No pages match that search.
Engineering blog
How we model composites, crashes, flows and heat, and where machine learning earns its place in the loop. Written by the engineers who run the projects.
How to simulate resistance spot welding: contact resistance, nugget growth, expulsion, shunting and electrode wear in steel and aluminium, the weld lobe, and the nugget and heat-affected-zone data a crash or fatigue model needs from the weld.
How to simulate arc welding distortion in large structures: the moving heat source, filler activation, thermal-metallurgical-mechanical coupling, inherent-strain shortcuts, weld sequence, fixturing and pre-set, and surrogate-assisted sequence optimisation.
How to simulate blast protection of vehicle occupants, shelters and explosives stores: what the NATO test methodology and the US explosives safety regulation say, where models break, how validation works and where surrogates help.
How to simulate fire on process structures: pool and jet fire heat flux, the adiabatic surface temperature that links CFD to thermal-structural FEA, critical temperature, passive fire protection, and surrogate-assisted screening of fire scenarios.
How dust explosion venting is sized, where the correlations reach their limits, and what a validated CFD and FEA workflow adds: vent duct and panel effects, thrust on the supports, and surrogate-assisted screening of vent designs.
How vapour cloud explosion simulation works for onshore and offshore process areas: congestion and confinement, CFD pressure fields, exceedance curves and design accidental loads, mapping loads onto a structural FE model, and where surrogates help.
A blast resistant design procedure for buildings and process modules: load definition, SDOF versus finite element response, response limits and protection levels, progressive collapse, where AI helps, and the validation record.
How ALE blast simulation works for air blast on structures: JWL explosive and Eulerian air, Lagrangian coupling and its leakage, cell size scaled with charge mass, mapping 1D or 2D runs into 3D, a decision table against empirical loads, SPH and CFD, and surrogate-assisted sensitivity.
How Hybrid III, THOR and WorldSID finite element dummy models are validated, what rating programmes say about virtual evidence, where human body models fit beside dummies, and where machine learning helps and does not.
Have a model that will not converge?
Send the geometry, the physics and the decision it has to support. We answer with an approach, not a sales call.
