Prediction of sound fields after propagation through sound barriers by CNN and DCNN algorithms

• Published in: The Journal of the Acoustical Society of America 2019-10, Vol.146 (4), p.2803-2803
• Main Authors: Kim, Jonghwan, Jung, Gwanghoon, Park, Junhong
• Format: Article
• Language: English
• ISSN: 0001-4966; 1520-8524
• DOI: 10.1121/1.5136711

Sound fields are affected by the size of the space and the surrounding environments. The ray tracing method analyze the sound propagation through geometric constraints. For prediction of sound fields in a real-time to investigate the influence of surrounding environments, an effective calculation method is required. During the prediction, it is necessary to calculate the sound field distribution according to the layout of equipment or sound reflecting devices to minimize noise in working spaces. This procedure is required for an optimal arrangement of barriers according to noise sources. To improve this repetitive work, it is devised to develop a machine learning model that predicts the change in the sound field according to the position of sound barrier through CNN and DCNN algorithms. Training data were generated by using a commercial tool. The proposed approach recognizes the characteristics of sound fields at spaces after acoustic barriers.