Skip to content

2
2026_Article_RESILINK_ML
Commit 34b4262c by Joaquín Irazábal González
Options

Update manuscript with new sections and files for surrogate optimization methodology

- Updated the RESILINK surrogate optimization manuscript PDF to reflect recent changes. - Added a new synctex.gz file for improved LaTeX editing and referencing. - Revised the manuscript text to include conclusions in the workflow summary. - Introduced a new section on dataset generation and surrogate modeling, detailing the use of FEM models and damage indicators. - Removed redundant section header for clarity in the design of experiments description.
parent f94d3488
Showing with 12 additions and 11 deletions
Manuscript/Optimal-Design-layout/RESILINK_surrogate_optimization_OptimalDesignLayout.aux
...@@ -52,6 +52,7 @@ ...@@ -52,6 +52,7 @@
\citation{Yoshida2002,Jia2014} \citation{Yoshida2002,Jia2014}
\citation{American2002} \citation{American2002}
\citation{American2017} \citation{American2017}
\BKM@entry{id=6,open,dest={73656374696F6E2E34},srcline={211}}{5C3337365C3337375C303030445C303030615C303030745C303030615C303030735C303030655C303030745C3030305C3034305C303030675C303030655C3030306E5C303030655C303030725C303030615C303030745C303030695C3030306F5C3030306E5C3030305C3034305C303030615C3030306E5C303030645C3030305C3034305C303030735C303030755C303030725C303030725C3030306F5C303030675C303030615C303030745C303030655C3030305C3034305C3030306D5C3030306F5C303030645C303030655C3030306C5C3030306C5C303030695C3030306E5C30303067}
\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces Flow chart of the proposed adaptive surrogate-assisted optimization framework.}}{3}{figure.caption.1}\protected@file@percent } \@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces Flow chart of the proposed adaptive surrogate-assisted optimization framework.}}{3}{figure.caption.1}\protected@file@percent }
\providecommand*\caption@xref[2]{\@setref\relax\@undefined{#1}} \providecommand*\caption@xref[2]{\@setref\relax\@undefined{#1}}
\newlabel{fig:MethodologyFlowChart}{{1}{3}{Flow chart of the proposed adaptive surrogate-assisted optimization framework}{figure.caption.1}{}} \newlabel{fig:MethodologyFlowChart}{{1}{3}{Flow chart of the proposed adaptive surrogate-assisted optimization framework}{figure.caption.1}{}}
...@@ -61,7 +62,6 @@ ...@@ -61,7 +62,6 @@
\newlabel{sec:fem}{{3}{3}{}{section.3}{}} \newlabel{sec:fem}{{3}{3}{}{section.3}{}}
\citation{Rastellini2016} \citation{Rastellini2016}
\citation{Rice1969,Bao2004,Wierzbicki2005,Bai2008} \citation{Rice1969,Bao2004,Wierzbicki2005,Bai2008}
\BKM@entry{id=6,open,dest={73656374696F6E2E34},srcline={215}}{5C3337365C3337375C303030445C303030615C303030745C303030615C303030735C303030655C303030745C3030305C3034305C303030675C303030655C3030306E5C303030655C303030725C303030615C303030745C303030695C3030306F5C3030306E5C3030305C3034305C303030615C3030306E5C303030645C3030305C3034305C303030735C303030755C303030725C303030725C3030306F5C303030675C303030615C303030745C303030655C3030305C3034305C3030306D5C3030306F5C303030645C303030655C3030306C5C3030306C5C303030695C3030306E5C30303067}
\BKM@entry{id=7,open,dest={73756273656374696F6E2E342E31},srcline={217}}{5C3337365C3337375C303030445C303030655C303030735C303030695C303030675C3030306E5C3030305C3034305C3030306F5C303030665C3030305C3034305C303030655C303030785C303030705C303030655C303030725C303030695C3030306D5C303030655C3030306E5C303030745C30303073} \BKM@entry{id=7,open,dest={73756273656374696F6E2E342E31},srcline={217}}{5C3337365C3337375C303030445C303030655C303030735C303030695C303030675C3030306E5C3030305C3034305C3030306F5C303030665C3030305C3034305C303030655C303030785C303030705C303030655C303030725C303030695C3030306D5C303030655C3030306E5C303030745C30303073}
\citation{Joseph2008} \citation{Joseph2008}
\newlabel{fig:H30_B29}{{\caption@xref {fig:H30_B29}{ on input line 159}}{4}{}{figure.caption.3}{}} \newlabel{fig:H30_B29}{{\caption@xref {fig:H30_B29}{ on input line 159}}{4}{}{figure.caption.3}{}}
......
