基于Stacking模型融合的高性能混凝土强度预测方法

胡以婵; 梁铭; 谢灿荣; 解威威; 翁贻令; 池浩; 彭浩; 罗雪霜

硅酸盐通报, 2023, 42 (11): 3914, 网络出版: 2023-12-11

基于Stacking模型融合的高性能混凝土强度预测方法

Strength Prediction Method of High Performance Concrete Based on Stacking Model Fusion

论文大纲

胡以婵 ^*梁铭谢灿荣解威威翁贻令池浩彭浩罗雪霜

作者单位

广西路桥工程集团有限公司, 南宁 530011

混凝土强度预测模型集成学习 stacking算法 XGBoost算法 RF算法 concrete strength prediction model ensemble learning stacking algorithm XGBoost algorithm RF algorithm

摘要

针对传统经验公式对高性能混凝土强度预测时存在偏差大、效率低等问题, 本文提出一种基于Stacking模型融合的高性能混凝土强度预测方法。首先, 通过数据清洗和归一化对1 030组高性能混凝土抗压强度试验数据进行预处理, 剔除异常数据及消除数据间量纲影响; 其次, 基于极端梯度提算法(XGBoost)、类别优先梯度提升算法、多层感知器和随机森林(RF)4种算法开展超参数优化、模型训练和评估, 采用决定系数、均方根误差和平均绝对误差对比分析4种基学习器对强度预测的整体效果, 在此基础上构建基于Stacking集成学习融合多种机器学习算法的高性能混凝土强度预测模型; 最后, 采用103组新的高性能数据集对模型进行验证, 并开展可解释分析。结果表明: 与其他基学习器的组合相比, XGBoost和RF融合模型的预测精度和性能均明显提高, 泛化性能较好, 且可解释分析显示最重要的输入特征变量是龄期和水泥, 说明模型内在的预测逻辑与工程实践的经验较吻合, 具有较高的合理性与可靠度。研究结果为进一步提高高性能混凝土强度的预测精度提供参考。

Abstract

Strength prediction method of high performance concrete based on stacking model fusion was proposed to address the issues of large deviations and low efficiency of traditional empirical formulas for high-performance concrete strength prediction. Firstly1 030 sets of high-performance concrete compressive strength test data were preprocessed through data cleaning and normalization to eliminate abnormal data and the dimensional influence among data. Secondlybased on extreme gradient boosting (XGBoost)category boostingmulti-layer perceptronand random forest (RF) algorithmshyperparameter optimizationmodel training and evaluation were conductedand the overall effect of the four base learners on strength prediction were compared and analyzed using coefficient of determination R2root mean square error and mean absolute error. Based on thisa Stacking ensemble learning model was constructedwhich fuses multiple machine learning algorithms for concrete strength prediction. Finallythe model was validated using 103 sets of new dataand interpretable analysis was performed. The results show that compared to other combinations of base learnersthe fusion model using XGBoost and RF significantly improves prediction accuracy and performanceand has good generalization performance. The interpretable analysis shows that the most important input feature variables are age and cementindicating that the internal prediction logic of the model is more in line with engineering practice experiencehaving high rationality and reliability. The research results provide reference for further improving the accuracy of high-performance concrete strength prediction.

参考文献

[1] 吴中伟. 混凝土科学技术近期发展方向的探讨［J］. 硅酸盐学报19797(3): 262-270.

[2] 李宗津孙伟潘金龙. 现代混凝土的研究进展［J］. 中国材料进展200928(11): 1-7+53.

[3] 俞桂良. 高强混凝土强度预测人工智能方法及应用［J］. 混凝土2010(10): 41-43.

[4] 季韬林挺伟林旭健. 基于人工神经网络的混凝土抗压强度预测方法［J］. 建筑材料学报20058(6): 677-681.

[5] 唐明述. 水泥混凝土与可持续发展［J］. 中国有色金属学报200414(增刊1): 164-172.

[6] LI Z ZYOON JZHANG Ret al. Machine learning in concrete science: applicationschallengesand best practices［J］. NPJ Computational Materials20228: 127.

[7] CHOU J STSAI C FPHAM A Det al. Machine learning in concrete strength simulations: multi-nation data analytics［J］. Construction and Building Materials201473: 771-780.

[8] NUNEZ IMARANI AFLAH Met al. Estimating compressive strength of modern concrete mixtures using computational intelligence: a systematic review［J］. Construction and Building Materials2021310: 125279.

[9] MOHTASHAM M MSARADAR ARAHMATI Ket al. Predictive models for concrete properties using machine learning and deep learning approaches: a review［J］. Journal of Building Engineering202363: 105444.

[10] KHAMBRA GSHUKLA P. Novel machine learning applications on fly ash based concrete: an overview［J］. Materials Today: Proceedings202380: 3411-3417.

[11] BEN C WFLAH MNEHDI M L. Machine learning prediction of mechanical properties of concrete: critical review［J］. Construction and Building Materials2020260: 119889.

[12] YUAN X ZTIAN Y ZAHMAD Wet al. Machine learning prediction models to evaluate the strength of recycled aggregate concrete［J］. Materials202215(8): 2823.

[13] SHANG M JLI H JAHMAD Aet al. Predicting the mechanical properties of RCA-based concrete using supervised machine learning algorithms［J］. Materials202215(2): 647.

[14] IFTIKHAR BALIH S CVAFAEI Met al. Predictive modeling of compressive strength of sustainable rice husk ash concrete: ensemble learner optimization and comparison［J］. Journal of Cleaner Production2022348: 131285.

