李鹏远,牛豪爽,刘毅豪

• 1:许昌职业技术学院
• 2:河南理工大学土木学院
• 3:许昌市数字化建造技术与装备重点实验室

摘要(Abstract)：

【目的】混凝土作为国民经济建设的基石，其抗压强度的精确预测对于工程结构的设计和安全至关重要。通过深度神经网络(DNN)模型预测混凝土抗压强度，并提出RF-NSGA-II算法以优化混凝土配合比，实现抗压强度和成本的双重优化。【方法】构建了包含不同隐藏层和神经元数量的15种DNN模型架构，评估其性能并选取最佳模型，采用超参数优化策略和贝叶斯优化策略，提升DNN模型的预测性能，比较DNN模型与支持向量回归(SVR)和随机森林(RF)模型的性能。基于RF-NSGA-II算法，优化混凝土配合比，以满足强度要求和成本控制。【结果】研究结果显示，最优模型为3个隐藏层和64个隐藏单元(3L-64u)的DNN模型，经过优化DNN模型在MAE和MSE上分别降低了18%和27%,优化后的DNN模型相比SVR和RF模型在MAE和MSE上分别减少了4%和12%、11%和15%。【结论】通过案例验证，DNN3-L64u-BOP模型预测结果与试验值吻合良好，RF-NSGA-II算法优化的混凝土配合比方案有效降低成本，满足工程强度要求。基于贝叶斯优化的DNN模型能较好地预测混凝土抗压强度，RF-NSGA-II算法在多目标优化混凝土配合比方面展示出优异的性能，具有实际工程应用价值。

关键词(KeyWords)： 混凝土;DNN;抗压强度;预测;优化;力学性能;影响因素;深度学习

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(42372331)

