基于ISSA-GRU的混凝土抗压强度预测

摘要/Abstract

摘要： 考虑到抗压强度对混凝土设计的重要影响,本文提出了改进麻雀搜索算法(ISSA)和门控循环单元(GRU)结合的ISSA-GRU预测模型,实现对高性能混凝土抗压强度的精准预测。对收集的数据集进行归一化处理后,利用基于光谱-理化值共生距离(SPXY)法对数据集进行训练集和测试集划分,采用GRU对高性能混凝土抗压强度进行回归预测,并通过引入动态惯性权重的ISSA,加强对GRU网络参数的寻优效率。结果表明,在使用相同数据样本的情况下,将ISSA-GRU模型与长短期记忆(LSTM)网络、核极限学习机(KELM)和支持向量回归(SVR)模型进行比较,其均方根误差RMSE分别降低了9.3%、37.5%、33.5%,平均绝对误差MAE分别降低了13.5%、38.5%、41.7%。同时,研究了训练集数据量和输入变量对模型预测性能的影响,研究结果表明,所提出的模型能高效寻找超参数,具有较高的预测精度和较好的适应性,为多样化原材料和混凝土特定性能的发展提供可行参考。

关键词: 高性能混凝土, 门控循环单元, 动态惯性权重, 麻雀搜索算法, 深度学习, 强度预测

Abstract: Considering the important influence of compressive strength on concrete design, an ISSA-GRU prediction model combining improved sparrow search algorithm (ISSA) and gate recurrent unit (GRU) was proposed to achieve accurate prediction of compressive strength of high-performance concrete. After normalizing the collected data set, the data set was divided into training set and testing set based on spectral-physicochemical value symbiotic distance (SPXY)method, GRU was used to predict the compressive strength of high-performance concrete, and enhances optimization efficiency of GRU network parameters by introducing ISSA with dynamic inertia weight. The results show that,in the case of using same data samples, the ISSA-GRU model is compared with the long short-term memory network (LSTM), kernel extreme learning machine (KELM) and support vector regression (SVR) models. The root mean square error (RMSE) is reduced by 9.3%, 37.5%, and 33.5%, respectively, and the mean absolute error (MAE) is reduced by 13.5%, 38.5%, and 41.7%, respectively. At the same time, the influences of the amount of training set data and input variables on prediction performance of the model were studied. The results show that the proposed model is efficient in finding GRU parameters, has high prediction accuracy and good adaptability, and provides a feasible reference for the development of diverse raw materials and specific properties of concrete.

Key words: high-performance concrete, gate recurrent unit, dynamic inertia weight, sparrow search algorithm, deep learning, strength prediction

中图分类号:

TU528.01

段妹玲, 张单, 袁锦虎, 孙爱军, 强晟. 基于ISSA-GRU的混凝土抗压强度预测[J]. 硅酸盐通报, 2023, 42(7): 2392-2400.

DUAN Meiling, ZHANG Dan, YUAN Jinhu, SUN Aijun, QIANG Sheng. Prediction of Compressive Strength of Concrete Based on ISSA-GRU[J]. BULLETIN OF THE CHINESE CERAMIC SOCIETY, 2023, 42(7): 2392-2400.

