基于不同机器学习方法的UHPC抗压强度预测
Prediction of UHPC Compressive Strength Based on Different Machine Learning Methods
超高性能混凝土(ultra-high performance concrete,简称UHPC)是一种新型水泥基复合材料,在抗压强度、韧性、延性和耐久性等方面均表现出远超常规混凝土的卓越性能。然而,目前尚缺乏系统化的UHPC配比设计规范,工程实践和科研领域主要依赖经验性结论进行配比设计。本文通过对UHPC相关文献的调研,收集了300组UHPC配合比及材性试验数据,建立包含水泥、粉煤灰、硅灰、石英砂、减水剂、钢纤维和水的用量共7个影响因素及此配比下UHPC抗压强度实测值的数据库。基于此,本文采用支持向量回归(SVR)、反向传播神经网络(BPNN)、随机森林(RF)及极限学习机(ELM)等机器学习方法,开展了UHPC抗压强度的预测模型训练,并基于机器学习方法提出了一种针对指定抗压强度的UHPC配比设计方法。研究结果表明,以上机器学习方法均能根据UHPC各组分的用量及配比关系对抗压强度进行较为准确的初步预测,训练集的决定系数R²均超过0.8,预测精度较高。通过该配比设计方法,研究者能够有效获得满足特定抗压强度要求的UHPC配比方案。
Ultra-High Performance Concrete (UHPC) is a high-performance cementitious composite material characterized by ultra-high compressive strength, superior toughness, enhanced ductility, and excellent durability. However, a systematic and standardized framework for UHPC mix design remains lacking. In both engineering applications and scientific research, mix proportioning still heavily relies on empirical approaches.To address this gap, this study compiled a comprehensive database consisting of 300 mix designs and corresponding experimental material properties from a thorough review of the existing UHPC literature. The database incorporates seven key influencing factors: cement content, fly ash content, silica fume content, quartz sand content, superplasticizer dosage, steel fiber content, and water content, along with their experimentally measured compressive strengths.Based on this dataset, several machine learning algorithms—including Random Forest (RF), Extreme Learning Machine (ELM), Support Vector Regression (SVR), and Backpropagation Neural Networks (BPNN)—were employed to develop predictive models for the compressive strength of UHPC. Furthermore, a machine learning-driven mix design methodology was proposed to achieve targeted compressive strength.The results demonstrate that all adopted machine learning models can provide reasonably accurate preliminary predictions of compressive strength based on the constituent proportions, with the coefficient of determination (R²) of the training set exceeding 0.8, indicating satisfactory predictive performance. The proposed method effectively generates UHPC mix proportions that meet specified strength requirements, offering a data-driven alternative to conventional empirical design.
| [1] |
YOO D Y,YOON Y S.A review on structure behavior,design,and application of ultra-high-performance fiber-reinforced concrete[J].Intertional Journal of Concrete Structure and Materials,2016,10(2):125-142. |
| [2] |
欧阳利军,石永超,丁斌,基于机器学习预测超高性能混凝土的抗压强度[J].混凝土与水泥制品,2025,347(3):1-7. |
| [3] |
OUYANG Lijun,SHI Yongchao,DING Bin,et al.Prediction of compressive strength of ultra-high performance concrete based on machine learning[J]. Concrete and Cement Products,2025,347(3):1-7.(in Chinese) |
| [4] |
刘凯华,郑佳凯,谢维力,基于机器学习的再生混凝土配合比设计方法[J].湖南大学学报(自然科学版),2023,50(9):88-96. |
| [5] |
LIU Kaihua,ZHENG Jiakai,XIE Weili,et al.Mix proportion design method of recycled concrete based on machine learning[J].Journal of Hunan University (Natural Science Edition),2023,50(9):88-96.(in Chinese) |
| [6] |
NADERPOUR H,RAFIEAN A H,FAKHARIAN P.Compressive strength prediction of environmentally friendly concrete using artificial neural networks[J].Journal of Building Engineering,2018(16):213-219. |
| [7] |
刘兵.基于机器学习的再生混凝土强度预测和LAA-LCC生态效率评价[D].北京:北京交通大学,2021. |