[15] FENG D CLIU Z TWANG X Det al. Machine learning-based compressive strength prediction for concrete: an adaptive boosting approach［J］. Construction and Building Materials2020230: 117000.

[16] ASTERIS P GSKENTOU A DBARDHAN Aet al. Predicting concrete compressive strength using hybrid ensembling of surrogate machine learning models［J］. Cement and Concrete Research2021145: 106449.

[17] FAROOQ FAHMED WAKBAR Aet al. Predictive modeling for sustainable high-performance concrete from industrial wastes: a comparison and optimization of models using ensemble learners［J］. Journal of Cleaner Production2021292: 126032.

[18] KANG M CYOO D YGUPTA R. Machine learning-based prediction for compressive and flexural strengths of steel fiber-reinforced concrete［J］. Construction and Building Materials2021266: 121117.

[19] 马高刘康. 基于BP神经网络CFRP约束混凝土抗压强度预测［J］. 湖南大学学报(自然科学版)202148(9): 88-97.

[20] NGUYEN K TNGUYEN Q DLE T Aet al. Analyzing the compressive strength of green fly ash based geopolymer concrete using experiment and machine learning approaches［J］. Construction and Building Materials2020247: 118581.

[21] 韩建军赵道松李建平. 基于BP神经网络的垃圾飞灰混凝土抗压强度预测模型［J］. 混凝土2022(9): 78-81.

[22] 徐潇航胡张莉刘加平等. 基于机器学习回归模型的三峡大坝混凝土强度预测［J］. 材料导报202337(2): 45-53.

[23] 梁宁慧游秀菲曹郭俊等. 基于机器学习的高温后聚丙烯纤维混凝土强度预测［J］. 硅酸盐通报202140(2): 455-464.

[24] 韩斌吉坤胡亚飞等. ANN-PSO-GA模型在湿喷混凝土强度预测及配合比优化中的应用［J］. 采矿与安全工程学报202138(3): 584-591.

[25] 汪声瑞胡畔陈思宝等. 基于耦合BAS-MLP混凝土抗压强度的预测［J/OL］. 建筑材料学报: 1-15 ［2023-04-03］. http://kns.cnki.net/kcms/detail/31.1764.TU.20230308.1421.004.html.

[26] 李杨刘庆华郭天添. 基于CART-SVR模型的混凝土抗压强度预测研究［J］. 混凝土2022(8): 40-44.

[27] 吴中伟. 绿色高性能混凝土与科技创新［J］. 建筑材料学报19981(1): 1-7.

[28] 马骏. 高速公路预制T梁用高性能混凝土性能研究［J］. 中国测试202147(6): 124-130.

[29] 秦涛韩方玉光鉴淼等. 铁路桥梁用高性能混凝土的力学性能试验研究［J］. 混凝土2021(3): 134-136.

[30] 武春丽. 两部门联合推广应用高性能混凝土鼓励绿色建筑、保障房、政府投资工程率先应用［J］. 混凝土2014(4): 151.

[31] 胥悦. 高性能混凝土技术在工业建筑工程中的实际应用［J］. 工业建筑202151(8): 256.

[32] 谷坤鹏王成启. 海工高性能混凝土常用胶凝材料抗硫酸盐侵蚀性能研究［J］. 水运工程2010(12): 8-13.

[33] 王跃全薛文强. 洋山港口工程中高性能混凝土施工质量控制［J］. 水运工程2005(6): 91-95.

[34] WOLPERT D H. Stacked generalization［J］. Neural Networks19925(2): 241-259.

[35] CHEN T QGUESTRIN C. XGBoost: a scalable tree boosting system［C］//Proceedings of the 22 nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. San FranciscoCaliforniaUSA. New York: ACM2016: 785-794.

[36] PROKHORENKOVA LGUSEV GVOROBEV Aet al. CatBoost: unbiased boosting with categorical features［EB/OL］. 2017: arXiv: 1706.09516. https://arxiv.org/abs/1706.09516.

[37] BREIMAN L. Random forests［J］. Machine Language200145(1): 5-32.

[38] WIDIASARI I RNUGROHO L EWIDYAWAN. Deep learning multilayer perceptron (MLP) for flood prediction model using wireless sensor network based hydrology time series data mining［C］//2017 International Conference on Innovative and Creative Information Technology (ICITech). SalatigaIndonesia. IEEE2017: 1-5.

[39] LUNDBERG S MLEE S I. A unified approach to interpreting model predictions［C］//Proceedings of the 31st International Conference on Neural Information Processing Systems. Long BeachCaliforniaUSA. New York: ACM2017: 4768-4777.

[40] YEH I C. Modeling of strength of high-performance concrete using artificial neural networks［J］. Cement and Concrete Research199828(12): 1797-1808.

[41] YEH I C. Modeling slump flow of concrete using second-order regressions and artificial neural networks［J］. Cement and Concrete Composites200729(6): 474-480.

胡以婵, 梁铭, 谢灿荣, 解威威, 翁贻令, 池浩, 彭浩, 罗雪霜. 基于Stacking模型融合的高性能混凝土强度预测方法[J]. 硅酸盐通报, 2023, 42(11): 3914. HU Yichan, LIANG Ming, XIE Canrong, XIE Weiwei, WENG Yiling, CHI Hao, PENG Hao, LUO Xueshuang. Strength Prediction Method of High Performance Concrete Based on Stacking Model Fusion[J]. Bulletin of the Chinese Ceramic Society, 2023, 42(11): 3914.

基于Stacking模型融合的高性能混凝土强度预测方法

关于本站 Cookie 的使用提示

全站搜索

基于Stacking模型融合的高性能混凝土强度预测方法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索