参考文献

[1] 缪昌文, 穆松. 混凝土技术的发展与展望[J]. 硅酸盐通报, 2020, 39(1): 1-11.
MIAO C W, MU S. Development and prospect of concrete technology[J]. Bulletin of the Chinese Ceramic Society, 2020, 39(1): 1-11 (in Chinese).
[2] 刘诚, 聂鑫, 汪家继, 等. 混凝土宏观本构模型研究进展[J]. 建筑结构学报, 2022, 43(1): 29-41.
LIU C, NIE X, WANG J J, et al. State-of-the-art of macroscopic constitutive models of concrete[J]. Journal of Building Structures, 2022, 43(1): 29-41 (in Chinese).
[3] LI D H, TANG Z L, KANG Q A, et al. Machine learning-based method for predicting compressive strength of concrete[J]. Processes, 2023, 11(2): 390.
[4] 王楠. 回弹法在C50～C60混凝土抗压强度检测应用中的探讨[J]. 建筑结构, 2020, 50(增刊2): 654-657.
WANG N. Discussion on application of rebound method in C50～C60 concrete compressive strength inspect[J]. Building Structure, 2020, 50(supplement 2): 654-657 (in Chinese).
[5] 郑士举, 刘辉, 蒋利学. 不同测试方式下混凝土抗压强度对比试验研究[J]. 建筑结构, 2022, 52(8): 105-111.
ZHENG S J, LIU H, JIANG L X. Comparative experimental research on concrete compression strength under different testing methods[J]. Building Structure, 2022, 52(8): 105-111 (in Chinese).
[6] 曹斐, 周彧, 王春晓, 等. 一种改进的支持向量回归的混凝土强度预测方法[J]. 硅酸盐通报, 2021, 40(1): 90-97.
CAO F, ZHOU Y, WANG C X, et al. An improved support vector regression method for concrete strength prediction[J]. Bulletin of the Chinese Ceramic Society, 2021, 40(1): 90-97 (in Chinese).
[7] 吴贤国, 刘鹏程, 陈虹宇, 等. 基于随机森林的高性能混凝土抗压强度预测[J]. 混凝土, 2022(1): 17-20+24.
WU X G, LIU P C, CHEN H Y, et al. Characteristic screening and prediction of high-performance concrete compressive strength based on random forest method[J]. Concrete, 2022(1): 17-20+24 (in Chinese).
[8] 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, 2023, 37(2): 45-53.
XU X H, HU Z L, LIU J P, et al. Concrete strength prediction of the Three Gorges Dam based on machine learning regression model[J]. Materials Reports, 2023, 37(2): 45-53 (in Chinese).
[9] 许开成, 毕丽苹, 陈梦成. 基于SPSS回归分析的锂渣混凝土抗压强度预测模型[J]. 建筑科学与工程学报, 2017, 34(1): 15-24.
XU K C, BI L P, CHEN M C. Prediction model of compressive strength of lithium slag concrete based on SPSS regression analysis[J]. Journal of Architecture and Civil Engineering, 2017, 34(1): 15-24 (in Chinese).
[10] 陈洪根, 龙蔚莹, 李昕, 等. 基于BP神经网络的粉煤灰混凝土抗压强度预测研究[J]. 建筑结构, 2021, 51(增刊2): 1041-1045.
CHEN H G, LONG W Y, LI X, et al. Prediction of compressive strength of fly ash concrete with BP neural network[J]. Building Structure, 2021, 51(supplement 2): 1041-1045 (in Chinese).
[11] 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+70.
GAO W. Influencing factors and deep learning prediction model of compression strength of recycled concrete[J]. Concrete, 2018(11): 58-61+70 (in Chinese).
[12] CHEN H G, LI X, WU Y Q, et al. Compressive strength prediction of high-strength concrete using long short-term memory and machine learning algorithms[J]. Buildings, 2022, 12(3): 302.
[13] 张静, 刘向东. 混沌粒子群算法优化最小二乘支持向量机的混凝土强度预测[J]. 吉林大学学报(工学版), 2016, 46(4): 1097-1102.
ZHANG J, LIU X D. Prediction of concrete strength based on least square support vector machine optimized by chaotic particle swarm optimization[J]. Journal of Jilin University (Engineering and Technology Edition), 2016, 46(4): 1097-1102 (in Chinese).
[14] RANJBAR I, TOUFIGH V, BOROUSHAKI M. A combination of deep learning and genetic algorithm for predicting the compressive strength of high-performance concrete[J]. Structural Concrete, 2022, 23(4): 2405-2418.
