Assessment of Anisotropic Acoustic Properties in Additively Manufactured Materials: Experimental, Computational, and Deep Learning Approaches

Ivan P. Malashin, Vadim Tynchenko, Dmitry Martysyuk, Nikita A. Shchipakov, Nikolay V. Krysko, Maxim Degtyarev, Vladimir A. Nelyub, Andrei P. Gantimurov, Aleksey S. Borodulin, Andrey Galinovsky

Abstract: The influence of acoustic anisotropy on ultrasonic testing reliability poses a challenge in evaluating products from additive technologies (AT). This study investigates how elasticity constants of anisotropic materials affect defect signal amplitudes in AT products. Experimental measurements on AT samples were conducted to determine elasticity constants. Using Computational Modeling and Simulation Software (CIVA), simulations explored echo signal changes across ultrasound propagation directions. The parameters A 13 (the ratio between the velocities of ultrasonic transverse waves with vertical and horizontal polarizations at a 45-degree angle to the growth direction), A 3 (the ratio for waves at a 90-degree angle), and A g (the modulus of the difference between A 13 and A 3 ) were derived from wave velocity relationships and used to characterize acoustic anisotropy. Comparative analysis revealed a strong correlation (0.97) between the proposed anisotropy coefficient A g and the amplitude changes. Threshold values of A g were introduced to classify anisotropic materials based on observed amplitude changes in defect echo signals. In addition, a method leveraging deep learning to predict A g based on data from other anisotropy constants through genetic algorithm (GA)-optimized neural network (NN) architectures is proposed, offering an approach that can reduce the computational costs associated with calculating such constants.