Manuscript/Optimal-Design-layout/RESILINK_surrogate_optimization_OptimalDesignLayout.fdb_latexmk
# Fdb version 4 # Fdb version 4
["bibtex RESILINK_surrogate_optimization_OptimalDesignLayout"] 1779262091.99058 "RESILINK_surrogate_optimization_OptimalDesignLayout.aux" "RESILINK_surrogate_optimization_OptimalDesignLayout.bbl" "RESILINK_surrogate_optimization_OptimalDesignLayout" 1779262094.94849 0 ["bibtex RESILINK_surrogate_optimization_OptimalDesignLayout"] 1779263430.67194 "RESILINK_surrogate_optimization_OptimalDesignLayout.aux" "RESILINK_surrogate_optimization_OptimalDesignLayout.bbl" "RESILINK_surrogate_optimization_OptimalDesignLayout" 1779263433.63411 0
"../wileyNJD-AMA.bib" 1778666628.59936 87574 acc8e4da0637e42f7c1b548d54a1e610 "" "../wileyNJD-AMA.bib" 1778666628.59936 87574 acc8e4da0637e42f7c1b548d54a1e610 ""
"./WileyNJD-AMA.bst" 1779200824.61362 19155 8fd474a7161c22bb6a795bd57510d2e1 "" "./WileyNJD-AMA.bst" 1779200824.61362 19155 8fd474a7161c22bb6a795bd57510d2e1 ""
"RESILINK_surrogate_optimization_OptimalDesignLayout.aux" 1779262094.58581 26187 353712e15f4b6bc2e20dc2fa9744b7f9 "pdflatex" "RESILINK_surrogate_optimization_OptimalDesignLayout.aux" 1779263433.29918 26187 af31c31eea3e0b065824db64a4408483 "pdflatex"
(generated) (generated)
"RESILINK_surrogate_optimization_OptimalDesignLayout.bbl" "RESILINK_surrogate_optimization_OptimalDesignLayout.bbl"
"RESILINK_surrogate_optimization_OptimalDesignLayout.blg" "RESILINK_surrogate_optimization_OptimalDesignLayout.blg"
(rewritten before read) (rewritten before read)
["pdflatex"] 1779262092.10124 "RESILINK_surrogate_optimization_OptimalDesignLayout.tex" "RESILINK_surrogate_optimization_OptimalDesignLayout.pdf" "RESILINK_surrogate_optimization_OptimalDesignLayout" 1779262094.94867 0 ["pdflatex"] 1779263430.78292 "/home/cimne/Articles_in_process/2026_Article_RESILINK_ML/Manuscript/Optimal-Design-layout/RESILINK_surrogate_optimization_OptimalDesignLayout.tex" "RESILINK_surrogate_optimization_OptimalDesignLayout.pdf" "RESILINK_surrogate_optimization_OptimalDesignLayout" 1779263433.63434 0
"../Figures/BayesianSearchCV.pdf" 1779190950.03781 230653 ee69df5dcd692912398f6eb14d4b39ea "" "../Figures/BayesianSearchCV.pdf" 1779190950.03781 230653 ee69df5dcd692912398f6eb14d4b39ea ""
"../Figures/Device.png" 1778506503.68273 158912 fe601a3780f3b901c7229f088234e566 "" "../Figures/Device.png" 1778506503.68273 158912 fe601a3780f3b901c7229f088234e566 ""
"../Figures/FEMsetup.png" 1778245828.41252 376415 13b71ef8c9b405795d5ed2ebd0411409 "" "../Figures/FEMsetup.png" 1778245828.41252 376415 13b71ef8c9b405795d5ed2ebd0411409 ""
...@@ -24,6 +24,7 @@ ...@@ -24,6 +24,7 @@
"../Figures/RBFOptimizationSurfaceEvolution/rbf_surface_evolution.png" 1778839162.67469 676554 eeb60734b5f2143a3de2966c25ad02d7 "" "../Figures/RBFOptimizationSurfaceEvolution/rbf_surface_evolution.png" 1778839162.67469 676554 eeb60734b5f2143a3de2966c25ad02d7 ""
