+高级检索
首页 > 过刊浏览>2023年第26卷第7期 >705-715. DOI:10.3969/j.issn.1007-9629.2023.07.002
PDF HTML阅读 XML下载 导出引用 引用提醒
基于耦合BAS-MLP的混凝土抗压强度预测
作者:
作者单位:

1.武汉理工大学 土木工程与建筑学院，湖北 武汉 430070;2.长江勘测规划设计研究有限责任公司，湖北 武汉 430010;3.华杰工程咨询有限公司 中南分公司，湖北 武汉 430000

作者简介:

汪声瑞（1955—），男，湖北武汉人，武汉理工大学副教授，硕士生导师，学士.E-mail：WSR_012@126.com

通讯作者:

胡 畔（1994—），男，湖北武汉人，武汉理工大学工程师，博士.E-mail：1204209248@qq.com

中图分类号:

TU528.1

基金项目:

中国工程院咨询研究基金资助项目（2019-XZ-19）


Prediction of Concrete Compressive Strength Based on Coupled BAS-MLP
Author:
Affiliation:

1.School of Civil Engineering and Architecture，Wuhan University of Technology，Wuhan 430070，China;2.Changjiang Survey Planning Design and Research，Co.，Ltd.，Wuhan 430010，China;3.Zhongnan Branch，Huajie Engineering Consulting Co.，Ltd.，Wuhan 430000，China

  • 摘要
    • |
  • 图/表
    • |
  • 访问统计
    • |
  • 参考文献 [18]
    • |
  • 相似文献 [20]
    • | | |
  • 文章评论
    摘要:

    针对1 030组混凝土抗压强度试验数据，通过天牛须搜寻算法（BAS）来训练多层神经网络（MLP），并与混合复杂进化方法（SCE）-MLP、多元宇宙优化算法（MVO）-MLP这2种耦合模型算法进行对比分析，得到可用于预测混凝土抗压强度的算法模型.结果表明：BAS可以显著提高MLP的训练精度和预测精度，该算法比SCE-MLP、MVO-MLP耦合模型算法更快、更准确；与人工神经网络（ANN）和支持向量机（SVM）个体学习算法相比，元启发式算法在混凝土抗压强度预测方便表现出良好的优越性.同时讨论了BAS-MLP模型中与训练数据集数量和输入变量数量相关的因素，发现使用1 030组数据的80%即可获得良好的预测结果.

    Abstract:

    1 030 sets of concrete compressive strength test data were used to train the multilayer perceptron（MLP） through beetle antennae search algorithm（BAS） and combined with shuffled complex evolution（SCE）， multi-verse optimizer（MVO） and the results of two MLP coupling algorithms were compared and analyzed， and an algorithm model that can be used to predict the compressive strength was obtained. The results show that BAS can significantly improve the training accuracy and prediction accuracy of MLP， and this model is faster， more accurate and stable than the other two coupled model algorithms. Compared with artifical neural network（ANN） and support vector machine（SVM） individual learning algorithms， the prediction of concrete compressive strength shows the superiority of meta-heuristic algorithm. The factors related to the number of training data sets and the number of input variables in the BAS-MLP model are discussed， and it is found that using 80% of the 1 030 sets of data can obtain good prediction results.

