A new study published in AI in Civil Engineering reports that an artificial neural network can synthesize experimental damage signals for inaccessible structural locations, using numerical simulations as raw material, with prediction accuracy of around 99 percent on a laboratory aluminum plate. The research, led by Anoop Sharma, Neetika Saha, and Pijush Topdar, addresses a gap in acoustic emission testing, which detects damage such as cracks, corrosion, and fatigue via stress waves. Laboratory signals are often produced with the pencil lead break, or Hsu-Nielsen, method, but this requires physical access to the tested point. The team modeled the structure numerically using the finite element method in the commercial software ABAQUS to obtain mechanical displacement-versus-time responses, and measured real electrical voltage-versus-time signals experimentally with an acoustic emission sensor, then trained a neural network to convert one into the other. The specimen was an isotropic aluminum plate measuring 1000 millimeters by 600 millimeters with a thickness of 1 millimeter, divided into four quadrants, each subdivided into square grids of 10 millimeters by 10 millimeters, giving 299 nodes per quadrant; the team exploited quadra symmetry and worked with Quadrant I. Pencil leads of 0.5 millimeter diameter were broken against the plate surface at an inclination of 45 degrees at each node.