检索结果-内蒙古大学图书馆

Fatigue Life Prediction of FRP-Strengthened Reinforced Concrete Beams Based on Soft Computing Techniques

MATERIALS 2025年第2期18卷 230-230页

作者： Zhang, Zhimei Wang, Xiaobo Shanghai Univ Sch Mech & Engn Sci Dept Civil Engn Shanghai 200444 Peoples R China

This paper establishes fatigue life prediction models using the soft computing method to address insufficient parameter consideration and limited computational accuracy in predicting the fatigue life of fiber-reinforced polymer (FRP) strengthened concrete beams. Five different input forms were proposed by collecting 117 sets of fatigue test data of FRP-strengthened concrete beams from the existing literature and integrating the outcomes from Pearson correlation analysis and significance testing. Using gene expression programming (GEP), the effects of various input configurations on the accuracy of model predictions were examined. The model prediction results were also evaluated using five statistical indicators. The GEP model used concrete compressive strength, the steel reinforcement stress range ratio to the yield strength, and the stiffness factor as input parameters. Subsequently, using the same input parameters, the Multi-Objective genetic Algorithm Evolutionary Polynomial Regression (MOGA-EPR) method was then employed to develop a fatigue life prediction model. Sensitivity analyses of the GEP and MOGA-EPR models revealed that both could precisely capture the fundamental connections between fatigue life and multiple contributing variables. Compared to existing models, the proposed ones have higher prediction accuracy with a coefficient of determination reaching 0.8, significantly enhancing the accuracy of fatigue life predictions for FRP-strengthened concrete beams.

关键词： multi-objective genetic algorithm evolutionary polynomial regression gene expression programming fatigue life prediction model fiber-reinforced polymer composite material FRP strengthened beams

来源：评论

学校读者我要写书评

暂无评论

DANCE: Distributed co-evolutionary design of velocity controllers for swarm intelligence robots in flocking and entrapping tasks

引用

SWARM AND EVOLUTIONARY COMPUTATION 2025年 94卷

作者： Wang, Chen Zhu, Cheng Zhu, Xianqiang Lei, Hongtao Zhang, Weiming Wu, Meng Natl Univ Def Technol Deya Rd Changsha 410073 Hunan Peoples R China Natl Key Lab informat Syst Engn Deya Rd Changsha 410073 Hunan Peoples R China

This study combined evolutionary algorithm and reinforcement learning to propose anew automated design method for generating swarm robots velocity controller model. It alternately evolves heterogeneous swarm and homogeneous swarm through a gene expression programming method that introduces reinforcement learning, and assembles function nodes and leaf nodes into new mathematical formulas during the evolution process. The method enable to realize the effect of swarm robots emerging to perform swarm tasks such as flocking and entrapping. What is more, anew swarm rule was discovered during the evolution process, which is used to realize the flocking of swarm robots at any angle. The experimental results show that the swarm motion controller automatically generated by the model has high task completion efficiency and strong generalization.

关键词： Swarm intelligence Multi-agent systems Robots Evolutionary computation Reinforcement learning gene expression programming

来源：评论

学校读者我要写书评

暂无评论

Machine learning-based prediction of elliptical double steel columns under compression loading

引用

JOURNAL OF BIG DATA 2025年第1期12卷 1-43页

作者： Ren, Yang Isleem, Haytham F. Almoghaye, Walaa J. K. Hamed, Abdelrahman Kamal Jangir, Pradeep Arpita Tejani, Ghanshyam G. Ezugwu, Absalom E. Soliman, Ahmed A. City Univ Seattle Master Comp Sci Seattle WA 98102 USA Univ York Dept Comp Sci York YO10 5DD England Nanjing Univ Sci & Technol Dept Management Sci & Engn Nanjing 210094 Jiangsu Peoples R China Horus Univ Egypt Fac Engn Civil Engn Dept New Damietta 34517 Egypt Chandigarh Univ Univ Ctr Res & Dev Mohali 140413 India Al Ahliyya Amman Univ Hourani Ctr Appl Sci Res Amman 19328 Jordan Graph Era Hill Univ Dept Math Dehra Dun India Graph Era Deemed Be Univ Dept Comp Sci & Engn Dehra Dun Uttarakhand India Saveetha Inst Med & Tech Sci Saveetha Sch Engn Dept Biosci Chennai 602105 India Yuan Ze Univ Dept Ind Engn & Management Taoyuan Taiwan North West Univ Unit Data Sci & Comp 11Hofman St ZA-2520 Potchefstroom South Africa Future Univ Egypt Fac Engn & Technol Struct Engn & Construct Management Dept Cairo 11835 Egypt Chitkara Univ Chitkara Ctr Res & Dev Baddi 174103 Himachal Prades India

