Improving Radiative Heat Flux Simulation in Fire Scenarios Using an Adaptive Resampling Backward Ray Tracing Method

Understanding thermal hazards and improving safety measures are two key aspects in virtual fire drill and evacuation simulation. Accurate simulation of radiative heat flux on the surface of human body in fire is critical for the two aspects. However, in these scenarios, the movement trajectories of firefighters and evacuees are uncertain due to fire conditions. Previous studies applied the backward Monte Carlo ray tracing method to calculate the radiant heat flux on moving targets with uncertain trajectories. However, this method is computationally expensive and lack of accuracy. To address these limitations, this study proposes an Adaptive Resampling Backward Ray Tracing Method (ARBRTM) for efficiently calculating surface radiative heat flux in fire. The ARBRTM integrates a uniform solid angle segmentation technique and an adaptive resampling method based on flame and hot gas edge detection, enabling high accuracy and computational efficiency. Validation against experimental data from pool fire tests demonstrates that ARBRTM achieves lower error rates compared to traditional backward Monte Carlo methods, even with fewer rays. The results highlight the method’s ability to capture significant variations in thermal radiation properties, improving simulation accuracy across varying distances and positions. This novel approach provides a robust and efficient tool for simulating radiative heat flux in complex fire scenarios, with potential applications in fire safety engineering and firefighter protection.