作者(Author): 李鹏远,牛豪爽,刘毅豪

DOI: 10.13928/j.cnki.wrahe.2025.04.016

参考文献(References)：

• [1] 陈璋，陈徐东，白银，等.粉煤灰对低胶材自密实混凝土强度及孔结构的影响[J].水利水电技术(中英文),2023,54(2):179-189.CHEN Zhang,CHEN Xudong,BAI Yin,et al.Influence of fly ash on the strength and pore structure of low-binder self-compacting concrete[J].Water Resources and Hydropower Engineering,2023,54(2):179-189.
• [2] 孟浩，杨维武，刘海峰.沙漠砂替代率对冻融环境下混凝土抗压性能的影响[J].宁夏工程技术，2023,22(1):49-55.MENG Hao,YANG Weiwu,LIU Haifeng.Influence of desert sand replacement rate on compressive performance of concrete in freeze-thaw environment[J].Ningxia Engineering Technology,2023,22(1):49-55.
• [3] 覃源，梁达，王玉龙，等.冷缝存在条件下混凝土力学性能试验研究[J].排灌机械工程学报，2023,41(7):677-681.QIN Yuan,LIANG Da,WANG Yulong,et al.Experimental study on mechanical properties of concrete with cold joints[J].Journal of Drainage and Irrigation Machinery Engineering,2023,41(7):677-681.
• [4] 张凯，李北星，李广，等.片麻岩石粉掺入方式对混凝土性能的影响[J].水利水电技术(中英文),2023,54(1):187-198.ZHANG Kai,LI Beixing,LI Guang,et al.Effect of incorporation method of gneiss powder on the performance of concrete[J].Water Resources and Hydropower Engineering,2023,54(1):187-198.
• [5] 侯艳娟，单世龙，李正宁，等.玻化粉煤灰微珠混凝土基本力学性能及水化热释放研究[J].铁道勘察，2023,49(4):7-12.HOU Yanjuan,SHAN Shilong,LI Zhengning,et al.Research on the basic mechanical properties and heat release of hydration of glassy fly ash microbead concrete[J].Railway Investigation and Surveying,2023,49(4):7-12.
• [6] 李温，王海龙，张佳豪，等.胶粉煤矸石混凝土力学特性及微观结构试验研究[J].排灌机械工程学报，2023,41(2):140-146.LI Wen,WANG Hailong,ZHANG Jiahao,et al.Experimental study on mechanical properties and microstructure of powdered coal gangue concrete[J].Journal of Drainage and Irrigation Machinery Engineering,2023,41(2):140-146.
• [7] MAJID K,FAISAL M J.Towards sustainable construction:Machine learning based predictive models for strength and durability characteristics of blended cement concrete[J].Materials Today Communications,2023,37:100-115.
• [8] GHOLAMREZA P,ZHONG T,NARIMAN S,et al.Using artificial intelligence methods to predict the compressive strength of concrete containing sugarcane bagasse ash[J].Construction and Building Materials,2023,409:124-138.
• [9] 许欣，马魁.基于不同掺量的PVA纤维混凝土力学性能研究[J].科技与创新，2023(2):83-85.XU Xin,MA Kui.Study on mechanical properties of PVA fiber concrete with different dosages[J].Science and Technology Innovation,2023(2):83-85.
• [10] 钱觉时，黄煜镔.混凝土强度尺寸效应的研究进展[J].混凝土与水泥制品，2003(3):1-5.QIAN Jueshi,HUANG Yubin.Research progress on the size effect of concrete strength[J].Concrete and Cement Products,2003(3):1-5.
• [11] 李清富，孙振华，张海洋.粉煤灰和硅粉对混凝土强度影响的试验研究[J].混凝土，2011(5):77-79.LI Qingfu,SUN Zhenhua,ZHANG Haiyang.Experimental study on the effect of fly ash and silica fume on the strength of concrete[J].Concrete,2011(5):77-79.
• [12] 张朝辉，王沁芳，杨娟.透水混凝土强度和透水性影响因素研究[J].混凝土，2008(3):7-9.ZHANG Zhaohui,WANG Qinfang,YANG Juan.Study on factors affecting strength and permeability of pervious concrete[J].Concrete,2008(3):7-9.
• [13] 李风增.基于人工神经元网络模型预测混凝土抗压强度[J].粉煤灰综合利用，2023,37(3):126-133.LI Fengzeng.Prediction of compressive strength of concrete based on artificial neural network model[J].Fly Ash Comprehensive Utilization,2023,37(3):126-133.
• [14] CHOPRA P,SHARMA R K,CHOPRA,M T.Comparison of machine learning techniques for the prediction of compressive strength of concrete[J].Advances in Civil Engineering,2018,2018:1-11.
• [15] 贾金青，胡玉龙，王东来，等.混凝土抗压强度与孔隙率关系的研究[J].混凝土，2015(10):56-59.JIA Jinqing,HU Yulong,WANG Donglai,et al.Study on the relationship between compressive strength and porosity of concrete[J].Concrete,2015(10):56-59.
