Home▸Acoustics▸Acoustic Emission Crack Localization: 2D Grid-Search Inversion

Acoustic Emission Crack Localization: 2D Grid-Search Inversion

Interactive 2D top-down acoustic-emission localizer: trigger a crack event on a plate, watch the attenuating elastic wave burst race outward to four piezoelectric sensors and solve for the source with a brute-force grid-search TDOA inversion — including amplitude-driven detection-latency bias and sensor dropout that a plain arrival-time model misses.

Acoustics2DAdvanced60 FPS📱 Mobile-adapted⇄ 3D version
2d-physics-ext-topic-64 ↗ Open standalone

Acoustic emission (AE) testing listens for the elastic wave bursts that damage itself releases — crack growth, fibre fracture, dislocation slip — and localizes the source purely from when each sensor hears it. This 2D companion to the 3D acoustic-emission simulator keeps the same real time-difference-of-arrival physics but computes it independently: the elastic burst's amplitude is modelled explicitly with 2D cylindrical spreading and material attenuation, so a weaker or more distant arrival crosses each sensor's fixed detection threshold late — a systematic bias distinct from random timing jitter — and can even drop out entirely if attenuation is high enough. Instead of the 3D version's closed-form linear solve, the source position here is recovered with a coarse-to-fine brute-force grid search over the candidate-position residual, the same class of nonlinear inversion real multi-sensor AE and seismic-location systems fall back on when more than the minimum number of sensors are in play. A dedicated arrival-time-vs-distance strip beneath the map — a genuinely 2D-native diagnostic with no 3D equivalent — makes the detection-latency bias visible as a graph, not just a localization-error number.

⚙ Under the hood

Trigger a crack event on a 2D structural plate and watch the attenuating elastic wave burst race outward to four piezoelectric sensors, with a brute-force grid-search TDOA inversion solving for the source live — including amplitude-driven detection-latency bias and sensor dropout that a plain arrival-time model misses.

acoustic emissionstructural health monitoringTDOAgrid-search inversionelastic wavesNDTsensor dropout

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)