16KDD Convolutional Neural Networks for Steady Flow Approximation
本文简介 以前空气动力学的计算方法是传统CFD算法，但是它需要的计算开销大。本文提出了用CNN的方法，预测2D和3D的非均匀稳定层流(non-uniform steady laminar flow)，在误差允许范围内比CFD快很多（在GPU上快4个数量级）。这使得可以在空气动力学产品的设计初期快速迭代。训练数据可以源于CFD结果和实测结果。 本文提出的模型模型步骤： 用SDF(Signed Dis
相关文章
- 1. Understanding Convolutional Neural Networks for NLP
- 2. Convolutional Neural Networks for Sentence Classification
- 3. Learning Convolutional Neural Networks for Graphs
- 4. Convolutional Neural Networks for Speech Recognition///2014
- 5. 《Convolutional Neural Networks for Sentence Classification》
- 6. [转] Understanding Convolutional Neural Networks for NLP
- 7. Convolutional Neural Networks for Visual Recognition
- 8. Recurrent Convolutional Neural Networks for Text Classification
- 9. Bag of Tricks for Convolutional Neural Networks
- 10. CS231n Convolutional Neural Networks for Visual Recognition
- 更多相关文章...
- • XSL-FO flow 对象 - XSL-FO 教程
- • Swift for 循环 - Swift 教程
- • YAML 入门教程
- • Flink 数据传输及反压详解
相关标签/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. Understanding Convolutional Neural Networks for NLP
- 2. Convolutional Neural Networks for Sentence Classification
- 3. Learning Convolutional Neural Networks for Graphs
- 4. Convolutional Neural Networks for Speech Recognition///2014
- 5. 《Convolutional Neural Networks for Sentence Classification》
- 6. [转] Understanding Convolutional Neural Networks for NLP
- 7. Convolutional Neural Networks for Visual Recognition
- 8. Recurrent Convolutional Neural Networks for Text Classification
- 9. Bag of Tricks for Convolutional Neural Networks
- 10. CS231n Convolutional Neural Networks for Visual Recognition