• [16] 方涛，王俊锋，张鑫月，等.基于人工神经网络的不同混凝土实时强度预测研究[J].混凝土，2016(12):38-43.FANG Tao,WANG Junfeng,ZHANG Xinyue,et al.Real-time strength prediction of different concretes based on artificial neural network[J].Concrete,2016(12):38-43.
• [17] 陈云龙，马菊荣，刘海峰，等.掺粉煤灰、沙漠砂高强混凝土抗压强度研究[J].混凝土，2014(7):80-84.CHEN Yunlong,MA Jurong,LIU Haifeng,et al.Study on compressive strength of high-strength concrete with fly ash and desert sand[J].Concrete,2014(7):80-84.
• [18] 肖建庄，林壮斌，朱军.再生骨料级配对混凝土抗压强度的影响[J].四川大学学报(工程科学版),2014,46(4):154-160.XIAO Jianzhuang,LIN Zhuangbin,ZHU Jun.Effect of recycled aggregate gradation on compressive strength of concrete[J].Journal of Sichuan University (Engineering Science Edition),2014,46(4):154-160.
• [19] 王振波，周潮，王家赫，等.高延性水泥基材料(ECC)开裂后渗透性研究综述[J].水利发展研究，2024,24(6):59-64.WANG Zhenbo,ZHOU Chao,WANG Jiahe,et al.Summary of research on the permeability of Engineered Cementitious Composites(ECC) after cracking[J].Water Resources Development Research,2024,24(6):59-64.
• [20] 徐国强，苏幼坡，韩佃利，等.基于BP神经网络的绿色混凝土抗压强度预测模型[J].混凝土，2013(2):33-35.XU Guoqiang,SU Youpo,HAN Dianli,et al.Prediction model of compressive strength of green concrete based on BP neural network[J].Concrete,2013(2):33-35.
• [21] 赵悟，王博，赵利军，等.改善搅拌过程提高混凝土强度[J].长安大学学报(自然科学版),2015,35(1):148-153.ZHAO Wu,WANG Bo,ZHAO Lijun,et al.Improving concrete strength by optimizing the mixing process[J].Journal of Chang′an University (Natural Science Edition),2015,35(1):148-153.
• [22] 陆岸典，唐欣薇，严振瑞，等.复合衬砌结构的预应力混凝土配比试验研究[J].水力发电学报，2022,41(11):10-20.LU Andian,TANG Xinwei,YAN Zhenrui,et al.Experimental study on mix proportion of prestressed concrete for composite lining structure[J].Journal of Hydroelectric Engineering,2022,41(11):10-20.
• [23] 朱正国，崔振伟，马超义，等.减震初期支护用钢纤维泡沫混凝土材料性能优化[J].中国铁道科学，2022(3):43-50.ZHU Zhengguo,CUI Zhenwei,MA Chaoyi,et al.Performance optimization of steel fiber foam concrete materials for initial support in shock absorption[J].China Railway Science,2022(3):43-50.
• [24] 曹检云，刘国坤，孙剑峰.大跨度悬臂浇筑钢筋混凝土拱桥多目标索力优化研究[J].公路工程，2022,47(3):30-34.CAO Jianyun,LIU Guokun,SUN Jianfeng.Multi-objective cable force optimization for long-span cantilever cast-in-place reinforced concrete arch bridges[J].Highway Engineering,2022,47(3):30-34.
• [25] ZAVALA G R,NEBRO A J,LUNA F,et al.A survey of multi-objective metaheuristics applied to structural optimization[J].Structural and Multidisciplinary Optimization,2014,49:
• [26]ZTA? A,?ZBAY E.Prediction and multi-objective optimization of high-strength concrete parameters via soft computing approaches[J].Expert Systems with Applications,2009,36(3):6145-6155.
• [27] 张海波，管学茂，勾密峰，等.废旧橡胶粒径对水泥砂浆和混凝土强度影响研究[J].硅酸盐通报，2012,31(4):931-934.ZHANG Haibo,GUAN Xuemao,GOU Mifeng,et al.Study on the effect of waste rubber particle size on the strength of cement mortar and concrete[J].Bulletin of the Chinese Ceramic Society,2012,31(4):931-934.
• [28] 李欣海.随机森林模型在分类与回归分析中的应用[J].应用昆虫学报，2013,50(4):1190-1197.LI Xinhai.Application of random forest model in classification and regression analysis[J].Journal of Applied Entomology,2013,50(4):1190-1197.
• [29] 方匡南，吴见彬，朱建平，等.随机森林方法研究综述[J].统计与信息论坛，2011,26(3):32-38.FANG Kuangnan,WU Jianbin,ZHU Jianping,et al.A review of random forest methods[J].Statistics and Information Forum,2011,26(3):32-38.
• [30] 姚登举，杨静，詹晓娟.基于随机森林的特征选择算法[J].吉林大学学报(工学版),2014,44(1):137-141.YAO Dengju,YANG Jing,ZHAN Xiaojun.Feature selection algorithm based on random forest[J].Journal of Jilin University (Engineering Edition),2014,44(1):137-141.
• [31] 李启明，喻泽成，余波，等.钢筋混凝土柱地震破坏模式判别的两阶段支持向量机方法[J].工程力学，2022,39(2):148-158.LI Qiming,YU Zecheng,YU Bo,et al.Two-stage SVM method for seismic damage mode discrimination of reinforced concrete columns[J].Engineering Mechanics,2022,39(2):148-158.