"../Figures/plot_FEM_validation/FEM_validation_comparison.png" 1778592241.42322 267019 5d803722bd23ee8001bfa19db371b882 "" "../Figures/plot_FEM_validation/FEM_validation_comparison.png" 1778592241.42322 267019 5d803722bd23ee8001bfa19db371b882 ""
"/etc/texmf/web2c/texmf.cnf" 1727440723.11971 475 c0e671620eb5563b2130f56340a5fde8 "" "/etc/texmf/web2c/texmf.cnf" 1727440723.11971 475 c0e671620eb5563b2130f56340a5fde8 ""
"/home/cimne/Articles_in_process/2026_Article_RESILINK_ML/Manuscript/Optimal-Design-layout/RESILINK_surrogate_optimization_OptimalDesignLayout.tex" 1779263427.6471 72819 b11d026d779d6b89261eefb89ad7ae12 ""
"/usr/share/texlive/texmf-dist/fonts/enc/dvips/stix2-type1/stix2-mathcal.enc" 1610747736 3356 2818ff063f8b5d0d850e1203330b9fda "" "/usr/share/texlive/texmf-dist/fonts/enc/dvips/stix2-type1/stix2-mathcal.enc" 1610747736 3356 2818ff063f8b5d0d850e1203330b9fda ""
"/usr/share/texlive/texmf-dist/fonts/enc/dvips/stix2-type1/stix2-mathex.enc" 1610747736 3374 7b7c21d9493820ef4980dfa6fd22840f "" "/usr/share/texlive/texmf-dist/fonts/enc/dvips/stix2-type1/stix2-mathex.enc" 1610747736 3374 7b7c21d9493820ef4980dfa6fd22840f ""
"/usr/share/texlive/texmf-dist/fonts/enc/dvips/stix2-type1/stix2-mathit-bold.enc" 1610747736 3402 463aa77cccaed5276381cbb20340cbef "" "/usr/share/texlive/texmf-dist/fonts/enc/dvips/stix2-type1/stix2-mathit-bold.enc" 1610747736 3402 463aa77cccaed5276381cbb20340cbef ""
...@@ -272,9 +273,9 @@ ...@@ -272,9 +273,9 @@
"/var/lib/texmf/web2c/pdftex/pdflatex.fmt" 1727447091 8213230 bd25039be121841657767565aabcb4cb "" "/var/lib/texmf/web2c/pdftex/pdflatex.fmt" 1727447091 8213230 bd25039be121841657767565aabcb4cb ""
"LETTERSP.STY" 1738614188 12163 b6de1fa5e93e58dc68ba3c77a1af2e61 "" "LETTERSP.STY" 1738614188 12163 b6de1fa5e93e58dc68ba3c77a1af2e61 ""
"NJDnatbib.sty" 1738614190 46229 b23b408004050376b05f17f766e34127 "" "NJDnatbib.sty" 1738614190 46229 b23b408004050376b05f17f766e34127 ""
"RESILINK_surrogate_optimization_OptimalDesignLayout.aux" 1779262094.58581 26187 353712e15f4b6bc2e20dc2fa9744b7f9 "pdflatex" "RESILINK_surrogate_optimization_OptimalDesignLayout.aux" 1779263433.29918 26187 af31c31eea3e0b065824db64a4408483 "pdflatex"
"RESILINK_surrogate_optimization_OptimalDesignLayout.bbl" 1779262092.09802 14639 9f914222de3a75e88dc9d37885f68819 "bibtex RESILINK_surrogate_optimization_OptimalDesignLayout" "RESILINK_surrogate_optimization_OptimalDesignLayout.bbl" 1779263430.77815 14368 7a9811f84f9b1403a796b13caa33e7da "bibtex RESILINK_surrogate_optimization_OptimalDesignLayout"
"RESILINK_surrogate_optimization_OptimalDesignLayout.tex" 1779261954.86016 72803 9401d1f00113e7744e150e52c7895502 "" "RESILINK_surrogate_optimization_OptimalDesignLayout.tex" 1779263427.6471 72819 b11d026d779d6b89261eefb89ad7ae12 ""
