In recent years, there has been a growing demand for greater efficiency in product design, leading to active research into design support technologies that utilize AI. One such example is the surrogate model, which is expected to be applied in design processes due to its ability to reduce analysis time by replacing numerical computations in CAE (Computer-Aided Engineering) with AI. In this paper, we report the development of a surrogate model targeting plate surface vibrations in power distribution panels and electric motors. Using deep learning, the model infers multiple natural frequencies and mode shape images from design variables such as dimensional values and the number of reinforcements. We also evaluated the inference accuracy of the surrogate model and identified several challenges.
Wada et al. (Wed,) studied this question.