Establishing a universal machine learning (ML) model in structural engineering is vital for understanding how various parameters, like geometry and material properties, influence a structure's behavior. This study aims to create a comprehensive ML model that considers the impact of different cross-sectional parameters on the ultimate load capacity (ULC) of concrete-filled steel tube (CFST) columns. This model assists engineers in making informed design decisions. The study employs a large dataset of 3094 data points with diverse geometric and material properties of CFST columns. After adjusting input features, robust boosting ML models (Catboost, LightGBM, and XGB) are meticulously fine-tuned using grid search and fivefold cross-validation. Monte Carlo simulation is used for further assessment. The results demonstrate that the most accurate XGB model delivers impressive accuracy, comparable to or better than existing literature models that focused on a single CFST column cross-section. The chosen XGB model is then utilized for feature importance analysis, local performance assessment, and sensitivity analysis through 1-D and 2-D partial dependence plots. These analyses help assess the input's contribution and effect on ULC prediction for CFST columns.
{"title":"Universal boosting ML approaches to predict the ultimate load capacity of CFST columns","authors":"Thuy-Anh Nguyen, Khuong Le Nguyen, Hai-Bang Ly","doi":"10.1002/tal.2071","DOIUrl":"https://doi.org/10.1002/tal.2071","url":null,"abstract":"Establishing a universal machine learning (ML) model in structural engineering is vital for understanding how various parameters, like geometry and material properties, influence a structure's behavior. This study aims to create a comprehensive ML model that considers the impact of different cross-sectional parameters on the ultimate load capacity (ULC) of concrete-filled steel tube (CFST) columns. This model assists engineers in making informed design decisions. The study employs a large dataset of 3094 data points with diverse geometric and material properties of CFST columns. After adjusting input features, robust boosting ML models (Catboost, LightGBM, and XGB) are meticulously fine-tuned using grid search and fivefold cross-validation. Monte Carlo simulation is used for further assessment. The results demonstrate that the most accurate XGB model delivers impressive accuracy, comparable to or better than existing literature models that focused on a single CFST column cross-section. The chosen XGB model is then utilized for feature importance analysis, local performance assessment, and sensitivity analysis through 1-D and 2-D partial dependence plots. These analyses help assess the input's contribution and effect on ULC prediction for CFST columns.","PeriodicalId":501238,"journal":{"name":"The Structural Design of Tall and Special Buildings","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138511039","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Akbar Rasoulnia, Vahid Broujerdian, Ali Ghamari, Abbas Ghadami
The present study was undertaken to characterize the structural behavior and ultimate flexural strength of steel plate I-girders under pure flexural moment at elevated temperatures. A novel design procedure along with flexural design curves was proposed to predict the flexural behavior of the I-girders and estimate corresponding ultimate flexural strengths. The main strategy of the procedure is to find an ambient-temperature equivalent of the I-girder by quantifying and formulating the effects of elevated temperatures. The proposed procedure comprises overall and partial phases. The former phase deals with the determination of equivalent laterally unbraced length, and the latter phase addresses the equivalent web and compression flange slenderness parameters. The calibration factor was defined to adapt the design curves to the effects of high compression flange slenderness parameters and residual stress at elevated temperatures. To generate comparative results, a numerical study was conducted by analyzing 216 finite element (FE) models. Fifty-four out of 216 FE models with different cross-sectional elements were dedicated to the I-girders fail by yield or local buckling failure mode, the results of which are reported in the present paper. Data fitting analysis was carried out to capture the variation of calibration factor with respect to compression flange slenderness parameters. By calibrating the proposed design procedure, the results were converged and, therefore, good conformity was reached between the numerical and parametric results.
{"title":"Flexural design curves of steel plate I-girders at elevated temperatures: I-girders with yield or local buckling failure mode","authors":"Akbar Rasoulnia, Vahid Broujerdian, Ali Ghamari, Abbas Ghadami","doi":"10.1002/tal.2070","DOIUrl":"https://doi.org/10.1002/tal.2070","url":null,"abstract":"The present study was undertaken to characterize the structural behavior and ultimate flexural strength of steel plate I-girders under pure flexural moment at elevated temperatures. A novel design procedure along with flexural design curves was proposed to predict the flexural behavior of the I-girders and estimate corresponding ultimate flexural strengths. The main strategy of the procedure is to find an ambient-temperature equivalent of the I-girder by quantifying and formulating the effects of elevated temperatures. The proposed procedure comprises overall and partial phases. The former phase deals with the determination of equivalent laterally unbraced length, and the latter phase addresses the equivalent web and compression flange slenderness parameters. The calibration factor was defined to adapt the design curves to the effects of high compression flange slenderness parameters and residual stress at elevated temperatures. To generate comparative results, a numerical study was conducted by analyzing 216 finite element (FE) models. Fifty-four out of 216 FE models with different cross-sectional elements were dedicated to the I-girders fail by yield or local buckling failure mode, the results of which are reported in the present paper. Data fitting analysis was carried out to capture the variation of calibration factor with respect to compression flange slenderness parameters. By calibrating the proposed design procedure, the results were converged and, therefore, good conformity was reached between the numerical and parametric results.","PeriodicalId":501238,"journal":{"name":"The Structural Design of Tall and Special Buildings","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138511032","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：