Sangyoung Yoon , Boohyun An , Chan Yeob Yeun , Ernesto Damiani , Malik Khalfan , Tae-Yeon Kim
{"title":"Solitary-wave-based deep learning for compressive strength estimation in cementitious materials","authors":"Sangyoung Yoon , Boohyun An , Chan Yeob Yeun , Ernesto Damiani , Malik Khalfan , Tae-Yeon Kim","doi":"10.1016/j.ijmecsci.2025.110170","DOIUrl":null,"url":null,"abstract":"<div><div>This study presents a novel deep learning (DL)-based approach for predicting the compressive strength of cementitious materials using highly nonlinear solitary waves (HNSWs) as input data. The proposed method leverages convolutional neural networks (CNNs) to classify compressive strength of mortar by transforming continuous measurements into discrete categories. Four different formats of HNSW signals are explored to evaluate the impact of signal preprocessing on model performance. Multiple mode testing is implemented to enhance the robustness of predictions, using multiple signals from the same class to reduce variability and stabilize results. The DL models were tested on datasets varying by water-to-cement (w/c) ratios and hydration time, achieving superior performance through signal slicing and frequency-domain transformations. Notably, the model achieved high prediction accuracy, with R² values up to 0.989 and RMSE as low as 0.930 MPa, demonstrating its reliability for predicting compressive strength in cementitious materials. Comparative analyses with benchmark architectures such as AlexNet, GoogleNet, and ResNet-18 highlight the effectiveness of the tailored CNN model, which consistently outperforms these benchmarks, especially in multiple mode testing.</div></div>","PeriodicalId":56287,"journal":{"name":"International Journal of Mechanical Sciences","volume":"291 ","pages":"Article 110170"},"PeriodicalIF":7.1000,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Mechanical Sciences","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0020740325002565","RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, MECHANICAL","Score":null,"Total":0}
引用次数: 0
Abstract
This study presents a novel deep learning (DL)-based approach for predicting the compressive strength of cementitious materials using highly nonlinear solitary waves (HNSWs) as input data. The proposed method leverages convolutional neural networks (CNNs) to classify compressive strength of mortar by transforming continuous measurements into discrete categories. Four different formats of HNSW signals are explored to evaluate the impact of signal preprocessing on model performance. Multiple mode testing is implemented to enhance the robustness of predictions, using multiple signals from the same class to reduce variability and stabilize results. The DL models were tested on datasets varying by water-to-cement (w/c) ratios and hydration time, achieving superior performance through signal slicing and frequency-domain transformations. Notably, the model achieved high prediction accuracy, with R² values up to 0.989 and RMSE as low as 0.930 MPa, demonstrating its reliability for predicting compressive strength in cementitious materials. Comparative analyses with benchmark architectures such as AlexNet, GoogleNet, and ResNet-18 highlight the effectiveness of the tailored CNN model, which consistently outperforms these benchmarks, especially in multiple mode testing.
期刊介绍:
The International Journal of Mechanical Sciences (IJMS) serves as a global platform for the publication and dissemination of original research that contributes to a deeper scientific understanding of the fundamental disciplines within mechanical, civil, and material engineering.
The primary focus of IJMS is to showcase innovative and ground-breaking work that utilizes analytical and computational modeling techniques, such as Finite Element Method (FEM), Boundary Element Method (BEM), and mesh-free methods, among others. These modeling methods are applied to diverse fields including rigid-body mechanics (e.g., dynamics, vibration, stability), structural mechanics, metal forming, advanced materials (e.g., metals, composites, cellular, smart) behavior and applications, impact mechanics, strain localization, and other nonlinear effects (e.g., large deflections, plasticity, fracture).
Additionally, IJMS covers the realms of fluid mechanics (both external and internal flows), tribology, thermodynamics, and materials processing. These subjects collectively form the core of the journal's content.
In summary, IJMS provides a prestigious platform for researchers to present their original contributions, shedding light on analytical and computational modeling methods in various areas of mechanical engineering, as well as exploring the behavior and application of advanced materials, fluid mechanics, thermodynamics, and materials processing.