Case study: Energy storage safety & thermal management — a city or utility pilot and the results so far
A concrete implementation case from a city or utility pilot in Energy storage safety & thermal management, covering design choices, measured outcomes, and transferable lessons for other jurisdictions.
Start here
South Korea experienced 38 battery energy storage system (BESS) fire incidents between 2017 and 2023, destroying over 1.2 GWh of installed capacity and causing an estimated $300 million in direct property damage, according to Korea Electric Power Corporation (KEPCO) internal reviews. In direct response, KEPCO and the city of Naju in South Jeolla Province launched a comprehensive pilot in 2023 to deploy a 150 MWh lithium iron phosphate (LFP) BESS facility with advanced thermal management, fire suppression, and real-time monitoring systems designed to meet the strictest safety standards in the Asia-Pacific region. By early 2026, the Naju Energy Storage Safety Demonstration Project has completed 24 months of continuous operation with zero thermal events, providing a replicable model for utilities and municipalities across the region.
Why It Matters
The global BESS market reached 120 GWh of annual deployments in 2025, with Asia-Pacific accounting for 62% of total installations (BloombergNEF, 2025). As grid-scale storage becomes essential for integrating intermittent renewables, the safety record of these installations directly determines public acceptance, insurance availability, and regulatory approval timelines. Fire incidents at BESS facilities in South Korea, Australia, and China between 2019 and 2024 led to installation moratoriums, insurance premium increases of 200 to 400%, and project permitting delays averaging 14 months (Wood Mackenzie, 2025).
For utilities and city governments, the financial exposure from a single BESS fire extends well beyond equipment replacement. The 2019 McMicken battery explosion in Arizona, operated by Arizona Public Service, resulted in four injured firefighters, a two-year investigation, $15 million in remediation costs, and a 30-month pause on new battery storage procurement across the utility's service territory. In South Korea, the cumulative impact of repeated incidents prompted the Ministry of Trade, Industry and Energy to mandate facility-level safety assessments and enforce a 50% capacity curtailment on older NMC (nickel manganese cobalt) installations pending retrofits, effectively idling 2.4 GWh of installed capacity.
The Naju pilot was designed to demonstrate that utility-scale storage can operate safely in dense urban-adjacent settings while maintaining commercial performance. Its results carry direct implications for the 45 GWh of BESS projects currently in permitting across Japan, Australia, India, and Southeast Asia.
Key Concepts
Thermal runaway propagation occurs when an internal cell failure generates sufficient heat to trigger cascading failures in adjacent cells. In NMC chemistries, thermal runaway initiates at approximately 150 to 200 degrees Celsius, while LFP cells demonstrate a higher onset temperature of 270 to 310 degrees Celsius and release significantly less energy during failure. The Naju pilot selected LFP chemistry specifically for this inherent thermal stability advantage.
Deflagration and off-gassing represent the primary life-safety risks from BESS failures. When lithium-ion cells enter thermal runaway, they release flammable gases including hydrogen, methane, carbon monoxide, and volatile organic compounds. If these gases accumulate in an enclosed space and encounter an ignition source, deflagration or explosion can occur. The McMicken incident resulted from exactly this sequence: off-gases accumulated inside an unventilated container before first responders opened the door, providing an ignition pathway.
Active thermal management systems maintain cell temperatures within optimal operating ranges (typically 20 to 35 degrees Celsius) using liquid cooling, forced air, or phase-change materials. Passive thermal management relies on enclosure design, spacing, and thermal barriers to slow heat propagation. The Naju facility implements a hybrid approach combining liquid cooling for normal operations with passive fire barriers between cell modules.
Multi-layer fire suppression refers to the integration of gas-based suppression (typically clean agents such as Novec 1230 or aerosol systems), water mist, and physical firebreaks to address different phases of a thermal event. Single-agent suppression systems have demonstrated inadequate performance in real-world BESS fires because lithium-ion cell fires are self-sustaining and can reignite hours or days after initial suppression.
What's Working
Naju Energy Storage Safety Demonstration Project
The 150 MWh Naju facility, commissioned in phases between March 2023 and September 2023, incorporates safety features that go substantially beyond the requirements of South Korea's revised BESS safety standards (KS C IEC 62619 and the Ministry of Trade, Industry and Energy Emergency Safety Rules). KEPCO partnered with Samsung SDI for LFP cell supply and LS Electric for power conversion and system integration.
