CNN-sindy based reduced order modeling of unsteady flow fields

Takaaki Murata, Kai Fukami, Koji Fukagata

研究成果: Conference contribution

抄録

We present a new framework of nonlinear reduced order model to extract low-dimensional modes and to predict their temporal evolutions. Autoencoder-type Convolutional Neural Network (CNN) which can learn nonlinearity of data is used to extract low-dimensional modes. For obtaining the temporal evolution of a mapped data by CNN, Sparse Identification of Nonlinear Dynamics (SINDy) is performed. The proposed method is applied to a circular cylinder wake at ReD = 100. The CNN trained using fluctuation components of velocity vector u, v shows better results than the snapshot Proper Orthogonal Decomposition in terms of the energy reconstruction rate. Although time-evolving flow fields reproduced by SINDy equation also show reasonable agreement with the reference direct numerical simulation, the errors of CNN and SINDy are accumulated through integral computation. The error of CNN can be reduced by devising a better network structure; however, the error of SINDy depends on the waveform of latent vector.

本文言語English
ホスト出版物のタイトルComputational Fluid Dynamics
出版社American Society of Mechanical Engineers (ASME)
ISBN(電子版)9780791859032
DOI
出版ステータスPublished - 2019
イベントASME-JSME-KSME 2019 8th Joint Fluids Engineering Conference, AJKFluids 2019 - San Francisco, United States
継続期間: 2019 7 282019 8 1

出版物シリーズ

名前ASME-JSME-KSME 2019 8th Joint Fluids Engineering Conference, AJKFluids 2019
2

Conference

ConferenceASME-JSME-KSME 2019 8th Joint Fluids Engineering Conference, AJKFluids 2019
CountryUnited States
CitySan Francisco
Period19/7/2819/8/1

ASJC Scopus subject areas

  • Fluid Flow and Transfer Processes

フィンガープリント 「CNN-sindy based reduced order modeling of unsteady flow fields」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル