Abstract
An analytical model configuration, in addition to air pressure analysis and post-processing, was conducted to measure the interior wind noise by changing the exterior vehicular design. Although wind noise can be calculated accurately through the current process, it requires three to five days for each design. In this study, a convolutional neural network (CNN), which is a class of deep neural networks designed for processing image data, was applied to predict the wind noise with vehicle design images from four different views. Feature maps were extracted from the CNN models trained with images of each view and concatenated to flow through a sequence of fully connected (FC) layers to predict the wind noise. Moreover, visualization of the significant vehicle parts for wind noise prediction was provided using a gradient-weighted class activation map (GradCAM). Finally, we compared the performance of various CNN-based models, such as ResNet, DenseNet, and EfficientNet, in addition to the architecture of the FC layers. The proposed method can predict the wind noise using vehicle images from different views with a root-mean-square error (RMSE) value of 0.206, substantially reducing the time and cost required for interior wind noise estimation.
Similar content being viewed by others
Abbreviations
- A-PLR:
-
a-pillar
- CAD:
-
computer-aided design
- CAE:
-
computer-aided engineering
- CNN:
-
convolutional neural network
- CFD:
-
computational fluid dynamics
- dB:
-
decibel
- FC:
-
fully connected layer
- GAP:
-
global average pooling
- GPU:
-
graphics processing unit
- Grad-CAM:
-
gradient-weighted class activation map
- Hybrid FE-SEA:
-
hybrid modeling method including finite element and statistical energy analysis
- ILSVRC:
-
ImageNet large scale visual recognition challenge
- LR:
-
learning rate
- NVH:
-
noise, vibration, and harshness
- MSE:
-
mean square error
- ReLU:
-
rectified linear unit
- RMSE:
-
root-mean-square error
- SPL:
-
sound pressure level
- XAI:
-
explainable artificial intelligence
References
Chen, S. M., Wang, D. F. and Zan, J. M. (2011). Interior noise prediction of the automobile based on hybrid FE-SEA method. Mathematical Problems in Engineering, 2011, 327170.
Cordioli, J. A., Trichês, M. and Gerges, S. N. (2004). Applications of the statistical energy analysis to vibro-acoustic modeling of vehicles. SAE Paper No. 2004-01-3352.
Cotoni, V., Shorter, P. and Langley, R. (2007). Numerical and experimental validation of a hybrid finite element-statistical energy analysis method. The J. Acoustic Society of America 122, 1, 259–270.
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold G., Gelly, S., Uszkreit. J. and Houlsby, N. (2020). An image is worth 16×16 words: Transformers for image recognition at scale. arXiv: 2010.11929.
Feng, Y., Feng, Y., You, H., Zhao, X. and Gao, Y. (2019). Meshnet: Mesh neural network for 3D shape representation. AAAI Conf. Artificial Intelligence, Honolulu, Hawaii, USA.
He, K., Zhang, X., Ren, S. and Sun, J. (2016). Deep residual learning for image recognition. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Las Vegas, Nevada, USA.
Huang, G., Liu, Z., Maaten, L. and Weinberger, K. Q. (2017). Densely connected convolutional networks. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, USA.
Jung, Y. W. and Kim, H. K. (2020). Prediction of nonlinear stiffness of automotive bushings by artificial neural network models trained by data from finite element analysis. Int. J. Automotive Technology 21, 6, 1539–1551.
Kadlowec, J., Wineman, A. S. and Hulbert, G. (2003). Elastomer bushing response: experiments and finite element modeling. Acta Mechanica 163, 1–2, 25–38.
Kingma, D. P. and Ba, J. (2014). Adam: A method for stochastic optimization. 3rd Int. Conf. Learning Representations (ICLR), San Diego, California, USA.
Krizhevsky, A., Sutskever, I. and Hinton, G. (2012). Imagenet classification with deep convolutional neural network. 26th Annual Conf. Neural Information Processing Systems (NIPS), Lake Tahoe, Nevada, USA.
Langley, R. S. (1989). A general derivation of the statistical energy analysis equations for coupled dynamic systems. J. Sound and Vibration 135, 3, 499–508.
LeCun, Y., Bengio, Y. and Hinto, G. (2015). Deep learning. Nature 521, 7553, 436–444.
LeCun, Y., Bottou, L., Bengio, Y. and Haffner, P. (1998). Gradient-based learning applied to document recognition. Proc. IEEE 86, 11, 2278–2324.
Lin, M., Chen, Q. and Yan, S. (2014). Network in network. 2nd Int. Conf. Learning Representations (ICLR), Banff, AB, Canada.
Nair, V. and Hinton, G. E. (2010). Rectified linear units improve restricted boltzmann machines. 27th Int. Conf. MachineLearning (ICML), Haifa, Israel.
Nefske, D. J., Wolf Jr, J. A. and Howell, L. J. (1982). Structural-acoustic finite element analysis of the automobile passenger compartment: A review of current practice. J. Sound and Vibration 80, 2, 247–266.
Putra, A., Munawir, A. and Farid, W. M. (2015). Corrected statistical energy analysis model for car interior noise. Advances in Mechanical Engineering 7, 1, 304283–304283.
Sarradj, E. (2004). Energy-based vibroacoustics: SEA and beyond. Proc. Joint Cong. CFA/DAGA, Strasbourg, France.
Selvaraju, R, R., Das, A., Vedantam R., Cogswell, M., Parikh, D. and Batra, D. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. IEEE Int. Conf. Computer Vision (ICCV), Venice, Italy.
Shin, Y., Lim, H. and Park, K. (2012). NVH CAE analysis for full vehicle model considering acustic package as well as body structure. Trans. Korean Society of Automotive Engineers Annual Conf., 1642–1643.
Shimobaba, T., Kakue, T. and Ito, T. (2018). Convolutional neural network-based regression for depth prediction in digital holograph. 27th Int. Symp. Industrial Electronics (ISIE), Cairns, QLD, Australia.
Tan, M. and Le, Q. (2019). Efficientnet: Rethinking model scaling for convolutional neural networks. Int. Conf. MachineLearning (ICML), Long Beach, CA, USA.
Umetani, N. (2017). Exploring generative 3D shapes using autoencoder networks. SIGGRAPH Asia (SA) Technical Briefs, Bangkok, Thailand.
Umetani, N. and Bickel, B. (2018). Learning three-dimensional flow for interactive aerodynamic design. ACM Trans. Graphics (TOG), 37, 4, 1–10.
Yoo, S., Lee, S., Kim, S., Hwang, K. H., Park, J. H. and Kang, N. (2020). Integrating deep learning into CAD/CAE system: Case study on road wheel design automation. arXiv: 2006.02138.
Acknowledgement
This research was supported by Brain Korea 21 FOUR. This research was also supported by a Korea TechnoComplex Foundation Grant (R2112651) and Korea University Grant (K2107521, K2107521). Please send any inquiry to the corresponding author, Sung Won Han (School of Industrial Management and Engineering, Korea University, Seoul, 02841, Republic of Korea. e]swhan@korea.ac.kr).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Park, H., Jung, H., Lee, M.S. et al. Interior Wind Noise Prediction and Visual Explanation System for Exterior Vehicle Design Using Combined Convolution Neural Networks. Int.J Automot. Technol. 23, 1013–1021 (2022). https://doi.org/10.1007/s12239-022-0088-9
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12239-022-0088-9