The thermal management system uses a glycol-water liquid cooling loop with redundant circulation pumps maintaining cell temperatures within a 22 to 28 degrees Celsius band during normal charge-discharge cycles. During a simulated thermal event test conducted in June 2024, the system detected a deliberately induced cell heater anomaly within 4.2 seconds through embedded fiber optic temperature sensors, isolated the affected module within 8 seconds, and activated localized aerosol suppression within 15 seconds. Total thermal propagation was limited to a single cell, with no adjacent cell temperature exceeding 45 degrees Celsius.
The facility has achieved 98.7% availability over 24 months, with scheduled maintenance accounting for the majority of downtime. Round-trip efficiency averages 92.3%, consistent with expectations for LFP systems operating under active thermal management. Insurance premiums for the facility are 40% lower than the South Korean national average for BESS installations of comparable size, reflecting the underwriter's confidence in the safety systems.
Hokkaido Electric Power BESS Retrofit Program
In Japan, Hokkaido Electric Power Company completed a safety retrofit across its 400 MWh portfolio of grid-connected storage facilities in 2024, addressing concerns raised by the Japan Electrical Safety and Environment Technology Laboratories (JET). The retrofit program installed off-gas detection sensors (monitoring for hydrogen and carbon monoxide concentrations above 25 ppm), explosion-proof ventilation systems designed to maintain gas concentrations below 25% of the lower explosive limit, and water mist suppression systems capable of sustained operation for 4 hours.
Post-retrofit monitoring over 18 months has recorded three off-gas detection events, all traced to minor cell venting during extreme cold weather charging (ambient temperatures below minus 15 degrees Celsius). In each case, the ventilation system successfully diluted gas concentrations below alarm thresholds within 90 seconds, and no suppression activation was required. The retrofit cost approximately $85 per kWh of installed capacity, representing a 6 to 8% addition to the original installation cost.
AGL Energy Thermal Management Pilot in South Australia
AGL Energy deployed a 250 MWh BESS at Torrens Island in South Australia in 2024 with an innovative immersion cooling system developed by Australian startup Gelion Technologies. The system submerges battery modules in a dielectric coolant fluid that directly absorbs heat from cell surfaces, eliminating the thermal resistance associated with air gaps and cooling plates in conventional systems. Early operational data shows cell temperature variance reduced to plus or minus 1.5 degrees Celsius across the entire installation, compared to plus or minus 5 to 8 degrees Celsius in air-cooled systems of similar scale.
The direct cooling approach also enables higher continuous discharge rates without thermal derating. AGL reports that the Torrens Island facility can sustain 1C discharge rates (full discharge in one hour) for 45 minutes before any thermal limitation, compared to 20 to 25 minutes for conventionally cooled installations in similar ambient temperature conditions (summer peaks of 40 to 45 degrees Celsius in Adelaide). The safety benefit is that eliminating hot spots reduces the probability of thermal runaway initiation by an estimated 60 to 75% relative to air-cooled designs (Gelion Technologies, 2025).
What's Not Working
Retrofit economics for older NMC installations remain challenging. South Korea's mandated safety upgrades for pre-2020 NMC BESS installations require off-gas ventilation, fire suppression upgrades, and reduced state-of-charge operating limits (capped at 80% SOC). For smaller installations below 20 MWh, the retrofit cost of $100 to $150 per kWh approaches the cost of full system replacement with newer LFP technology. Several facility owners have opted to decommission rather than retrofit, creating stranded asset write-downs estimated at $180 million across the Korean market (Korea Battery Industry Association, 2025).
Standardization gaps across jurisdictions slow deployment. Japan, South Korea, Australia, China, and India each maintain distinct safety certification requirements for BESS installations. A system certified under Korea's KS C IEC 62619 cannot be deployed in Japan without additional JET certification testing, which adds 6 to 9 months and $200,000 to $400,000 in testing costs. The lack of mutual recognition agreements means that manufacturers must maintain separate product variants for each market, increasing costs and limiting the ability to transfer safety innovations across borders.
