Automatic Design Support System for Compact Acoustic Devices Using Deep Neural Network
Kai Hirai, Kai Nakamura, Yoshinobu Kajikawa, Kenta Iwai · 2018
An appropriate acoustic structure with a desired frequency response is rarely obtained through the acoustic equivalent circuit analysis in the case of compact acoustic devices. Thus, skilled acoustic engineers must design structures based on their know-how, and the time and cost are increased. Therefore, we propose an automatic design support system for compact acoustic devices introducing deep neural network. In the proposed system, the acoustic characteristics of candidate structures are analyzed by the acoustic FDTD method and an optimal candidate is obtained by learned deep neural network. We demonstrate the effectiveness of the proposed system through some comparisons between desired and designed frequency responses.