This paper presents a comprehensive investigation into the prediction of axial load capacity (P) for elliptical double steel columns (EDSCs) using a diverse set of machine learning models (MLMs). These include Artificial Neural Network (ANN), gene expression programming (GEP), Support Vector Regression (SVR), Random Forest (RF), and AdaBoost. Among the models, AdaBoost demonstrated superior performance, achieving an R2 of 0.996 and a MAPE of 0.013 during training, outperforming other models under identical conditions. Using a dataset of 119 finite element models derived from prior experimental research, the study validates the proposed solution through k-fold cross-validation, feature importance analysis, and detailed comparisons with experimental data. A Graphical User Interface (GUI) was developed specifically for the AdaBoost model due to its superior accuracy and efficiency, offering engineers a practical and accessible tool for axial load prediction in EDSC design. This research highlights the significance of using advanced machine learning techniques for structural engineering applications, providing valuable insights for the optimization of EDSC performance and design under varying conditions.

关键词： Elliptical double steel columns Machine learning Artificial neural networks gene expression programming Support vector regression Random forest AdaBoost

来源：评论

学校读者我要写书评

暂无评论

Application of enhanced hybrid optimization models for discharge prediction of cosine sharp-crested weirs

引用

FLOW MEASUREMENT AND INSTRUMENTATION 2025年 102卷

作者： Akhgar, Samira Azimi, Amir H. Foroudi, Ali Lakehead Univ Dept Civil Engn Thunder Bay ON P7B 5E1 Canada Quchan Univ Technol Fac Engn Sci Dept Civil Engn Quchan Iran

The discharge coefficient plays a crucial role in estimation of discharge over sharp-crested weirs. This study employs various hybrid optimization algorithms to predict the discharge coefficient over half-cycle and full-cycle cosine sharp-crested weirs. A Support Vector Machine (SVM) algorithm was utilized, with its parameters optimized using the Gray Wolf Optimization (GWO) algorithm. The performance of the GWO-SVM hybrid model was compared against the Gaussian Process Regression (GPR) model. Additionally, gene expression programming (GEP) was applied to derive the best predictive equations for both weir types. For this purpose, a broad range of laboratory data, comprising geometric and hydraulic information (110 data sets for half-cycle and 270 data sets for full-cycle cosine weirs), has been considered. The results indicated that the GWO-SVM model demonstrated high accuracy, particularly in peak discharge predictions, achieving errors below 4 %. The proposed models incorporated the influence of the approach Froude number, a critical factor often overlooked in prior discharge prediction models. Sensitivity analysis revealed that the ratio of weir height to upstream water head (ho/P) and Froude number play significant roles in prediction of discharge coefficient for cosine sharp-crested weirs. The frequency histogram of prediction for half- and full-cycle cosine sharp-crested weirs showed that the GEP model exhibited a uniform error distribution with a slight underprediction tendency, whereas regression models were less reliable for both weir types. Overall, the AI-based models outperformed conventional regression approaches, effectively minimizing underpredictions and overpredictions, and providing robust discharge coefficient predictions.

关键词： Discharge coefficient Sharp-crested weirs Support vector machine SVM gene expression programming GEP Gray wolf optimization GWO Gaussian process regression GPR

来源：评论

学校读者我要写书评

暂无评论

Machine-learnt closure models based on computational space training frameworks for large-eddy simulations in natural convection

引用

INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER 2025年 236卷

作者： Liu, Liyuan Lav, Chitrarth Sandberg, Richard D. Univ Melbourne Dept Mech Engn Melbourne Vic 3010 Australia Ctr Proc Innovat Coxon Bldg John Walker Rd Sedgefield TS21 3FE England CFD Methodol Grp Racing Bulls Team Falmouth England

Natural convection heat transfer at very high Rayleigh numbers (Ra) Ra ) is a commonly occurring phenomenon in many industry applications. Practically, Large Eddy Simulations (LES) are often chosen as an economical method for the numerical prediction of these flows. However, developing accurate LES models that generalize well to complex geometries poses a challenge, particularly for data-driven methods. Thus, in the current study, machine-learnt closure models with the general choice of length scales for different geometries are proposed, where the subgrid-scale (SGS) stress and heat-flux models developed by using gene expression programming (GEP) are built in the computational space. Two geometrically distinct natural convection cases (Rayleigh-B & eacute;nard convection and the concentric horizontal annulus) are chosen to develop and generalize the models. Subsequently, the formulation between the SGS closures, the total and the resolved large-scale turbulent stress and heat-flux is derived in the compressible LES context. The a-priori results show that the GEP models developed in computational space significantly improve both the SGS stress and SGS heat-flux prediction while being robust to complex flows. Similarly, the a-posteriori results demonstrate that the GEP models perform better than the wall-adapting local eddy-viscosity (WALE) model in the prediction of the mean SGS stress and the SGS heat-flux. The data-driven approach for turbulence model development presented clearly offers a promising models' generalization for LES in the prediction of SGS stress and heat-flux.

关键词： Natural convection gene expression programming Subgrid-scale stress Subgrid-scale heat-flux Complex turbulent flows

来源：评论

学校读者我要写书评

暂无评论

AI-powered optimization of engineered cementitious composites properties and CO2 emissions for sustainable construction

引用

CASE STUDIES IN CONSTRUCTION MATERIALS 2025年 22卷

作者： Chang, Qiuying Zhao, Chuanhai Alateah, Ali H. Alinsaif, Sadiq Sufian, Muhammad Ahmad, Ayaz Changchun Univ Architecture & Civil Engn Sch Civil Engn Changchun 130607 Jilin Peoples R China Univ Hafr Al Batin Coll Engn Dept Civil Engn Hafar al Batin 39524 Saudi Arabia Univ Hafr Al Batin Coll Comp Sci & Engn Hafar Al Batin 39524 Saudi Arabia Southeast Univ Sch Civil Engn Nanjing 210096 Peoples R China COMSATS Univ Islamabad Dept Civil Engn Abbottabad Campus Islamabad 22060 Pakistan

Machine learning (ML) approaches are ideal for meeting the complex requirements of structural elements, as they have shown tremendous potential for improving material performance and promoting sustainability in the building industry. Recently, ML techniques have become costeffective and efficient for predicting how material properties influence the proposed mixture characteristics, cost, and time. ML approaches in civil engineering should be further developed to optimize the incorporation of fly ash and slag into mixtures. In this study, machine intelligence algorithms, including an adaptive neuro-fuzzy inference system (ANFIS), a hybrid genetic algorithm-artificial neural network (GA-ANN), and gene expression programming (GEP), were used to predict the tensile strain capacity (TSC) and tensile strength (TS) of engineered cementitious composites (ECCs). The dataset comprised samples for TSC and TS with 11 parameters. To assess feature relevance, a sensitivity analysis (SA) and a complete parametric analysis using the best-performing model were conducted. The results demonstrated that GEP achieved the highest prediction accuracy for TS and TSC, with R2 values of 0.93 and 0.94 for training and 0.91 and 0.92 for testing, respectively. The models' performance was evaluated using MSE, RMSE, and MARE values for tensile strength and strain capacity during both training and testing phases, with GEP emerging as the best model, followed by ANFIS and GA-ANN. Sensitivity analysis was performed to determine the contribution of input variables. For CO2 emission prediction, the GEP model achieved the highest accuracy with R2 at 0.97, followed by ANFIS and GA-ANN, with ANFIS having the lowest accuracy with R2 at 0.95. Additionally, a practical graphical user interface (GUI) tool was developed to apply the study's findings, supporting sustainability efforts in the construction industry.

