NISSAN_TECHNICAL_REVIEW_89 (2023)

91/111

Technical Awards：2022 JSAE Award The Outstanding Technical Paper Award - Surrogate Model Development for Prediction of Car Aerodynamics Using Machine LearningNISSAN TECHNICAL REVIEW No.88 (2022)ReferencesSourceCFD results.3 The error (MAPE) in the CD value estimated by the proposed model was 1.4%.4 The proposed model is a useful surrogate model that can replace CFD because the fl ow fi elds and CD values for various shape designs can be estimated within a signifi cantly short period of time. Using the proposed model, it is expected that the resources required for aerodynamic analyses can be reduced considerably.(1) K. Akasaka, et al.: Simultaneous Estimation of Aerodynamic and Thermal Performances Using CFD, Proceedings of SAE Seminar, No.96-05, pp.11–14 (2005).(2) M. Arai, et al.: Development of the Aerodynamics of the New Nissan Murano, SAE Technical Paper, 2015-01-1542 (2015).(3) N. Umetani, et al.: Learning Three-Dimensional Flow for Interactive Aerodynamic Design,ACM Trans. Graph., Vol. 37, No. 4, Article 89 (2018).(4) A. Guo, et al: Convolutional Neural Networks for Steady Flow Approximation, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp.481–490 (2016).(5) Y. LeCun, et al.: Gradient-based learning applied to document recognition, Proceedings of the IEEE, 86(11), pp.2278–2324 (1998).(6) K. Akasaka, et al.: Development of A Tool for Interactive Prediction of Car Drag Coeffi cient Using Machine Learning, at The 34th Annual Conference of the Japanese Society for Artifi cial Intelligence, 2O6-GS-13-02 (2020).(7) K. He, et al: Deep residual learning for image recognition, Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 770-778 (2016).(8) Ioffe, et al.: Batch normalization: Accelerating deep network training by reducing internal covariate shift, arXiv preprint arXiv:1502.03167 (2015).(9) Ulyanov, et al. :Instance normalization: The missing ingredient for fast stylization, arXiv preprint arXiv:1607.08022 (2016).(10) D. Maturana, et al.: VoxNet: A 3D Convolutional Neural Network for real-time object recognition, IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Hamburg, pp. 922-928 (2015).(11) M. Lin, et al.: Network in network, arXiv:1312.4400 (2013).公益社団法人自動車技術会自動車技術会論文集Vol.52, No.3 文献番号：202142488699