    表 3 不同输入变量组合的评价指标结果Table 3 Evaluation index results of different of input variable combinations
    图1 MLP结构及其参数图解Fig.1 Illustration of MLP structure and its parametes
    图2 基于SCE-MLP和MVO-MLP复杂度的灵敏度分析Fig.2 Sensitivity analysis based on complexity of SCE-MLP and MVO-MLP
    图3 SCE-MLP、MVO-MLP和BAS-MLP算法的收敛曲线Fig.3 Convergence curves of SCE-MLP， MVO-MLP and BAS-MLP algorithm
    图4 各算法获得的训练样本计算结果及相应的直方图Fig.4 Calculation results and corresponding histogram obtained for the training samples from various algorithms
    图5 混凝土抗压强度预测值与实际值之间的关系Fig.5 Relationship between the predicted and actual compressive strength of concrete
    图7 混凝土抗压强度试验数据和预测数据随养护时间的变化趋势Fig.7 Example of compressive strength trends with respect to curing age of concrete
    图8 BAS-MLP算法与ANN和SVM个体学习算法的对比Fig.8 Comparison of BAS-MLP algorithm compared with ANN and SVM individual learning algorithms
    图9 输入变量数量对评价指标的影响Fig.9 Influence of number of input variables on evaluation index
    表 2 不同训练数据集数量的评价指标结果Table 2 Evaluation index results for different training dataset amounts
    表 1 混凝土抗压强度及关键因素描述性统计数据Table 1 Descriptive statistics data of concrete compressive strength and key factors
    参考文献
    [1] 徐富强, 陶有田. 基于IFOA-RBF算法的混凝土抗压强度预测[J].巢湖学院学报,2014,16(6):7-11.XU Fuqiang, TAO Youtian.On the prediction of concrete compressive strength based on the algorithm of IFOA-RBF[J]. Journal of Chaohu College,2014,16(6):7-11 (in Chinese)
    [2] HENIGAL A, ELBELTGAI E, ELDWINY M, et al. Artificial neural network model for forecasting concrete compressive strength and slump in Egypt[J]. Journal of Al-Azhar University Engineering Sector,2016, 39(11):435-446.
    [3] CHOU J S, TSAI C F, PHAM A D, et al. Machine learning in concrete strength simulations:Multi-nation data analytics[J].Construction and Building Matererials,2014,73:771-780.
    [4] YEH I C. Analysis of strength of concrete using design of experiments and neural networks[J].Journal Materials Civil Engineering,2006,18(4):597-604.
    [5] NEHDI M, GREENOUGH T. Modeling shear capacity of RC slender beams without stirrups using genetic algorithms[J]. Smart Structure System,2007, 3(1):51-68.
    [6] PRAYOGO D, CHENG M Y, WIDJAJA J, et al. Prediction of concrete compressive strength from early age test result using an advanced metaheuristic-based machine learning technique[C]//Proceedings of the International Symposium on Automation and Robotics in Construction (ISARC 2017). Taipei:IAARC Publications, 2017:34.
    [7] PHAM A D,HOANG N D, NGUYEN Q T. Predicting compressive strength of high-performance concrete using metaheuristic-optimized least squares support vector regression[J].Computer Civil Engeering, 2016, 30(3):06015002.
    [8] TIBERTI G,GERMANO F,MUDADU A, et al. An overview of the flexural post-cracking behavior steel fiber reinforced concrete[J]. Structural Concrete,2017,19(3):695-718.
    [9] VENKATESHWARAN A,TAN K H, LI Y. Residual flexural strengths of steel fiber reinforced concrete with multiple hooked-end fibers[J]. Structural Concrete,2017,19(2):352-365.
    [10] BUI D K, NGUYEN T , CHOU J S, et al.A modified firefly algorithm-artificial neural network expert system for predicting compressive and tensile strength of high-performance concrete[J].Construction and Building Matererials,2018,180:320-333.
    [11] TIEN BUI D, ABDULLAHI M A M, GHAREH S,et al. Fine-tuning of neural computing using whale optimization algorithm for predicting compressive strength of concrete[J]. Engineering with Computers,2019, 37(1):701-712.
    [12] DUAN J, ASTERIS P G, NGUYEN H, et al. A novel artificial intelligence technique to predict compressive strength of recycled aggregate concrete using ICA-XGBoost model[J]. Engineering with Computers,2020,37(4):3329-3346.
    [13] JIANG X Y, LI S. BAS:Beetle antennae search algorithm for optimization problems[J]. International Journal of Robotics and Control,2018, 1(1):1.
    [14] 张涛,王才进,刘松玉. 基于ANN的岩土体热阻系数预测模型研究[J].建筑材料学报,2020,23(2):382-391.ZHANG Tao, WANG Caijin, LIU Songyu. Prediction model of thermal resistivity of geomaterial based on artificial neural network[J]. Journal of Building Materials,2020,23(2):382-391.(in Chinese).
    [15] YEH I C. Modeling of strength of high performance concrete using artificial neural networks[J]. Cement and Concrete Research,1998,28(12):1797-1808.
    [16] 陈庆, 马瑞, 蒋正武. 基于GA-BP神经网络的UHPC抗压强度预测与配合比设计[J].建筑材料学报,2020,23(1):177-183.CHEN Qing, MA Rui, JIANG Zhengwu. Compressive strength prediction and mix proportion design of UHPC based on GA-BP neural network[J]. Journal of Building Materials,2020,23(1):177-183.(in Chinese).
    [17] CAMPOS P G, OLIVEIRA E M J, LUDER MIR T B, et al. MLP Networks for classification and prediction with rule extraction mechanism[C]//IEEE International Joint Conference on Networks Proceedings. Budapest:[s.n.],2004:1387-1392.
    [18] 刘富成. 基于随机森林和智能算法预测混凝土抗压强度的方法:112069567[P].2020-12-11.LIU Fucheng. Prediction method of concrete compressive strength based on random forest and intelligent algorithm:112069567[P]. 2020-12-11.(in Chinese)
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

汪声瑞,胡畔,陈思宝,肖约.基于耦合BAS-MLP的混凝土抗压强度预测[J].建筑材料学报,2023,26(7):705-715

复制
分享
文章指标
  • 点击次数:139
  • 下载次数: 343
  • HTML阅读次数: 10
  • 引用次数: 0
历史
  • 收稿日期:2022-09-26
  • 最后修改日期:2023-01-05
  • 在线发布日期: 2023-09-15
文章二维码
您是第 2051312 位访问者
主管单位 ：教育部
主办单位 ：同济大学
地址 ：上海市四平路1239号同济大学综合楼1701室
电话 ：021-65981597 传真 ：021-65981597
建筑材料学报 ® 2025 版权所有