ventilation exacerbated fire intensity and smoke dispersion, leading to earlier temperature peaks and no improvement in visibility. This finding supports the practice of sealing the BESS container during fire incidents to enhance灭火 effectiveness. The air exchange rate $A_e$ in a ventilated system can be modeled as: $$ A_e = \frac{Q_v}{V_c} $$ where $Q_v$ is the volumetric flow rate and $V_c$ is the container volume. Higher $A_e$ values correlated with increased combustion rates, validating the need for a closed environment in battery energy storage system safety protocols.
In conclusion, this comprehensive analysis demonstrates that fire protection facilities play a pivotal role in mitigating risks in containerized battery energy storage systems. The water sprinkler system is effective for temperature control but may elevate CO levels, while the gas灭火 system excels in rapid fire suppression and visibility restoration. The combined gas-water system offers the best overall performance, integrating cooling and chemical inhibition. Furthermore, maintaining a sealed container during fires is crucial to prevent escalation. These insights, derived from numerical simulations, provide a foundation for designing safer BESS installations. Future work should explore additional variables, such as battery types, container geometries, and environmental conditions, to refine fire safety strategies further. As the adoption of battery energy storage systems continues to grow, optimizing these protection measures will be essential for sustainable energy infrastructure development.