Details of the Researcher

PHOTO

Yosuke Yamanaka
Section
Graduate School of Engineering
Job title
Assistant Professor
Degree
  • Ph.D. in Engineering (Tohoku University)

e-Rad No.
71006286

Research History 4

  • 2026/04 - Present
    Tohoku University Graduate School of Engineering Department of Civil and Environmental Engineering Assistant Professor

  • 2024/04 - 2026/03
    Japan Society for the Promotion of Science Research fellow, PD

  • 2024/04 - 2026/03
    Nihon University College of Industrial Technology Department of Mechanical Engineering Research fellow

  • 2022/04 - 2024/03
    Japan Society for the Promotion of Science Reseach fellow, DC2

Education 1

  • Tohoku University Graduate School of Engineering Department of Civil and Environmental Engineering

    2021/04 - 2024/03

Professional Memberships 3

  • Japan Society of Civil Engineers

    2020 - Present

  • The Japan Society for Computational Engineering and Science

    2019 - Present

  • The Japan Society of Mechanical Engineers

    2024 -

Research Interests 4

  • Continuum Mechanics

  • Machine Learning

  • Multiscale analysis

  • Computational Homogenization

Research Areas 3

  • Informatics / Computational science /

  • Manufacturing technology (mechanical, electrical/electronic, chemical engineering) / Machine materials and mechanics /

  • Social infrastructure (civil Engineering, architecture, disaster prevention) / Structural and seismic engineering /

Awards 5

  1. The PhD Thesis Award

    2026/05 RBF-based surrogate models for computational homogenization of nonlinearcomposites

  2. 第30回計算工学講演会 若手優秀講演フェロー表彰

    2025/11 日本計算工学会 機械学習に基づく複合梁の弾塑性マルチスケール解析

  3. 第29回計算工学講演会 若手優秀講演フェロー表彰

    2024/11 日本計算工学会 非深層学習手法を用いた代理均質化モデルによる粘塑性複合材料のマルチスケール解析

  4. 第24回応用力学シンポジウム 講演賞

    2021/10 公益社団法人 土木学会(応用力学委員会) 熱硬化性樹脂の硬化と有限変形粘弾性の完全陰的熱・機械連成増分型変分法

  5. 工学研究科長賞

    2021/03 東北大学 大学院 工学研究科

Papers 8

  1. Surrogate Computational Homogenization of Viscoelastic Composites

    Yosuke Yamanaka, Norio Hirayama, Kenjiro Terada

    International Journal for Numerical Methods in Engineering 126 (6) 2025/03/18

    Publisher: Wiley

    DOI: 10.1002/nme.70008  

    ISSN: 0029-5981

    eISSN: 1097-0207

    More details Close

    ABSTRACT We establish a surrogate model for computational homogenization of composite materials consisting of multiple viscoelastic constituents. A surrogate macroscopic material is identified by performing interpolation using radial basis functions (RBFs) and cubic spline functions on a constitutive database generated by a series of microscopic analyses or, equivalently, numerical material tests (NMTs) on a unit cell to represent anisotropic stress relaxation behavior as well as its dependence on strain rate and temperature. After briefly reviewing the two‐scale boundary value problem derived based on homogenization theory, an RBF interpolation with a normalized kernel is formulated for a discrete dataset, and an optimization algorithm is applied to determine the three hyperparameters so as to achieve proper interpolation. Then, we formulate the surrogate homogenization model (SHM) using the interpolants to substitute for the macroscopic viscoelastic response. To show the specific procedure of the proposed surrogate modeling and demonstrate the performance of the created SHM, a representative numerical example is presented in the offline and online stages. In the offline stage, NMTs are carried out in the space of training data, including the temperature, to generate a dataset as long as the macroscopic stresses can be learned exhaustively, and then optimization is performed to determine the set of hyperparameters. Then, to validate the created SHM, its responses to unseen loading and temperature histories are compared with the corresponding NMT results. In the online stage, the created SHM is used to carry out a macroscopic analysis of a simple structure under specific loading and temperature conditions, and subsequently, the obtained macroscopic stresses are compared with those obtained from the localization analysis results. The results are also compared with those obtained from the single‐scale direct numerical analyses.

  2. Radial basis function-based surrogate computational homogenization for elastoplastic composites at finite strain

    Akari Nakamura, Yosuke Yamanaka, Reika Nomura, Shuji Moriguchi, Kenjiro Terada

    Computer Methods in Applied Mechanics and Engineering 436 117708-117708 2025/03

    Publisher: Elsevier BV

    DOI: 10.1016/j.cma.2024.117708  

    ISSN: 0045-7825

  3. Surrogate modeling for the homogenization of elastoplastic composites based on RBF interpolation

    Yosuke Yamanaka, Seishiro Matsubara, Norio Hirayama, Shuji Moriguchi, Kenjiro Terada

    Computer Methods in Applied Mechanics and Engineering 415 2023/10/01

    DOI: 10.1016/j.cma.2023.116282  

    ISSN: 0045-7825

  4. A method of fully implicit coupled analyses for thermoset resin subjected to cure based on variationally consistent formulation for finite thermo-viscoelasticity