• [32] 熊先勇，曾亚林，付慧建，等.基于支持向量机和改进粒子群算法的钢管混凝土拱桥可靠度分析[J].公路工程，2023,48(2):55-61.XIONG Xianyang,ZENG Yalin,FU Huijian,et al.Reliability analysis of concrete-filled steel tube arch bridges based on SVM and improved particle swarm optimization[J].Highway Engineering,2023,48(2):55-61.
• [33] 侍莘莘.基于支持向量机的建筑工程混凝土质量可靠性分析[J].安徽建筑，2024,31(1):77-79.SHI Xinxin.Reliability analysis of concrete quality in construction engineering based on SVM[J].Anhui Architecture,2024,31(1):77-79.
• [34] 高蔚.基于深度学习的再生混凝土抗压强度预测[J].混凝土，2018(11):58-61.GAO Wei.Prediction of compressive strength of recycled concrete based on deep learning[J].Concrete,2018(11):58-61.
• [35] 王竣禾，姜勇.基于深度强化学习的动态装配算法[J].智能系统学报，2023,18(1):2-11.WANG Junhe,JIANG Yong.Dynamic assembly algorithm based on deep reinforcement learning[J].Journal of Intelligent Systems,2023,18(1):2-11.
• [36] 章玉容，余威龙，麻雪晴，等.基于深度学习的粉煤灰混凝土氯离子浓度预测[J].北京工业大学学报，2023,49(2):205-212.ZHANG Yuyong,YU Weilong,MA Xueqing,et al.Prediction of chloride ion concentration in fly ash concrete based on deep learning[J].Journal of Beijing University of Technology,2023,49(2):205-212.
• [37] 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，2005(6):677-681.JI Tao,LIN Tingwei,LIN Xujian.Prediction method for compressive strength of concrete based on artificial neural network[J].Journal of Building Materials,2005(6):677-681.
• [38] 申嘉荣，徐千军.建筑材料微观结构的人工识别研究[J].水利发展研究，2023,23(10):72-79.SHEN Jiarong,XU Qianjun.Research of artificial identification of microstructure of building materials[J].Water Resources Development Research,2023,23(10):72-79.
• [39] 冷发光，冯乃谦，邢峰.提高普通混凝土强度和耐久性的正交试验研究[J].混凝土，2000(1):18-23.LENG Faguang,FENG Naiqian,XING Feng.Orthogonal experiment study on improving strength and durability of ordinary concrete[J].Concrete,2000(1):18-23.
• [40] 梁子超，李智炜，赖铿，等.10折交叉验证用于预测模型泛化能力评价及其R软件实现[J].中国医院统计，2020,27(4):289-292.LIANG Zichao,LI Zhiwei,LAI Keng,et al.Evaluation of generalization ability of prediction model using 10-fold cross-validation and its implementation in R software[J].Chinese Hospital Statistics,2020,27(4):289-292.
• [41] 胡局新，张功杰.基于K折交叉验证的选择性集成分类算法[J].科技通报，2013,29(12):115-117.HU Juxin,ZHANG Gongjie.Selective ensemble classification algorithm based on K-fold cross-validation[J].Science and Technology Bulletin,2013,29(12):115-117.
• [42] 王重仁，韩冬梅.基于超参数优化和集成学习的互联网信贷个人信用评估[J].统计与决策，2019,35(1):87-91.WANG Chongren,HAN Dongmei.Internet credit personal credit assessment based on hyperparameter optimization and ensemble learning[J].Statistics and Decision,2019,35(1):87-91.
• [43] 崔佳旭，杨博.贝叶斯优化方法和应用综述[J].软件学报，2018,29(10):3068-3090.CUI Jiaxu,YANG Bo.Review of Bayesian optimization methods and applications[J].Journal of Software,2018,29(10):3068-3090.
• [44] 丁昌伟，王新，陈同俊，等.贝叶斯优化的XGBoost在小断层地震解释中的应用[J].煤炭学报，2023,48(6):2530-2539.DING Changwei,WANG Xin,CHEN Tongjun,et al.Application of Bayesian optimized XGBoost in small fault earthquake interpretation[J].Journal of China Coal Society,2023,48(6):2530-2539.
• [45] 吴贤国，杨赛，王成龙，等.基于RF-NSGA-II的建筑能耗多目标优化[J].土木工程与管理学报，2021,38(4):263-270.WU Xianguo,YANG Sai,WANG Chenglong,et al.Multi-objective optimization of building energy consumption based on RF-NSGA-II[J].Journal of Civil Engineering and Management,2021,38(4):263-270.
• [46] 向芷恒，王秉哲，雪景州，等.基于NSGA-II算法解决多目标优化实际应用的研究[J].应用数学进展，2023,12(10):4195-4207.XIANG Zhiheng,WANG Bingzhe,XUE Jingzhou,et al.Research on solving practical multi-objective optimization problems based on NSGA-II algorithm[J].Advances in Applied Mathematics,2023,12(10):4195-4207.