关键词： gene expression programming Sensitivity analyses Graphical user interface Co2 emission

来源：评论

学校读者我要写书评

暂无评论

Predictive modeling for durability characteristics of blended cement concrete utilizing machine learning algorithms

引用

CASE STUDIES IN CONSTRUCTION MATERIALS 2025年 22卷

作者： Fu, Bo Lei, Hua Ullah, Irfan El-Meligy, Mohammed El Hindi, Khalil Javed, Muhammad Faisal Ahmad, Furqan North Minzu Univ Coll Civil Engn Yinchuan 750021 Peoples R China Hohai Univ Dept Civil & Transportat Engn Nanjing Peoples R China Appl Sci Private Univ Appl Sci Res Ctr Amman Jordan Jadara Univ Res Ctr POB 733 Irbid Jordan King Saud Univ Coll Comp & Informat Sci Dept Comp Sci Riyadh 11543 Saudi Arabia GIK Inst Engn Sci & Technol Dept Civil Engn Swabi 23640 Pakistan Western Caspian Univ Baku Azerbaijan UNHCR Kabul Afghanistan

Chloride penetration and carbonation resistance are critical durability attributes that assess concrete's ability to withstand challenging environmental conditions. However, determining these parameters requires time-consuming and resource-intensive physical experiments. Accordingly, this study employed gene expression programming (GEP) and multi-expression programming (MEP) to develop a robust model for predicting these parameters, providing mathematical equations for their estimation. Additionally, the study to develop a graphical user interface that would allow for predictions based solely on input values, thereby eliminating the need for extensive physical testing. To thoroughly assess the effectiveness of the proposed GEP and MEP models, a range of statistical metrics were employed, including the coefficient of determination (R2), adjusted R2, root mean square error (RMSE), mean absolute error (MAE), and root mean square error to observation's standard deviation ratio (RSR), along with engineering indices like the a10-index and a20-index. Both GEP and MEP models consistently demonstrated outstanding performance across all statistical indicators for both carbonation rate and chloride penetration. The GEP model showed high precision in modeling chloride penetration with an R2 of 0.954, MAE of 0.252, and RMSE of 1.050, and for carbonation rate with an R2 of 0.99, MAE of 0.230, and RMSE of 1.100. Similarly, the MEP model performed well, achieving an R2 of 0.913, MAE of 0.489, and RMSE of 1.434 for chloride penetration, and an R2 of 0.985, MAE of 0.560, and RMSE of 1.440 for carbonation rate. In addition, the SHapley Additive exPlanation (SHAP) method was employed to comprehend the model estimations. In predicting chloride penetration, cement to water ratio (C/B) emerged as the most impactful feature, followed by silica fume to binder ratio (SF/B) and water to binder ratio (W/B) in terms of importance. For carbonation rate, W/B stood out as the most influential, wit

关键词： Blended cement concrete Machine learning gene expression programming Multiple expression programming

来源：评论

学校读者我要写书评

暂无评论

MODELING AND OPTIMIZATION OF STANDARD CONCRETE CONTAINING GRANULE BLAST FURNACE SLAG: A gene expression MODELING BASED MULTI-RESPONSE WEIGHTED NON-LINEAR programming APPLICATION

引用

INTERNATIONAL JOURNAL OF INDUSTRIAL ENGINEERING-THEORY APPLICATIONS AND PRACTICE 2018年第4期25卷 490-506页

作者： Simsek, Baris Cankiri Karatekin Univ Fac Engn Dept Chem Engn Cankiri Turkey

In this study, gene expression models have been obtained for prediction of slump flow, 3rd day compressive strength, 7th day compressive strength and 28th day compressive strength of normal weight concrete containing granule blast furnace slag using geneXproTools software package version 5.0 with data collected from concrete quality control laboratory. Optimal mixture proportions of normal weight concrete containing granule blast furnace slag were determined by multi-response weighted non-linear programming using using non-linear gene expression models via Matlab version 2015a. It is concluded that proposed gene expression models based multi-response weighted non-linear programming methodology is the first application that is effective in order to determine optimal mixture proportions when compared to other methods such as fuzzy modeling and artificial neural network. In addition, this methodology provides the possibility of optimal use of resources such as raw materials for laboratory personnel without the need for additional experiments causing a loss of time.