    Yosuke Yamanaka, Seishiro Matsubara, Shuji Moriguchi, Kenjiro Terada

    International Journal of Solids and Structures 268 2023/04/15

    DOI: 10.1016/j.ijsolstr.2023.112161  

    ISSN: 0020-7683

  5. Surrogate Model-based Multiscale Analysis of Elastoplastic Composites with Nonperiodicity

    Akari Nakamura, Yosuke Yamanaka, Yuichi Shintaku, Norio Hirayama, Shuji Moriguchi, Kenjiro Terada

    Transactions of the Japan Society for Computational Engineering and Science 2023 2023

    DOI: 10.11421/jsces.2023.20230008  

    eISSN: 1347-8826

  6. Thermo-mechanical coupled incremental variational formulation for thermosetting resins subjected to curing process

    Yosuke Yamanaka, Seishiro Matsubara, Risa Saito, Shuji Moriguchi, Kenjiro Terada

    International Journal of Solids and Structures 216 30-42 2021/05/01

    DOI: 10.1016/j.ijsolstr.2021.01.014  

    ISSN: 0020-7683

  7. A decoupling scheme for two-scale finite thermoviscoelasticity with thermal and cure-induced deformations

    Risa Saito, Yosuke Yamanaka, Seishiro Matsubara, Tomonaga Okabe, Shuji Moriguchi, Kenjiro Terada

    International Journal for Numerical Methods in Engineering 122 (4) 1133-1166 2021/02/28

    DOI: 10.1002/nme.6575  

    ISSN: 0029-5981

    eISSN: 1097-0207

  8. Incremental variational formulation of time evolution of cure-degree along with viscoelasticity for thermosetting resins

    Yosuke Yamanaka, Seishiro Matsubara, Risa Saito, Shuji Moriguchi, Kenjiro Terada

    Transactions of the Japan Society for Computational Engineering and Science 2020 1 2020

    DOI: 10.11421/jsces.2020.20200011  

    eISSN: 1347-8826

Show all ︎Show first 5

Misc. 3

  1. 非深層学習の代理均質化モデルによる非線形複合材料の三次元有限変形マルチスケール解析—Three-Dimensional Multiscale Analysis Using RBF-Based Surrogate Homogenization Model at Finite Strain

    中村 明莉, 山中 耀介, 野村 怜佳, 森口 周二, 寺田 賢二郎

    計算工学講演会論文集 = Proceedings of the Conference on Computational Engineering and Science / 日本計算工学会 編 30 1075-1079 2025/06

    Publisher: 東京 : 日本計算工学会

  2. A surrogate computational homogenization for viscoplastic composites based on RBF interpolation

    山中耀介, 森口周二, 寺田賢二郎

    計算工学講演会論文集(CD-ROM) 29 2024

    ISSN: 1342-145X

  3. RBF-based surrogate model for computational homogenization of inelastic composites in finite deformation framework

    中村明莉, 山中耀介, 森口周二, 寺田賢二郎

    計算工学講演会論文集(CD-ROM) 29 2024

    ISSN: 1342-145X

Research Projects 2

  1. データ駆動型マルチスケール解析と構成材料の分布推定によるCFRPの設計最適化

    山中 耀介

    Offer Organization: 日本学術振興会

    System: 科学研究費助成事業

    Category: 特別研究員奨励費

    Institution: 日本大学

    2024/04/23 - 2027/03/31

  2. CFRPのプロダクトライフサイクル予測を可能にするマルチスケール計算科学の創生

    山中 耀介

    Offer Organization: 日本学術振興会

    System: 科学研究費助成事業

    Category: 特別研究員奨励費

    Institution: 東北大学

    2023/03/08 - 2024/03/31

    More details Close

    令和五年度は,複合材料の材料非線形性を考慮したデータ駆動型マルチスケール解析手法の開発を実施した. 複合材料の力学挙動を予測・評価するために開発された均質化法に基づくマルチスケール解析手法は,ミクロ構造の非均質性を反映して製品スケール(マクロスケール)の力学シミュレーションが可能である.しかし,従来の手法は計算コストの高さや定式化の難しさといった課題を抱えている.近年ではこれらの課題を克服するために,機械学習を取り入れたマルチスケール解析手法,すなわちデータ駆動型マルチスケール解析手法の研究が実施されている. 本研究の成果として,弾塑性,粘塑性,粘弾性といった,樹脂性複合材料にしばしば見られる材料非線形性を表現可能なマルチスケール解析手法の開発に成功した.これにより,本研究成果をより発展させ,実用化することでFRP製品の変形・強度特性評価が可能となることが期待される.また,これまでに開発した,樹脂の硬化の影響を考慮したFRP製品の熱・機械・化学連成シミュレーション手法と組み合わせることで,FRP製品の製造から使用状況までを一貫して解析することが可能である. 提案手法の特徴として,機械学習手法として非深層学習の一種であるRBF補間を使用した点が挙げられる.ニューラルネットに代表される深層学習と比べ,非深層学習モデルは学習プロセスが単純であり,また,得られたモデルの精度から教師データが十分であるか否かの検証も容易である.加えて,製品スケールの解析(マクロ解析)を実施する際に,従来使用されてきた有限要素法などのプログラムを侵食しないことも提案手法の特徴である.これらの特徴により,提案手法は汎用解析ソフトウェアにも容易に実装が可能であり,使用も比較的容易であるため,将来的には社会実装が期待される.