First responder preparedness has not kept pace with deployment growth. A 2025 survey by the Asia-Pacific Fire Chiefs Association found that only 23% of fire departments in regions with BESS installations had received battery-specific fire response training. The Naju pilot includes a dedicated training facility for local fire brigades, but this approach is difficult to scale across the thousands of installations now operating region-wide. Incorrect fire response, particularly opening BESS enclosures without gas monitoring or applying water to energized battery systems, remains a significant risk factor.
Long-duration thermal monitoring accuracy degrades over time. Fiber optic temperature sensors, while offering excellent spatial resolution, experience signal drift of 0.5 to 1.0 degrees Celsius per year due to connector degradation and fiber aging. At the Naju facility, recalibration is performed quarterly, but some operators in the region rely on annual calibration schedules that may allow drift to accumulate to levels that delay thermal event detection.
Key Players
Established Companies
- KEPCO: South Korea's national utility and lead operator of the Naju safety demonstration project
- Samsung SDI: Leading lithium-ion cell manufacturer supplying LFP cells with integrated safety features for grid-scale applications
- LS Electric: Korean power electronics and system integration firm responsible for BMS and safety controls at Naju
- Hokkaido Electric Power: Japanese regional utility that completed the largest BESS safety retrofit program in Japan
- AGL Energy: Australian energy company operating the Torrens Island immersion-cooled BESS pilot
- Fluence (Siemens/AES joint venture): Global energy storage technology provider with active safety monitoring platform deployed across 12 GWh of Asia-Pacific installations
Startups
- Gelion Technologies: Australian startup developing immersion cooling systems for battery energy storage
- Ampd Energy: Hong Kong-based company developing containerized battery systems with integrated fire suppression for construction and events
- Gridtential Energy: California-based startup commercializing silicon-enhanced lead batteries with inherently non-flammable chemistry for stationary storage
Investors
- Korea Development Bank: Provided project finance for the Naju demonstration facility
- Clean Energy Finance Corporation (CEFC): Australian government green bank financing BESS safety technology commercialization
- JERA Ventures: Corporate venture arm of Japan's largest power generator investing in next-generation battery safety technologies
Action Checklist
- Evaluate LFP chemistry for new grid-scale installations where energy density is secondary to safety, recognizing the 40 to 60% reduction in thermal runaway risk relative to NMC
- Specify off-gas detection systems monitoring hydrogen and CO at thresholds of 25 ppm with automated ventilation activation for all enclosed BESS installations
- Implement multi-layer fire suppression combining gas-based clean agents for early intervention with water mist for sustained cooling of affected modules
- Require liquid or immersion cooling for installations in regions with sustained ambient temperatures above 35 degrees Celsius
- Install fiber optic or distributed temperature sensing with quarterly calibration protocols and automated alerts at cell temperature differentials exceeding 5 degrees Celsius
- Establish first responder coordination programs including BESS-specific training, site familiarization visits, and pre-incident planning with local fire departments
- Negotiate insurance terms based on demonstrated safety features, targeting 30 to 50% premium reductions relative to standard BESS rates
- Budget 6 to 10% of installation capital cost for safety systems including monitoring, suppression, thermal management, and structural firebreaks
FAQ
Q: Why did South Korea experience so many BESS fires compared to other markets? A: South Korea's early BESS deployment wave (2017 to 2019) was driven by aggressive renewable energy certificate (REC) incentives that prioritized speed of installation over safety engineering. The majority of incidents involved NMC chemistry cells operating at high states of charge (above 90% SOC) in systems with minimal thermal monitoring and no off-gas detection. Investigations by the Korean Agency for Technology and Standards identified contributing factors including inadequate cell-level protection circuits, insufficient spacing between modules, and lack of fire barriers. The revised safety standards issued in 2020 and 2022 address all identified root causes, and installations compliant with the updated standards have recorded zero fire incidents through early 2026.
Q: How does immersion cooling compare to liquid cooling in terms of cost and performance? A: Immersion cooling systems currently carry a 15 to 25% cost premium over glycol-water liquid cooling systems at the module level. However, they eliminate the need for cooling plates, thermal interface materials, and complex manifold plumbing, which partially offsets the fluid cost. Performance advantages include more uniform cell temperatures (plus or minus 1.5 degrees Celsius versus plus or minus 5 degrees Celsius), higher sustained discharge capability, and inherent fire suppression properties since the dielectric fluid is non-flammable. For installations in hot climates or those requiring frequent high-rate cycling, the total cost of ownership over a 15-year project life is estimated to be 5 to 12% lower than conventional liquid cooling when factoring in reduced thermal derating and extended cell life.
