A devastating explosion shook a grain storage silo, deforming steel walls and scattering debris hundreds of meters away. To determine the origin of the incident, forensic investigators turned to a complete digital workflow: LiDAR scanning with a Leica RTC360, fluid dynamics simulation in FLACS, and visualization in CloudCompare and Blender. The goal was to reconstruct the suspended dust cloud and locate the ignition source.
Forensic reconstruction with LiDAR and CFD simulation 🔥
The team deployed a Leica RTC360 scanner to capture the post-explosion geometry with millimeter precision. The resulting point clouds were processed in CloudCompare, where they were aligned and segmented to measure the plastic deformation of the steel and the direction of the fragments. These blast wave vectors, recorded in the metal structure, were exported as boundary conditions to the FLACS software. There, the ignition of a grain dust cloud was simulated, varying the position of the heat source. The simulation that matched the observed deformation patterns pointed to an overheated bearing in the bucket elevator as the ignition point.
Lessons from an invisible dust cloud 💡
The case demonstrates that combining 3D scanning and CFD simulation not only identifies the culprit but validates forensic hypotheses that would be impossible to confirm with traditional methods. By recreating the explosion in Blender for visualization, engineers were able to clearly communicate how a small spark in a bearing triggered a chain reaction inside the silo. This approach is becoming an indispensable tool for preventing industrial catastrophes, allowing ventilation and dust suppression systems to be redesigned based on real data.
How to integrate data from a post-explosion LiDAR scan with FLACS simulations to validate the ignition source hypothesis in a grain silo?
(PS: Simulating catastrophes is fun until the computer crashes and you are the catastrophe.)