Loading [MathJax]/jax/output/SVG/jax.js

基于基因表达式编程的FRP约束混凝土极限轴向应变预测

邓楚兵, 薛新华

邓楚兵, 薛新华. 基于基因表达式编程的FRP约束混凝土极限轴向应变预测[J]. 复合材料学报, 2023, 40(1): 510-520. DOI: 10.13801/j.cnki.fhclxb.20220125.002
引用本文: 邓楚兵, 薛新华. 基于基因表达式编程的FRP约束混凝土极限轴向应变预测[J]. 复合材料学报, 2023, 40(1): 510-520. DOI: 10.13801/j.cnki.fhclxb.20220125.002
DENG Chubing, XUE Xinhua. Prediction of ultimate axial strain of FRP-confined concrete based on gene expression programming[J]. Acta Materiae Compositae Sinica, 2023, 40(1): 510-520. DOI: 10.13801/j.cnki.fhclxb.20220125.002
Citation: DENG Chubing, XUE Xinhua. Prediction of ultimate axial strain of FRP-confined concrete based on gene expression programming[J]. Acta Materiae Compositae Sinica, 2023, 40(1): 510-520. DOI: 10.13801/j.cnki.fhclxb.20220125.002

基于基因表达式编程的FRP约束混凝土极限轴向应变预测

详细信息
    通讯作者:

    薛新华,博士,教授,博士生导师,研究方向为岩土工程 E-mail: xuexinhua@scu.edu.cn

  • 中图分类号: TU37

Prediction of ultimate axial strain of FRP-confined concrete based on gene expression programming

  • 摘要: 纤维增强树脂复合材料(FRP)以其质量轻、强度高、耐腐蚀和施工方便等优势被广泛应用于混凝土结构性能提升和受损构件加固中。FRP约束混凝土的极限条件是选择FRP种类、选择FRP厚度及确定包裹层数等必须要考虑的因素,现有极限应力模型的预测结果能够较好反地映真实情况,而现有极限轴向应变模型的预测精度偏低,故本文对极限轴向应变进行了研究。由于影响FRP约束混凝土极限轴向应变的因素较多,许多研究人员提出的模型在输入参数的选择上存在较大差异,故本文在通过基因表达式编程建立极限轴向应变模型的同时还探讨了不同输入形式对模型预测精度的影响。采用决定系数及平均绝对误差等5种统计指标对模型预测结果进行评价,并将其与现有模型进行对比分析。研究结果表明:原始数据和新数据组合的输入形式对应的模型具有最高的预测精度,因此在模型输入参数的选择上不能仅考虑原始数据或者新数据;与其他研究人员所提模型相比,本文所提模型预测精度更高,其决定系数为0.893,平均绝对误差等指标均在0.35以下。
    Abstract: Fiber reinforced polymer (FRP) is widely used in enhancing the performance of concrete structures and strengthening damaged components due to its advantages of light weight, high strength, corrosion resistance and convenient construction. The ultimate conditions of FRP-confined concrete are the important factors that must be considered in the selection of FRP types, FRP thickness and the number of covering layer. The prediction results of the existing ultimate stress model can better reflect the real situation, while the prediction accuracy of the existing ultimate axial strain model is low, so the ultimate axial strain was studied. Since there are many factors that affect the ultimate axial strain of FRP-confined concrete, the models proposed by many researchers have large differences in the choice of input parameters. Therefore, the influence of different input forms on the prediction accuracy of ultimate axial strain model was discussed while the ultimate axial strain model was established by gene expression programming. Five statistical indicators such as coefficient of determination and mean absolute error were used to evaluate the prediction results of model, which was compared with the existing prediction models. The research results show that the model corresponding to the input form of the combination of original data and new data has the highest prediction accuracy, so the selection of model input parameters should not only consider the original data or new data. Compared with the models proposed by other researchers, the prediction accuracy of the model proposed in this article is the highest. The coefficient of determination is 0.893, and the mean absolute error and other indicators are all below 0.35.
  • 图  1   基因表达式编程(GEP)语言示意图