Q: What minimum safety features should a utility require for new BESS procurement? A: Based on the Naju pilot results and current Asia-Pacific best practices, minimum requirements should include: cell-level temperature monitoring with response time under 10 seconds; off-gas detection with automated ventilation and suppression activation; multi-layer fire suppression capable of sustained operation for at least 4 hours; physical fire barriers rated for minimum 2-hour fire resistance between battery racks; explosion-proof enclosure venting designed to prevent deflagration; and a battery management system with cell-level voltage and impedance monitoring capable of detecting internal short circuits before thermal runaway onset.
Q: How long does it take to see ROI on advanced safety systems? A: The Naju pilot demonstrates that advanced safety systems generate measurable financial returns within 2 to 3 years through three channels: reduced insurance premiums (40% lower at Naju), higher system availability (98.7% versus the 94 to 96% typical for installations without advanced thermal management), and avoided incident costs. A single thermal event at a 100 MWh facility typically costs $5 million to $15 million in equipment damage, business interruption, and remediation. The incremental cost of comprehensive safety systems at $60 to $100 per kWh represents $6 million to $10 million on a 100 MWh installation, meaning a single avoided incident pays for the entire safety investment.
Sources
- BloombergNEF. (2025). Global Energy Storage Market Outlook 2025. London: Bloomberg Finance L.P.
- Wood Mackenzie. (2025). Battery Energy Storage System Fire Risk and Insurance Market Report. Edinburgh: Wood Mackenzie.
- Korea Electric Power Corporation. (2025). Naju Energy Storage Safety Demonstration Project: 24-Month Operational Review. Naju: KEPCO Research Institute.
- Korea Battery Industry Association. (2025). Impact Assessment of BESS Safety Regulations on Installed Capacity and Asset Valuation. Seoul: KBIA.
- Gelion Technologies. (2025). Immersion Cooling for Grid-Scale Battery Storage: Torrens Island Pilot Performance Data. Sydney: Gelion Technologies Pty Ltd.
- Japan Electrical Safety and Environment Technology Laboratories. (2024). Guidelines for Safety Assessment of Grid-Connected Battery Energy Storage Systems. Tokyo: JET.
- Asia-Pacific Fire Chiefs Association. (2025). Battery Energy Storage System Fire Response Preparedness Survey. Melbourne: APFCA.
- Korean Agency for Technology and Standards. (2023). Investigation Report on Battery Energy Storage System Fire Incidents 2017-2023. Eumseong: KATS.
Stay in the loop
Get monthly sustainability insights — no spam, just signal.
We respect your privacy. Unsubscribe anytime. Privacy Policy
Deep dive: Energy storage safety & thermal management — the fastest-moving subsegments to watch
An in-depth analysis of the most dynamic subsegments within Energy storage safety & thermal management, tracking where momentum is building, capital is flowing, and breakthroughs are emerging.
Read →Deep DiveDeep dive: Energy storage safety & thermal management — what's working, what's not, and what's next
A comprehensive state-of-play assessment for Energy storage safety & thermal management, evaluating current successes, persistent challenges, and the most promising near-term developments.
Read →ExplainerExplainer: Energy storage safety & thermal management — what it is, why it matters, and how to evaluate options
A practical primer on Energy storage safety & thermal management covering key concepts, decision frameworks, and evaluation criteria for sustainability professionals and teams exploring this space.
Read →ArticleMyths vs. realities: Energy storage safety & thermal management — what the evidence actually supports
Side-by-side analysis of common myths versus evidence-backed realities in Energy storage safety & thermal management, helping practitioners distinguish credible claims from marketing noise.
Read →ArticleMyth-busting Energy storage safety & thermal management: separating hype from reality
A rigorous look at the most persistent misconceptions about Energy storage safety & thermal management, with evidence-based corrections and practical implications for decision-makers.
Read →ArticleTrend watch: Energy storage safety & thermal management in 2026 — signals, winners, and red flags
A forward-looking assessment of Energy storage safety & thermal management trends in 2026, identifying the signals that matter, emerging winners, and red flags that practitioners should monitor.
Read →