"USG.cls" 1779262083.08888 95443 5ea137a0bee5c1a9c59eb5fe3a22630a "" "USG.cls" 1779262083.08888 95443 5ea137a0bee5c1a9c59eb5fe3a22630a ""
"algorithm.sty" 1738614188 3250 ec924db429a99ca4ed9c3769acfa5f18 "" "algorithm.sty" 1738614188 3250 ec924db429a99ca4ed9c3769acfa5f18 ""
"algorithmicx.sty" 1738614188 26750 ce139c05a983e19ddca355b43e29c395 "" "algorithmicx.sty" 1738614188 26750 ce139c05a983e19ddca355b43e29c395 ""
......
Manuscript/Optimal-Design-layout/RESILINK_surrogate_optimization_OptimalDesignLayout.fls
...@@ -3,7 +3,7 @@ INPUT /etc/texmf/web2c/texmf.cnf ...@@ -3,7 +3,7 @@ INPUT /etc/texmf/web2c/texmf.cnf
INPUT /usr/share/texmf/web2c/texmf.cnf INPUT /usr/share/texmf/web2c/texmf.cnf
INPUT /usr/share/texlive/texmf-dist/web2c/texmf.cnf INPUT /usr/share/texlive/texmf-dist/web2c/texmf.cnf
INPUT /var/lib/texmf/web2c/pdftex/pdflatex.fmt INPUT /var/lib/texmf/web2c/pdftex/pdflatex.fmt
INPUT RESILINK_surrogate_optimization_OptimalDesignLayout.tex INPUT /home/cimne/Articles_in_process/2026_Article_RESILINK_ML/Manuscript/Optimal-Design-layout/RESILINK_surrogate_optimization_OptimalDesignLayout.tex
OUTPUT RESILINK_surrogate_optimization_OptimalDesignLayout.log OUTPUT RESILINK_surrogate_optimization_OptimalDesignLayout.log
INPUT ./USG.cls INPUT ./USG.cls
INPUT USG.cls INPUT USG.cls
......
Manuscript/Optimal-Design-layout/RESILINK_surrogate_optimization_OptimalDesignLayout.tex
...@@ -125,7 +125,7 @@ All these works demonstrate the increasing interest in applying FEM-based and da ...@@ -125,7 +125,7 @@ All these works demonstrate the increasing interest in applying FEM-based and da
The present work addresses this gap through a damage-aware surrogate-assisted optimization methodology in which the objective is not only to maximize distortion or energy dissipation, but also to balance dissipative performance with damage indicators derived from FEM simulations. The proposed approach combines: (i) experimentally calibrated nonlinear FEM models used as numerical ground truth; (ii) supervised surrogate models trained to predict local damage and distortion indicators; (iii) a Differential Evolution (DE) optimizer; and (iv) an adaptive FEM validation and retraining loop. The present work addresses this gap through a damage-aware surrogate-assisted optimization methodology in which the objective is not only to maximize distortion or energy dissipation, but also to balance dissipative performance with damage indicators derived from FEM simulations. The proposed approach combines: (i) experimentally calibrated nonlinear FEM models used as numerical ground truth; (ii) supervised surrogate models trained to predict local damage and distortion indicators; (iii) a Differential Evolution (DE) optimizer; and (iv) an adaptive FEM validation and retraining loop.