    Figure  1.   Diagram of gene expression programming (GEP) language

    d0—Ultimate axial strain of unconfined concrete; d1—Stiffness ratio; d2—Strain ratio; d3—Ultimate axial strain of FRP-confined concrete; 3Rt—Function symbol; c2—Constant

    图  2   FRP约束混凝土极限轴向应变模型最优参数确定

    Figure  2.   Determination of optimal parameters of ultimate axial strain model of FRP-confined concrete

    图  3   不同输入形式下的FRP约束混凝土极限轴向应变模型的预测结果与试验结果对比

    Figure  3.   Comparison between experimental and prediction results of ultimate axial strain models of FRP-confined concrete under different input forms

    图  4   模型C (GEP模型)的表达式树(Sub-ET)

    Figure  4.   Expression trees (Sub-ET) of model C (GEP model)

    Inv—Inverse; c0, c3, c7—Constant

    图  5   FRP约束混凝土GEP模型的参数敏感性分析结果

    Figure  5.   Parameter sensitivity analysis results of GEP model for FRP-confined concrete

    图  6   各FRP约束混凝土极限轴向应变模型的预测结果与试验结果对比

    Figure  6.   Comparison between experimental and prediction results of ultimate axial strain models of FRP-confined concrete

    表  1   FRP约束混凝土的极限轴向应变模型

    Table  1   Ultimate axial strain models of FRP-confined concrete

    Model and yearUltimate axial strain
    Ahmad et al [23]
    (2020)
    εcu=(1.85+7.46ρ1.171ερ0.71k)εco
    Yu et al[24]
    (2011)
    εcu=0.0033+0.6(Elfco)0.8(εh,rup)1.45
    Benzaid et al [25]
    (2010)
    εcu=2εco+7.6flfcoεco
    Teng et al [26]
    (2009)
    εcu=1.75εco+6.5εcoρ1.45ερ0.8k
    Al-Tersawy et al [27] (2007) εcu=2εco+8.16(flfco)0.34εco
    Ilki et al [28]
    (2004)
    εcu=εco+20(flfco)0.5εco
    Xiao et al[29]
    (2000)
    εcu=εh,rup0.00057(Elfco)0.8
    Samaan et al [30]
    (1998)
    εcu=fcc0.872fco0.371fl6.258245.61fco0.2+1.3456EFRPTD
    Mander et al [31]
    (1988)
    εcu=(1+5(fccfco1))εco
    Notes:fcc—Peak strength of FRP-confined concrete; εcu—Ultimate axial strain of FRP-confined concrete; fco—Peak strength of unconfined concrete; εco—Peak strain of unconfined concrete; D—Diameter of the concrete core; T—Total thickness of the FRP jacket; EFRP—Elastic modulus of the FRP jacket; εh,rup—Hoop rapture strain of the FRP jacket; fl—Confining stress; ρk—Stiffness ratio; ρε—Strain ratio; El—Restraint stiffness; See Table 2 for variable units.
    下载: 导出CSV

    表  2   FRP约束普通混凝土圆柱体试验数据的统计参数

    Table  2   Statistical parameters of test data of FRP-confined normal concrete cylinder

    ParameterMinMaxMedianAverageStandard
    deviation
    D/mm100200152146.08017.816
    H/mm200788305314.10694.341
    fco/MPa26.255.24240.5777.100
    εco/%0.160.420.2390.2490.047
    Ec/GPa24.2135.1429.7729.6922.731
    T/mm0.115.210.4950.7940.882
    EFRP/GPa13.6629.6105154.604113.36
    εh,rup/%0.193.091.081.2080.504
    ρk2.55538.1149.32410.6837.153
    ρε0.0100.3690.0496.4305.659
    fl/MPa0.90513.4354.3334.9222.154
    εcu/%0.45.551.611.8301.005
    Notes: H—Height of the concrete core; Ec—Elastic modulus of the concrete core.
    下载: 导出CSV

    表  3   FRP约束普通混凝土圆柱体试验数据的皮尔逊相关性分析结果

    Table  3   Results of Pearson correlation analysis of test data of FRP-confined normal concrete cylinder

