High performance ultrasonic field simulation on complex geometries

Abstract

Abstract. Ultrasonic field simulation is a key ingredient for the design of new testing methods as well as a crucial step for NDT inspection simulation. As presented in a previous paper [1], CEA-LIST has worked on the acceleration of these simulations focusing on simple geometries (planar interfaces, isotropic materials). In this context, significant accelerations were achieved on multicore processors and GPUs (Graphics Processing Units), bringing the execution time of realistic computations in the 0.1 s range.In this paper, we present recent works that aim at similar performances on a wider range of configurations. We adapted the physical model used by the CIVA platform to design and implement a new algorithm providing a fast ultrasonic field simulation that yields nearly interactive results for complex cases. The improvements over the CIVA pencil-tracing method include adaptive strategies for pencil subdivisions to achieve a good refinement of the sensor geometry while keeping a reasonable number of ray-tracing operations. Also, interpolation of the times of flight was used to avoid time consuming computations in the impulse response reconstruction stage.To achieve the best performance, our algorithm runs on multi-core superscalar CPUs and uses high performance specialized libraries such as Intel Embree for ray-tracing, Intel MKL for signal processing and Intel TBB for parallelization. We validated the simulation results by comparing them to the ones produced by CIVA on identical test configurations including mono-element and multiple-element transducers, homogeneous, meshed 3D CAD specimens, isotropic and anisotropic materials and wave paths that can involve several interactions with interfaces. We show performance results on complete simulations that achieve computation times in the 1s range.

New semi-analytical method for fast transcranial ultrasonic field simulation

“…The propagation of a pencil is illustrated on the left part of figure 3. The pencil method was later adapted by Chouh et al (2016) for performance purposes, to avoid computing the propagation and interaction matrices. Instead, the paraxial rays of every pencil are cast, using Intel Embree raytracing library, so as to reconstitute their contribution.…”

“…We adapted the pencil method to our path computation method (Angla et al 2023). The principle is very similar to the one proposed by Chouh et al (2016), but instead of casting rays to compute the paths, they are found by time of flight minimization. The advantage of computing the ultrasonic paths by time-of-flight minimization over using ray-tracing, is that each path is defined between a given source point, on the transducer, and the current computation point.…”

“…They yield: the minimum t min and maximum t max times of flight of the sample, the directions of the paraxial rays and the total Fresnel coefficient (Daniel et al 1996) T A of the central ray. The paraxial ray directions are used to compute the beam divergence DF (Chouh et al 2016). For an input amplitude A 0 , the impulse response associated with the surface sample dS is a rectangular signal of amplitude:…”