| [8] |
LIU Bing.Recycled aggregate concrete strength prediction and eco-efficiency evaluation of LCA-LCC Based on Machine Learning[D].Beijing:Beijing Jiaotong University,2021.(in Chinese) |
| [9] |
胡定宇.基于机器学习和优化算法的混凝土性能预测及配合比设计[D].重庆:重庆交通大学,2022. |
| [10] |
HU Dingyu.Concrete performance prediction and mix proportion design based on machine learning and optimization algorithm[D].Chongqing:Chongqing Jiaotong University,2022.(in Chinese) |
| [11] |
鄢亦斌,刘开志,高立强,基于机器学习的超高性能混凝土性能预测[J].混凝土与水泥制品,2024,343(11):49-53,59.DOI:10.19761/j.1000-4637. 2024.11.049.06. |
| [12] |
YAN Yibin,LIU Kaizhi,GAO Liqiang,et al.Performance prediction of ultra-high performance concrete based on machine learning[J].Concrete and Cement Products,2024,343(11):49-53,59.DOI:10.19761/j.1000-4637.2024.11.049.06.(in Chinese) |
| [13] |
陈玉超,潘杰,刘明涛,超高减水型聚羧酸减水剂的合成及其在UHPC中的应用研究[J].新型建筑材料,2023,50(3):127-130. |
| [14] |
CHEN Yuchao,PAN Jie,LIU Mingtao,et al.Synthesis of ultra-high water-reducing polycarboxylate superplasticizer and its application in UHPC[J].New Building Materials,2023,50(3):127-130.(in Chinese) |
| [15] |
李永.本土化材料对UHPC力学性能及微观结构的影响[D].兰州:兰州交通大学,2022. |
| [16] |
LI Yong.Influence of localized materials on mechanical properties and microstructure of UHPC[D].Lanzhou:Lanzhou Jiaotong University,2022.(in Chinese) |
| [17] |
许文英,尹江涛,张利俊,UHPC用超高性能减水剂的合成与性能研究[J].新型建筑材料,2022,49(1):81-84. |
| [18] |
XU Wenying,YIN Jiangtao,ZHANG Lijun,et al.Study on synthesis and properties of ultra-high performance water reducer for UHPC[J].New Building Materials,2022,49(1):81-84.(in Chinese) |
| [19] |
余睿,范定强,孙美娟,钢纤维掺量及其3D空间结构对超高性能混凝土性能的影响[J].硅酸盐学报,2021,49(11):2313-2321. |
| [20] |
YU Rui,FAN Dingqiang,SUN Meijuan,et al.The influence of steel fiber content and its 3D spatial structure on the performance of ultra-high performance concrete[J].Journal of Silicate,2021,49(11):2313-2321.(in Chinese) |
| [21] |
刘奎启.超高性能混凝土减水剂优选试验研究[J].市政技术,2021,39(12):138-141,146. |
| [22] |
LIU Kuiqi.Experimental study on optimization of super high performance concrete water reducing agent[J].Municipal Technology,2021,39(12):138-141,146.(in Chinese) |
| [23] |
樊俊江,於林锋,韩建军.配比参数对UHPC流动性及抗压强度的影响试验研究[J].新型建筑材料, 2019,46(4):5-8. |
| [24] |
FAN Junjiang,YU Linfeng,HAN Jianjun.Experimental study on the influence of mixing parameters on the fluidity and compressive strength of UHPC[J].New Building Materials,2019,46(4):5-8.(in Chinese) |
| [25] |
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):176-183,191. |
| [26] |
CHEN Qing,MA Rui,JIANG Zhengwu,et al.Compressive strength prediction and mix proportion design of UHPC based on GA-BP neural network[J].Journal of Building Materials,2020,23(1):176-183,191.(in Chinese) |
| [27] |
高宝成,陶博文.基于SVR算法的混凝土强度预测[J].城市住宅,2019,26(4):143-146. |
| [28] |
GAO Baocheng,TAO Bowen.Prediction of concrete strength based on SVR algorithm[J].Urban housing, 2019,26(4):143-146.(in Chinese) |
| [29] |
HUANG G,ZHU Q,SIEW C.Extreme learning machine:Theory and applications[J].Neurocomputing,2006,70:489-501. |
| [30] |
曹恩华,包腾飞,胡绍沛,基于变量筛选优化极限学习机的混凝土坝变形预测模型[J].长江科学院院报,2022,39(7):59-65. |
| [31] |
CAO Enhua,BAO Tengfei,HU Shaopei,et al.Deformation prediction model of concrete dam based on variable screening optimization extreme learning machine[J].Journal of Yangtze River Academy of Sciences,2022,39(7):59-65.(in Chinese) |
国家自然科学基金(52378186)
/
| 〈 |
|
〉 |