    ρkρεflεcu
    D−0.223**−0.064−0.017−0.121
    H−0.041−0.131−0.108−0.127
    fco−0.211**0.0390.136−0.029
    εco−0.163*−0.181*−0.180*0.236**
    Ec−0.198**0.0890.151−0.003
    T0.161*0.356**0.553**0.297**
    EFRP0.509**−0.544**−0.031−0.165*
    εh,rup−0.395**0.926**0.0860.377**
    ρk1−0.334**0.669**0.379**
    ρε−0.334*10.165*0.301**
    fl0.669**0.165*10.653**
    εcu0.379**0.301*0.653**1
    Notes: **—Correlation is significant at the 0.01 level (2-tailed);
    *—Correlation is significant at the 0.05 level (2-tailed).
    下载: 导出CSV

    表  4   FRP约束混凝土极限轴向应变模型的参数设置

    Table  4   Parameters setting of ultimate axial strain model of FRP-confined concrete

    Parameter typesSettingParameter typesSetting
    Population
    size
    50Gene transposition rate0.3
    Head
    length
    12Gene recombination rate0.3
    Gene
    number
    3One-point recombination rate0.4
    Chromosome length45Two-point recombination rate0.4
    Connection functionAddition (+)Insertion sequence transposition rate0.3
    Mutation
    rate
    0.044Root insertion sequence transposition rate0.3
    下载: 导出CSV

    表  5   FRP约束混凝土极限轴向应变模型的输入形式

    Table  5   Input informs of ultimate axial strain model of FRP-confined concrete

    No.ModelUltimate axial strain εcu
    1Aεcu=f(D,H,fco,εco,Ec,T,EFRP,εh,rup)
    2Bεcu=f(fl,ρε,ρk)
    3Cεcu=f(εco,ρε,ρk)
    4Dεcu=f(fco,εco,fl)
    5Eεcu=f(εco,T,EFRP,εh,rup,fl,ρε,ρk)
    下载: 导出CSV

    表  6   不同输入形式下的FRP约束混凝土极限轴向应变模型的预测结果

    Table  6   Prediction results of ultimate axial strain models of FRP-confined concrete under different input forms

    ModelR2MAERRSERMSEMAPE
    A 0.487 0.414 0.721 0.682 0.257
    B 0.561 0.533 0.841 0.796 0.299
    C 0.893 0.228 0.334 0.316 0.160
    D 0.705 0.362 0.543 0.514 0.260
    E 0.740 0.543 0.762 0.721 0.353
    Notes: R2—Determination coefficient; MAE—Mean absolute error; RRSE—Relative square root error; RMSE—Root mean square error; MAPE—Mean absolute percentage error.
    下载: 导出CSV

    表  7   FRP约束混凝土GEP模型的参数重要性分析结果

    Table  7   Parameter importance analysis results of GEP model for FRP-confined concrete

    ModelParameterR2Conclusion

    GEP
    Without εco 0.496
    ρk>ρε>εco
    Without ρk 0.112
    Without ρε 0.397
    Without εco 0.522
    Teng et al [26] Without ρk 0.118 ρk>ρε>εco
    Without ρε 0.245
    下载: 导出CSV

    表  8   各FRP约束混凝土极限轴向应变模型的预测结果

    Table  8   Prediction results of ultimate axial strain models of FRP-confined concrete

    ModelR2MAERRSERMSEMAPE
    GEP 0.893 0.228 0.334 0.316 0.160
    Ahmad et al[23] 0.743 0.327 0.524 0.496 0.206
    Yu et al[24] 0.611 0.394 0.721 0.683 0.199
    Benzaid et al[25] 0.664 0.814 1.129 1.068 0.410
    Teng et al[26] 0.701 0.357 0.563 0.533 0.208
    Al-Tersawy et al[27] 0.550 0.513 0.767 0.726 0.348
    Ilki et al[28] 0.682 0.881 1.029 0.974 0.671
    Xiao et al[29] 0.572 0.412 0.742 0.702 0.213
    Samaan et al[30] 0.478 1.713 2.061 1.950 0.971
    Mander et al[31] 0.681 0.379 0.544 0.613 0.204
    下载: 导出CSV
  • [1]