[15] XU Y, WANG J J, XIA A Q, et al. Continuous wavelet analysis of leaf reflectance improves classification accuracy of mangrove species[J]. Remote Sensing, 2019, 11(3): 254.
[16] YANG Z F, XIAO H, ZHANG L, et al. Fast determination of oxide content in cement raw meal using NIR spectroscopy with the SPXY algorithm[J]. Analytical Methods, 2019, 11(31): 3936-3942.
[17] PAN M Y, ZHOU H N, CAO J Y, et al. Water level prediction model based on GRU and CNN[J]. IEEE Access, 2020, 8: 60090-60100.
[18] HUSSAIN B, AFZAL M K, AHMAD S, et al. Intelligent traffic flow prediction using optimized GRU model[J]. IEEE Access, 2021, 9: 100736-100746.
[19] XUE J K, SHEN B. A novel swarm intelligence optimization approach: sparrow search algorithm[J]. Systems Science & Control Engineering, 2020, 8(1): 22-34.
[20] KHALEEL M I. Efficient job scheduling paradigm based on hybrid sparrow search algorithm and differential evolution optimization for heterogeneous cloud computing platforms[J]. Internet of Things, 2023, 22: 100697.
[21] MENG K, CHEN C, XIN B. MSSSA: a multi-strategy enhanced sparrow search algorithm for global optimization[J]. Frontiers of Information Technology & Electronic Engineering, 2022, 23(12): 1828-1847.
[22] LI Q Z, LI Z N, SHI Z Y, et al. Magnetic object recognition with magnetic gradient tensor system heading-line surveys based on kernel extreme learning machine and sparrow search algorithm[J]. Measurement, 2022, 203: 111967.
[23] 高林钢, 李同春, 林潮宁, 等. 基于改进鲸鱼优化算法的重力坝变形参数反演[J]. 水资源与水工程学报, 2020, 31(3): 193-199.
GAO L G, LI T C, LIN C N, et al. Parameter inversion of gravity dam deformation based on improved whale optimization algorithm[J]. Journal of Water Resources and Water Engineering, 2020, 31(3): 193-199 (in Chinese).
[24] DUA D, GRAFF C. UCI machine learning repository[DB/OL]. (2020-07-22) [2020-08-03]. http://archive.ics.uci.edu/ml.
[25] 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.
LI Y, LIU Q H, GUO T T. Prediction method of concrete compressive strength based on CART-SVR[J]. Concrete, 2022(8): 40-44 (in Chinese).
[26] 张恺, 尹志刚, 程国敏, 等. 聚羧酸减水剂及不同掺合料对混凝土耐久性的试验研究[J]. 硅酸盐通报, 2018, 37(9): 2974-2978+2984.
ZHANG K, YIN Z G, CHENG G M, et al. Experimental study on the durability of concrete with polycarboxylate superplasticizer and different admixtures[J]. Bulletin of the Chinese Ceramic Society, 2018, 37(9): 2974-2978+2984 (in Chinese).
[27] 杨泽波, 刘勇, 陈清蓉, 等.混凝土用矿物掺合料超细化作用机理[J/OL]. 材料导报, 2023(增刊1): 1-9 [2023-04-16]. http://kns.cnki.net/kcms/detail/50.1078.TB.20230207.1554.018.html.
YANG Z B, LIU Y, CHEN Q R, et al. Superrefinement mechanism of mineral admixture for concrete[J/OL]. Materials Reports, 2023(supplement 1): 1-9 [2023-04-16]. http://kns.cnki.net/kcms/detail/50.1078.TB.20230207.1554.018.html (in Chinese).
[28] LI Q F, SONG Z M. High-performance concrete strength prediction based on ensemble learning[J]. Construction and Building Materials, 2022, 324: 126694.
[29] MOON S, MUNIRA C A. Utilization of prior information in neural network training for improving 28-day concrete strength prediction[J]. Journal of Construction Engineering and Management, 2021, 147(5): 04021028.
[30] FENG D C, LIU Z T, WANG X D, et al. Machine learning-based compressive strength prediction for concrete: an adaptive boosting approach[J]. Construction and Building Materials, 2020, 230: 117000.