Seongjun Hong, Yongmin Kwon, Dongju Shin, Jangseop Park, Namwoo Kang
{"title":"DeepJEB:基于三维深度学习的合成喷气发动机支架数据集","authors":"Seongjun Hong, Yongmin Kwon, Dongju Shin, Jangseop Park, Namwoo Kang","doi":"arxiv-2406.09047","DOIUrl":null,"url":null,"abstract":"Recent advancements in artificial intelligence (AI) have significantly\ninfluenced various fields, including mechanical engineering. Nonetheless, the\ndevelopment of high-quality, diverse datasets for structural analysis still\nneeds to be improved. Although traditional datasets, such as simulated jet\nengine bracket dataset, are useful, they are constrained by a small number of\nsamples, which must be improved for developing robust data-driven surrogate\nmodels. This study presents the DeepJEB dataset, which has been created using\ndeep generative models and automated engineering simulation pipelines, to\novercome these challenges. Moreover, this study provides comprehensive 3D\ngeometries and their corresponding structural analysis data. Key experiments validated the effectiveness of the DeepJEB dataset,\ndemonstrating significant improvements in the prediction accuracy and\nreliability of surrogate models trained on this data. The enhanced dataset\nshowed a broader design space and better generalization capabilities than\ntraditional datasets. These findings highlight the potential of DeepJEB as a\nbenchmark dataset for developing reliable surrogate models in structural\nengineering. The DeepJEB dataset supports advanced modeling techniques, such as\ngraph neural networks (GNNs) and high-dimensional convolutional networks\n(CNNs), leveraging node-level field data for precise predictions. This dataset\nis set to drive innovation in engineering design applications, enabling more\naccurate and efficient structural performance predictions. The DeepJEB dataset\nis publicly accessible at: https://www.narnia.ai/dataset","PeriodicalId":501570,"journal":{"name":"arXiv - CS - Computational Geometry","volume":"20 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-06-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"DeepJEB: 3D Deep Learning-based Synthetic Jet Engine Bracket Dataset\",\"authors\":\"Seongjun Hong, Yongmin Kwon, Dongju Shin, Jangseop Park, Namwoo Kang\",\"doi\":\"arxiv-2406.09047\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Recent advancements in artificial intelligence (AI) have significantly\\ninfluenced various fields, including mechanical engineering. Nonetheless, the\\ndevelopment of high-quality, diverse datasets for structural analysis still\\nneeds to be improved. Although traditional datasets, such as simulated jet\\nengine bracket dataset, are useful, they are constrained by a small number of\\nsamples, which must be improved for developing robust data-driven surrogate\\nmodels. This study presents the DeepJEB dataset, which has been created using\\ndeep generative models and automated engineering simulation pipelines, to\\novercome these challenges. Moreover, this study provides comprehensive 3D\\ngeometries and their corresponding structural analysis data. Key experiments validated the effectiveness of the DeepJEB dataset,\\ndemonstrating significant improvements in the prediction accuracy and\\nreliability of surrogate models trained on this data. The enhanced dataset\\nshowed a broader design space and better generalization capabilities than\\ntraditional datasets. These findings highlight the potential of DeepJEB as a\\nbenchmark dataset for developing reliable surrogate models in structural\\nengineering. The DeepJEB dataset supports advanced modeling techniques, such as\\ngraph neural networks (GNNs) and high-dimensional convolutional networks\\n(CNNs), leveraging node-level field data for precise predictions. This dataset\\nis set to drive innovation in engineering design applications, enabling more\\naccurate and efficient structural performance predictions. The DeepJEB dataset\\nis publicly accessible at: https://www.narnia.ai/dataset\",\"PeriodicalId\":501570,\"journal\":{\"name\":\"arXiv - CS - Computational Geometry\",\"volume\":\"20 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-06-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Computational Geometry\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2406.09047\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Computational Geometry","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2406.09047","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
DeepJEB: 3D Deep Learning-based Synthetic Jet Engine Bracket Dataset
Recent advancements in artificial intelligence (AI) have significantly
influenced various fields, including mechanical engineering. Nonetheless, the
development of high-quality, diverse datasets for structural analysis still
needs to be improved. Although traditional datasets, such as simulated jet
engine bracket dataset, are useful, they are constrained by a small number of
samples, which must be improved for developing robust data-driven surrogate
models. This study presents the DeepJEB dataset, which has been created using
deep generative models and automated engineering simulation pipelines, to
overcome these challenges. Moreover, this study provides comprehensive 3D
geometries and their corresponding structural analysis data. Key experiments validated the effectiveness of the DeepJEB dataset,
demonstrating significant improvements in the prediction accuracy and
reliability of surrogate models trained on this data. The enhanced dataset
showed a broader design space and better generalization capabilities than
traditional datasets. These findings highlight the potential of DeepJEB as a
benchmark dataset for developing reliable surrogate models in structural
engineering. The DeepJEB dataset supports advanced modeling techniques, such as
graph neural networks (GNNs) and high-dimensional convolutional networks
(CNNs), leveraging node-level field data for precise predictions. This dataset
is set to drive innovation in engineering design applications, enabling more
accurate and efficient structural performance predictions. The DeepJEB dataset
is publicly accessible at: https://www.narnia.ai/dataset