关键词： artificial neural network granule blast furnace slag (GBFS) fuzzy logic gene expression programming multi-response optimization weighted non-linear programming

来源：评论

学校读者我要写书评

暂无评论

Optimized punching shear design in steel fiber-reinforced slabs: Machine learning vs. evolutionary prediction models

引用

ENGINEERING STRUCTURES 2025年 322卷

作者： Albostami, Asad S. Mohamad, Safaa A. Alzabeebee, Saif Al-Hamd, Rwayda Kh. S. Al-Bander, Baidaa Oryx Universal Coll Partnership Liverpool John Moo Sch Engn & Construct Doha Qatar Mustansiriyah Univ Coll Engn Highway & Transportat Engn Dept Baghdad Iraq Univ Al Qadisiyah Coll Engn Rd & Transport Engn Dept Al Diwaniyah Al Qadisiyah Iraq Keele Univ Sch Comp Newcastle ST5 5BG England Abertay Univ Fac Social & Appl Sci Dept Built Environm & Life Sci Dundee Scotland

This research paper focuses on utilizing Artificial Neural Networks (ANN), Multi-Objective genetic Algorithm Evolutionary Polynomial Regression (MOGA-EPR), and gene expression programming (GEP) to predict the punching shear strength of Steel Fibre-Reinforced Concrete (SFRC) slabs. In order to formulate predictions, research and analysis were carried out making use of a dataset, this dataset included several parameters that impact on punching shear strength, including SFRC slabs longitudinally and transversely, using ANN, GEP, and MOGA-EPR methods. The developed models exhibited very good performance, as the soft computing techniques (GEP and MOGA-EPR) achieved R2 values of 0.91 to 0.93, while the ANN technique was higher at 0.95. Furthermore, two case studies were incorporated to carry out cost analyses of the models in real-world applications. It was shown that the efficiency of the Machine Learning (ML) models in reducing the costs of materials is relatively high, as they were capable of better predictions than the standard methods employed by the codes.

关键词： Machine learning Soft computing Artificial neural networks Multi-objective genetic algorithm evolutionary polynomial regression gene expression programming Punching shear Steel fibre reinforced concrete

来源：评论

学校读者我要写书评

暂无评论

A hybrid machine learning approach for predicting fiber-reinforced polymer-concrete interface bond strength

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2025年 148卷

作者： Wahab, Sarmed Salami, Babatunde Abiodun Danish, Hassan Nisar, Saad Alateah, Ali H. Alsubeai, Ali Univ Engn & Technol Taxila Dept Civil Engn Taxila Pakistan Cardiff Metropolitan Univ Cardiff Sch Management Cardiff CF5 2YB Wales HITEC Univ Taxila Dept Civil Engn Taxila Pakistan Univ Teknol PETRONAS Dept Civil & Environm Engn Perak Malaysia Univ Hafr Al Batin Coll Engn Dept Civil Engn POB 1803 Hafar al Batin 39524 Saudi Arabia Jubail Ind Coll Dept Civil Engn Royal Commiss Jubail Jubail Ind City 31961 Saudi Arabia

The interfacial bond strength between fiber-reinforced polymer (FRP) sheets and concrete is crucial for structural design. This study presented a novel approach using ensemble learning models to predict bond strength and analyze input parameters' influence. No previous research used gene expression programming (GEP) for developing bond strength models in single shear tests. This research introduced GEP to develop an expression for estimating bond strength, comparing its performance with existing empirical models used in design codes. Six ensemble machine learning models were tested: extreme gradient boosting (XGBoost), light gradient boosting (LightGBM), categorical boosting (CatBoost), adaptive boosting (AdaBoost), random forest (RF), and explainable boosting machine (EBM), using 855 samples. CatBoost demonstrated superior performance with R2 = 0.98, RMSE = 1.61 kN, and MAE = 1.18 kN. The study utilized EBM's interpretability for parametric analysis through local and global explanations. Results showed FRP material and geometric properties had greater impact on bond strength than concrete properties. The novel GEP-developed empirical expression achieved higher accuracy compared to existing empirical models, with R2 = 0.812, RMSE = 4.63 kN, and MAE = 3.58 kN. The GEP model primarily relied on FRP's material and geometric properties, aligning with parametric analysis findings. Based on the results, both the CatBoost ensemble learning model and GEP model are recommended for estimating FRP-concrete interfacial bond strength.

关键词： Fiber-reinforced polymer concrete Interfacial bond strength Extreme gradient boosting Light gradient boosting Categorical boosting Adaptive boosting Random forest gene expression programming

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：