Figure \ref{fig:MethodologyFlowChart} summarizes the proposed workflow. The different stages of the methodology, together with the surrogate modelling, optimization strategy, validation procedure and corresponding results, are described in the following sections. Figure \ref{fig:MethodologyFlowChart} summarizes the proposed workflow. The different stages of the methodology, together with the surrogate modelling, optimization strategy, validation procedure and corresponding results and conclusions, are described in the following sections.
\begin{figure*}[htbp] \begin{figure*}[htbp]
\centering \centering
...@@ -208,12 +208,12 @@ The model was calibrated and validated against cyclic experimental tests perform ...@@ -208,12 +208,12 @@ The model was calibrated and validated against cyclic experimental tests perform
\label{fig:FEM_validation_comparison} \label{fig:FEM_validation_comparison}
\end{figure*} \end{figure*}
\section{Dataset generation and surrogate modelling}\label{sec:surrogates}
Once validated, the FEM model is used to generate the datasets required for surrogate training and optimization. The optimization strategy relies on local damage indicators in the dissipative windows and surrounding frame, together with local distortion measures associated with the activation of the dissipative mechanism. Since these quantities are difficult to measure experimentally, the use of a high-fidelity FEM model provides valuable access to the internal state variables and local fields governing damage evolution and energy dissipation. Once validated, the FEM model is used to generate the datasets required for surrogate training and optimization. The optimization strategy relies on local damage indicators in the dissipative windows and surrounding frame, together with local distortion measures associated with the activation of the dissipative mechanism. Since these quantities are difficult to measure experimentally, the use of a high-fidelity FEM model provides valuable access to the internal state variables and local fields governing damage evolution and energy dissipation.
The damage indicator adopted in this work is the Triaxial Failure Damage Map (TFDMap) \cite{Rastellini2016}. This stress-triaxiality-based indicator evaluates the proximity of each material point to ductile failure by comparing its stress triaxiality and accumulated equivalent plastic strain with a reference failure envelope \cite{Rice1969,Bao2004,Wierzbicki2005,Bai2008}. In this study, the TFDMap is used as a post-processing damage-screening indicator, not as a constitutive fracture criterion. Its purpose is therefore not to explicitly predict crack initiation, but to compare geometrical configurations and ensure that optimized designs remain within acceptable damage levels. The damage indicator adopted in this work is the Triaxial Failure Damage Map (TFDMap) \cite{Rastellini2016}. This stress-triaxiality-based indicator evaluates the proximity of each material point to ductile failure by comparing its stress triaxiality and accumulated equivalent plastic strain with a reference failure envelope \cite{Rice1969,Bao2004,Wierzbicki2005,Bai2008}. In this study, the TFDMap is used as a post-processing damage-screening indicator, not as a constitutive fracture criterion. Its purpose is therefore not to explicitly predict crack initiation, but to compare geometrical configurations and ensure that optimized designs remain within acceptable damage levels.
\section{Dataset generation and surrogate modelling}\label{sec:surrogates}
\subsection{Design of experiments}\label{subsec:doe} \subsection{Design of experiments}\label{subsec:doe}
The FEM campaign is planned to cover the admissible parameter domain of each device family while ensuring a homogeneous exploration of the multidimensional space. The design variables are the window thicknesses $t_{w,i}$, whose combinations are generated through a Design of Experiments (DoE) strategy based on Latin Hypercube Sampling (LHS) optimized with the maximin criterion \cite{Joseph2008}. This approach provides a near-random yet space-filling distribution of samples, reducing clustering and improving the representation of the admissible domain. The FEM campaign is planned to cover the admissible parameter domain of each device family while ensuring a homogeneous exploration of the multidimensional space. The design variables are the window thicknesses $t_{w,i}$, whose combinations are generated through a Design of Experiments (DoE) strategy based on Latin Hypercube Sampling (LHS) optimized with the maximin criterion \cite{Joseph2008}. This approach provides a near-random yet space-filling distribution of samples, reducing clustering and improving the representation of the admissible domain.
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment