The landscape of material science and manufacturing has transformed significantly with the advent of machine learning (ML). This toolkit of data driven methods accelerated the discovery and production of new materials by accurately predicting the complicated physical processes and mechanisms that are not fully described by existing material theories. Yet, with an array of intricate ML models at our disposal, the pressing question remains: Which ML algorithm is best suited for our needs? In this line of research, we aim to provide insights to strategically select appropriate models aligned with specific design challenges.

(Review paper)
Junhyeong Lee, Donggeun Park, Mingyu Lee, Hugon Lee, Kundo Park, Ikjin Lee, and Seunghwa Ryu* "Machine learning-based inverse design methods considering data characteristics and design space size in materials design and manufacturing: a review", Materials Horizons (2023). 

(AI-based inference: regression & classification)
image-to-number: Charles Yang+, Youngsoo Kim+, Seunghwa Ryu*, and Grace Gu*, "Using Convolutional Neural Networks to Predict Composite Properties beyond the Elastic Limit", MRS Communications 9, 609 (2019).
image-to-array: Charles Yang+, Youngsoo Kim+, Seunghwa Ryu*, and Grace Gu*, "Prediction of composite microstructure stress-strain curves using convolutional neural networks", Materials & Design 189, 108509 (2020)
image-to-image: Donggeun Park, Jiyoung Jung, Grace Gu, and Seunghwa Ryu*, "A generalizable and interpretable deep supervised neural network to predict strain field of composite in unseen design space", Materials & Design 223, 111192 (2022)
array-to-image: Jiyoung Yoon, Junhyeong Lee, Giyoung Kim, Seunghwa Ryu*, and Jinhyoung Park* "Deep Neural Network-Based Structural Health Monitoring Technique for Real-time Crack Detection and Localization using Strain Gauge Sensors", Scientific Reports 12, 20204 (2022).
image-to-high-res-image: Donggeun Park, Jiyoung Jung, and Seunghwa Ryu* "Double Generative Network (DGNet) Pipeline for Structure- Property Relation of Digital Composites" Composite Structures 319, 117131 (2023)
time-series-classification:  Dabin Kim, Ziyue Yang, Jaewon Cho, Donggeun Park, Dong Hwi Kim, Jinkee Lee, Seunghwa Ryu, Sang-Woo Kim, Miso Kim, "High-Performance Piezoelectric Yarns for Artificial Intelligence-enabled Wearable Sensing and Classification", EcoMat (2023).
image-classification:  Juhee Lee+ , Tae Gu Lee+, Ha Neul Lee, Hyoungsoo Kim, Yoo Kyung Kang, Seunghwa Ryu*, and Hyun Jung Chung* "Simple and multiplexed detection of nucleic acid targets based on fluorescent ring patterns and deep learning analysis", ACS Applied Materials and Interfaces (2023). 

image-to-movie:  Dongeun Park, Jaemin Lee, Hugon Lee, Grace X. Gu, and Seunghwa Ryu*  "Deep Generative Spatiotemporal Learning for Integrating Fracture Mechanics in Composite Materials: Inverse Design, Discovery, and Optimization", Materials Horizons (2024),. 

(Case1: data-driven DL-based optimization via interpolation)
anisotropic adhesive pillar shape:  Yongtae Kim†, Jinwook Yeo†, Kundo Park, Aymeric Destree, Zhao Qin, and Seunghwa Ryu* "Designing Directional Adhesive Pillar using Deep Learning-Based Optimization, 3D printing, and Testing",  Mechanics of Materials (2023). Invited.
adhesive pillar shape: Yongtae Kim, Charles Yang, Youngsoo Kim, Grace Gu, and Seunghwa Ryu*, "Designing Adhesive Pillar Shape with Deep Learning-Based Optimization", ACS Applied Materials & Interface 12, 24458 (2020).
wide bandgap phononic crystal: Wabi Demeke, Jiyoung Jung, Byungki Ryu, Wonju Jeon, and Seunghwa Ryu "Design of Aluminum Plate Phononic Crystals with Wide Bandgaps Using Deep Neural Network-based Free-form Optimization", Extreme Mechanics Letters (2023).

(Case2: data-driven DL-based optimization via extrapolation)
meta-structure for vibration energy harvesting: Sangryun Lee, Wonjae Choi, Jeongwon Park, Daesu Kim, Sahn Nahm, Wonju Jeon, Grace X. Gu, Miso Kim*, and Seunghwa Ryu* "Machine Learning-enabled Development of High Performance Gradient-index Phononic Crystals for Energy Focusing and Harvesting", Nano Energy 103, 107846 (2022)
grid composite pattern: Yongtae Kim+, Youngsoo Kim+, Charles Yang, Kundo Park, Grace X. Gu, and Seunghwa Ryu*, "Deep Learning Framework for Material Design Space Exploration using Active Transfer Learning and Data Augmentation", 
npj Computational Materials 7, 140 (2021)
grid composite pattern with multiple ingredients: Donggeun Park+, Minwoo Park+, and Seunghwa Ryu "Expanding Design Spaces in Digital Composite Materials: A Multi-Input Deep Learning Approach Enhanced by Transfer Learning and Multi-kernel Network", Advanced Theory and Simulations (2023)
grid composite pattern considering full crack propagation: Donggeun Park, Jaemin Lee, Kundo Park, and Seunghwa Ryu "HGNet: A Hierarchical Multi-Task Learning Approach for Accelerated Composite Material Design and Discovery",  Advanced Engineering Materials (2023).
thermoelectric composite stack: Wabi Demeke, Yongtae Kim, Jiyoung Jung, Jaywan Chung, Byungki Ryu* and Seunghwa Ryu* "Neural network-assisted optimization of segmented thermoelectric power generators using active learning based on a genetic optimization algorithm", Energy Reports 8, 6633 (2022)  
thermoelectric composite stack: Wabi Demeke, Byungki Ryu*, and Seunghwa Ryu* "Machine Learning-based Optimization of Segmented Thermoelectric Power Generators Using Temperature-Dependent Performance Properties",  Applied Energy (2023).

(Case3: data-driven optimization via multi-fidelity datasets)
homogenization dataset & FEM dataset: Jiyoung Jung, Yongtae Kim, Jinkyoo Park, and Seunghwa Ryu* "Transfer learning for enhancing the homogenization-theory-based prediction of elasto-plastic response of particle/fiber-reinforced composites",  Composite Structures 285, 115210 (2022)
injection molding process optimization for material alteration:  Hugon Lee+, Mingyu Lee+, Jiyoung Jung, Ikjin Lee*, and Seunghwa Ryu* "Multi-fidelity optimization of injection molding process parameters using near by prior information under insufficient data environment", Advanced Theory and Simulations (2024),

(Case4: data-driven optimization with small datasets)
composite pattern with single objective: Kundo Park, Youngsoo Kim, Minki Kim, Chihyeon Song, Jinkyoo Park*, and Seunghwa Ryu*, "Designing staggered platelet composite structure with Gaussian Process Regression based Bayesian optimization",  Composites Science and Technology 220, 109254 (2022)
composite pattern with multiple objectives: Kundo Park, Chihyeon Song, Jinkyoo Park, and Seunghwa Ryu "Multi-objective Bayesian optimization for the designing of nacre-inspired composites: optimizing and understanding the biomimetics through AI", Materials Horizons (2023).
auxetic pattern with multiple objectives: Hyunggwi Song†, Eunjeong Park†, Hongjae Kim, Chung-Il Park, Taek-Soo Kim, Yoon Young Kim, and Seunghwa Ryu*" Free-form optimization of pattern shape for improving mechanical characteristics of a concentric tube", Materials & Design (2023).
injection molding process with multiple objectives:Jiyoung Jung, Kundo Park, Byeonjin Cho, Jinkyoo Park, and Seunghwa Ryu* "Optimization of Injection Molding Process using Multi-objective Bayesian Optimization and Constrained Generative Inverse Design Networks",  Journal of Intelligent Manufacturing (2022)

Piezoelectric yarn optimization for extensibility and strength: Ziyue Yang, Kundo Park, Jisoo Nam, Jaewon Cho, Yong-Il Kim, Hyeonsoo Kim, Seunghwa Ryu, and Miso Kim* "Laminate-inspired high strength piezoelectric self-powered sensing yarns via multi-objective Bayesian optimization", Advanced Science (2024)

Our team specializes in designing theoretical and computational models for materials at various scales. We conduct atomistic simulations to understand the mechanical behavior of materials, using these insights as inputs for larger-scale continuum and composite modeling. Additionally, we develop multiphysics simulation frameworks and integrate them into existing Finite Element Analysis (FEA) codes. These simulations are further enhanced by incorporating experimental data from our group and others. To elevate manufacturing quality, we apply computational modeling tools, including those for 3D printing processes and injection molding. Furthermore, we incorporate advanced fatigue life prediction models into commercial software, which, when combined with AI algorithms, enhances the durability of structures under cyclic loading.

We develop micromechanics-based homogenization theory for the prediction of effective physical behavior of composite materials. It combines distinct phases like fiber and matrix into an equivalent, single material description, reducing computational costs. Mathematical methods help translate microscale actions into overall characteristics, aiding in the optimized design of new composites.

(Review paper)
Seunghwa Ryu+*, Sangryun Lee+, Jiyoung Jung, Jinyeop Lee,  and Youngsoo Kim, "Micromechanics-based homogenization of effective properties of composites with anisotropic matrix and interfacial imperfection", Frontiers in Materials 6, 21 (2019).

(Various Physical Properties)
thermoelectric: Jiyoung Jung, Sangryun Lee, Byungki Ryu, and Seunghwa Ryu*, "Investigation of effective thermoelectric properties of composite with interfacial resistance using micromechanics-based homogenisation", International Journal of Heat and Mass Transfer 144, 118620 (2019).
anisotropic thermoelectric: Jiyoung Jung , Wabi Demeke , Sangryun Lee, Jaywan Chung , Byungki Ryu*, and Seunghwa Ryu*, "Micromechanics-based theoretical prediction for thermoelectric properties of anisotropic composites and porous media",  International Journal of Thermal Sciences 165, 106918 (2021).
piezoelectric:  Sangryun Lee, Jiyoung Jung, Seunghwa Ryu*, "Micromechanics-based prediction of the effective properties of piezoelectric composite having interfacial imperfections", Composite Structures 240, 112076 (2020).
thermal conductivity: Sangryun Lee+, Jinyeop Lee+, Byungki Ryu, and Seunghwa Ryu* "A micromechanics-based analytical solution for the effective thermal conductivity of composites with orthotropic matrices and interfacial thermal resistance", Scientific Reports 8, 7266 (2018).

(Liquid Metal Inclusions)
liquid inclusion: Jiyoung Jung, Seunghee Jeong*, Klas Hjort, and Seunghwa Ryu*, "Investigation of Thermal Conductivity for Liquid Metal Composites Based on Mean-Field Homogenization Theory", Soft Matter 16, 5840 (2020).

