Thermal Runaway Propagation and Fire Safety Modeling of Large-Scale Marine Battery Rooms

Abstract

The transition to fully electric ships is an essential step toward reducing greenhouse gas emissions in the maritime industry. This change requires significant modifications to the operations and construction of ships. This thesis makes a contribution to the study of the thermal runaway (TR) behavior of lithium iron phosphate (LiFePO2 or LFP) battery systems. The objective is to gain an understanding of how fire, heat, and gas spread throughout a marine battery room during a TR event. To investigate the process by which a fire spreads from a single cell to an entire battery room, a thermal runaway simulation model was developed in Spreadsheet. The model is based on experimental heat release rate (HRR) and total heat release (THR) data from recent fire tests. It employs a hierarchical structure that extends from the cell to the module to the rack to the room. In addition, it incorporates flame propagation that is based on the direction of the flame, variable state-ofcharge levels ranging from 0% to 100%, and battery room sizes. For each scenario, the tool computes the peak HRR, the total energy release, combustion duration, and the structural heat load on the steel in the room. The analysis show that state of charge (SOC) is the strongest lever for safety: 100% SOC gives the fastest roof failure, 50% SOC extends survival by about 30-50%, and at 0% SOC many layouts do not reach roof melting at all. Rack spacing controls room density and creates a practical trade-off more spacing lowers heat concentration and increases melt time, but also increases steel mass and ship space. A moderate spacing band (≈0.5-0.9 m) offers the best balance for design, especially for 15-25 MWh rooms. The seawater flooding analysis further showed that early flooding can completely prevent roof failure, while late flooding has limited effect and may even increase explosion risks. The tool therefore gives designers a simple, validated way to compare layouts, operating modes and to set safe design limits for marine battery rooms. In doing so, it directly supports compliance with DNV’s 5 MWh threshold rule for single spaces, while offering quantified methods to justify larger 10-25 MWh installations where compensatory measures are required.