期刊文献+

耦合聚类的数据驱动稀薄流非线性本构计算方法

Numerical method of data-driven rarefied nonlinear constitutive relations coupled with clustering
原文传递
导出
摘要 临近空间环境下的稀薄流动具有典型非平衡特征,现有数值模拟方法在面对多尺度共存的复杂流动现象时难以同时高效、准确描述。结合近些年兴起的机器学习理论与建模方法,基于数据驱动非线性本构关系(DNCR)为稀薄非平衡流动快速求解提供了一种全新的物理建模思路。为进一步提升DNCR方法的泛化性能、模型训练与计算效率,本文在DNCR方法中首次引入高斯混合模型(GMM)与稀疏主成分分析(SPCA)在回归模型训练预测前对训练集与预测集数据进行聚类预处理,建立了耦合聚类模型的数据驱动非线性本构计算方法。相关算例预测结果表明:在DNCR可解释性方面,GMM与SPCA方法可以不依靠人工经验通过流场数据提取出不同流动中的主导物理量,提升了DNCR方法的可解释性与鲁棒性;在计算效率方面,GMM与SPCA方法能在大量流场样本下对数据点进行精准聚类,排除冗余信息的干扰,提高了方法的并行计算效率,降低了回归模型的训练预测时间,同时有利于后期添加新样本数据时模型的更新与维护;在计算精度方面,GMM与SPCA方法在简单特征流动预测中能够在保持现有预示精度的前提下提高计算效率,并在面对复杂流动特征耦合预测场景时有效提高DNCR方法预测精度。 Owing to the rarefied nonequilibrium effect encountered in near space environment,the existing numerical methods are difficult to describe the multi-scale flow phenomena efficiently and accurately at the same time.By integrating the machine learning methods proposed in recent years,Data-driven Nonlinear Constitutive Relations(DNCR)presented a new data-driven modeling approach for solving the nonequilibrium problem.To further improve the generalization capability and model training efficiency of the DNCR method,Gaussian Mixture Model(GMM)and Sparse Principal Component Analysis(SPCA)for preprocessing the data of training set and prediction set are introduced to the DNCR method for the first time in this paper.The prediction results of relevant cases show that the dominant parameters in different flow fields are extracted by GMM and SPCA without relying on artificial experience,which could improve the interpretability and robustness of DNCR.On the other hand,GMM and SPCA can accurately cluster the data points under a large number of flow field samples to eliminate the interference of redundant information and reduce the training and prediction time of the regression model,which is crucial to the updating and maintenance of the model when adding new sample data in future.For prediction accuracy,the coupled DNCR could improve the computational efficiency without losing accuracy in simple flows,and could further elevate the precision significantly in complex flows coupled with different flow characteristics.
作者 尧少波 蒋励剑 赵文文 卢铮 吴昌聚 陈伟芳 YAO Shaobo;JIANG Lijian;ZHAO Wenwen;LU Zheng;WU Changju;CHEN Weifang(School of Aeronautics and Astronautics,Zhejiang University,Hangzhou 310027,China;Institute of Remote Sensing Satellite,China Academy of Space Technology,Beijing 100094,China)
出处 《航空学报》 EI CAS CSCD 北大核心 2022年第S02期43-56,共14页 Acta Aeronautica et Astronautica Sinica
基金 国家自然科学基金(U20B2007) 国家数值风洞项目(NNW2019ZT3-A08) 中央高校基本科研业务费(226-2022-00172)
关键词 高斯混合模型 稀疏主成分分析 数据驱动 树模型 稀薄非平衡流 Gaussian mixture model sparse principal component analysis data driven tree model rarefied nonequilibrium flow
  • 相关文献

参考文献6

二级参考文献66

  • 1张振亚,王进,程红梅,王煦法.基于余弦相似度的文本空间索引方法研究[J].计算机科学,2005,32(9):160-163. 被引量:54
  • 2李洁,高新波,焦李成.基于特征加权的模糊聚类新算法[J].电子学报,2006,34(1):89-92. 被引量:114
  • 3熊俊涛,乔志德,韩忠华.基于响应面法的跨声速机翼气动优化设计[J].航空学报,2006,27(3):399-402. 被引量:56
  • 4吕丽丽,张伟伟,叶正寅.高超声速再入体表面热流计算[J].应用力学学报,2006,23(2):259-262. 被引量:18
  • 5Guha S,Rastogi R,Shim K.CURE:An Efficient Clustering Algorithm for Large Databases[C].Seattle:Proceedings of the ACM SIGMOD Conference,1998.73-84.
  • 6Guha S,Rastogi R,Shim K.ROCK:A Robust Clustering Algorithm for Categorical Attributes[C].Sydney:Proceedings of the 15th ICDE,1999.512-521.
  • 7Karypis G,Han E-H,Kumar V.CHAMELEON:A Hierarchical Clustering Algorithm Using Dynamic Modeling[J].IEEE Computer,1999,32(8):68-75.
  • 8Ester M,Kriegel H-P,Sander J,et al.A Density-based Algorithm for Discovering Clusters in Large Spatial Databases with Noise[C].Portland:Proceedings of the 2nd ACM SIGKDD,1996.226-231.
  • 9Hinneburg A,Keim D.An Efficient Approach to Clustering Large Multimedia Databases with Noise[C].New York:Proceedings of the 4th ACM SIGKDD,1998.58-65.
  • 10Wang W,Yang J,Muntz R.STING:A Statistical Information Grid Approach to Spatial Data Mining[C].Athens:Proceedings of the 23rd Conference on VLDB,1997.186-195.

共引文献1341

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部