(Interface & Interphase)
atomistic-continuum scale bridging:  Sangryun Lee, Jiyoung Jung, Youngsoo Kim, Yongtae Kim, and Seunghwa Ryu* "Multiscale Modeling Framework to Predict Effective Stiffness of Crystalline-Matrix Nanocomposite",  International Journal of Engineering Science 161, 103457 (2021).
interface & interphase model test:  Sangryun Lee, Jiyoung Jung, and Seunghwa Ryu* "Applicability of Interface Spring and Interphase Models in Micromechanics for Predicting Effective Stiffness of Polymer-Matrix Nanocomposite",  Extreme Mechanics Letters 49, 101489 (2022)
effective moduli with interfacial damage: Sangryun Lee, Youngsoo Kim, Jinyeop Lee, and Seunghwa Ryu*, "Applicability of interface spring model in micromechanics analysis with interfacial imperfection for predicting modified Eshelby tensor and effective modulus", Mathematics and Mechanics of Solids 24, 2944 (2019).
modifed Eshelby tensor: Sangryun Lee+, Jinyeop Lee+, and Seunghwa Ryu* "Modified Eshelby tensor for an anisotropic matrix with interfacial damage", Mathematics and Mechanics of Solids 24, 1749 (2019).

(Nonlinear Responses)
viscoelastic-viscoplastic: Jiyoung Jung, Youngsoo Kim, Sangryun Lee, Issam Doghri*, and Seunghwa Ryu* "Improved incrementally affine method for viscoelastic-viscoplastic composite by utilizing an adaptive scheme", Composite Structures 297, 115982 (2022)
hyperelastic: Youngsoo Kim†, Jiyoung Jung†, Sangryun Lee, Issam Doghri*, and Seunghwa Ryu* "Adaptive Affine Homogenization Method for Visco-hyperelastic Composites with Interfacial Damage", Applied Mathematical Modelling 107, 72 (2022)

We create simulation frameworks for both 3D printing procedures and the mechanical properties of 3D-printed constructs, to produce high-quality structures. By integrating AI techniques, we fine-tune the 3D printing process to achieve the best blend of print quality and efficiency. Additionally, we optimize the design of 3D-printed structures to strike an ideal balance between lightness, durability, and resilience.

(under construction)

<Manufacturing>
- 미래모빌리티를 위한 3D프린팅 복합재 최적설계기법 개발 (2022.03.-2025.02.)
      sponsor :  한국연구재단  (NRF of Korea)  중견연구 / 연구책임
- 머신러닝적용 재료모델링 및 설계최적화기술을 이용한 전기자동차용 플라스틱 테일게이트 개발 (2021.12.-2024.12)
      sponsor :  중기 소재부품장비전략협력기술개발사업 / 연구책임
- 다상소재 혁신 생산공정 연구센터 (2023.06.-2030.05.)
      sponsor :  한국연구재단  (NRF of Korea)  ERC / 공동연구
- AI기반 배면가열선 자동 생성 알고리즘 개발 연구 (2023.04.-2023.12)
      sponsor :  HD한국조선해양  / 연구책임
- 차세대 반도체 공정 장비용 고내구성, 고신뢰성 부품 개발을 위한 신규 소재 및 AI 기반 설계 기술 연구 (2022.09.-2024.08.)
      sponsor :  한솔아이원스  / 용역연구

<Biomedical>
- 개인 맞춤형 의료진단기술 기반의 심뇌혈관 M3DT 기술개발 및 검증 (2023.04.-2027.12)
      sponsor :  식약처 컴퓨터모델링 기반 의료기기 평가체계 구축사업  / 공동연구
- i-Lab-on-Human 기반 체내 생체 신호 모니터링 시스템 개발 (2022.01.-2025.12.)
      sponsor : 나노종합기술원 / 위탁연구  
-  항생제 내성인자 구분 및 검출 신호 패턴 분석 기술 개발 (2022.01.-2024.12.)
      sponsor : 한국생명공학연구원 / 위탁연구  

<Materials>
- 다각도-멀티스케일 데이터 융합형 리튬이차전지 설계 인공지능 플랫폼 개발 (2023.04.-2027.12)
      sponsor :  한국연구재단  (NRF of Korea)  STEAM 연구사업 / 공동연구

<Modeling Methods>
-  화학작용제 복합전달현상 모델 연구 (2023.02.-2025.02.)
      sponsor : 국방과학연구소 / 위탁연구