    GULER S. Axial behavior of FRP-wrapped circular ultra-high performance concrete specimens[J]. Structural Engineering and Mechanics,2014,50(6):709-722. DOI: 10.12989/sem.2014.50.6.709

    [2]

    WANG W, WU C, LIU Z, et al. Compressive behavior of ultra-high performance fiber rein-forced concrete (UHPFRC) confined with FRP[J]. Composite Structures,2018,204:419-437. DOI: 10.1016/j.compstruct.2018.07.102

    [3]

    ZHOU Y, LIU X, XING F, et al. Axial compressive behavior of FRP-confined light weight aggregate concrete: An experimental study and stress-strain relation model[J]. Construction and Building Materials,2016,119:1-15. DOI: 10.1016/j.conbuildmat.2016.02.180

    [4] 李家齐, 李诗娴, 胡皓, 等. FRP约束玄武岩纤维再生混凝土圆柱力学性能试验研究[J]. 混凝土, 2018(12):57-61. DOI: 10.3969/j.issn.1002-3550.2018.12.015

    LI Jiaqi, LI Shixian, HU Hao, et al. Experimental study and finite element analysis on mechanical properties of FRP confined basalt fiber reinforced recycled concrete circular columns[J]. Concrete,2018(12):57-61(in Chinese). DOI: 10.3969/j.issn.1002-3550.2018.12.015

    [5]

    LAM L, TENG J G. Design-oriented stress-strain model for FRP-confined concrete[J]. Construction and Building Materials,2003,17(6-7):471-489. DOI: 10.1016/S0950-0618(03)00045-X

    [6]

    AHMAD S M, SHAH S P. Stress-strain curves of concrete confined by spiral reinforcement[J]. Civil and Environmental Engineering,1982,79(6):484-490.

    [7]

    MOZUMDER R A, ROY B, LASKAR A I. Support vector regression approach to predict the strength of FRP confined concrete[J]. Arabian Journal for Science and Engineering,2017,42(3):1129-1146. DOI: 10.1007/s13369-016-2340-y

    [8]

    KAMGAR R, NADERPOUR H, JANKOWSKI R. A proposed soft computing model for ultimate strength estimation of FRP-confined concrete cylinders[J]. Applied Science-Basel,2020,10(5):1769. DOI: 10.3390/app10051769

    [9]

    CEVIK A, CABALAR A F. A genetic-programming-based formulation for the strength enhancement of fiber-reinforced-polymer-confined concrete cylinders[J]. Journal of Applied Polymer Science,2008,110(5):3087-3095. DOI: 10.1002/app.28839

    [10]

    CEVIK A, GOGUS M T, GUZELBZY I H, et al. Soft computing based formulation for strength enhancement of CFRP confined concrete cylinders[J]. Advance in Engineering Software,2010,41(4):527-536. DOI: 10.1016/j.advengsoft.2009.10.015

    [11]

    GANDOMI A H, ALAVI A H, SAHAB M G. New formulation for compressive strength of CFRP confined concrete cylinders using linear genetic programming[J]. Materials and Structures,2010,43(7):963-983. DOI: 10.1617/s11527-009-9559-y

    [12]

    MANSOURI I, OZBAKKALOGLU T, KISI O, et al. Predicting behavior of FRP-confined concrete using neuro fuzzy, neural network, multivariate adaptive regression splines and M5 model tree techniques[J]. Materials and Structures,2016,49(10):4319-4334. DOI: 10.1617/s11527-015-0790-4

    [13]

    ZHAO J L, YU T, TENG J G. Stress-strain behavior of FRP-confined recycled aggregate concrete[J]. Journal of Composites Construction,2014,19(3):4014-4054.

    [14]

    CHEN G M, HE Y H, JIANG T, et al. Behavior of CFRP-confined recycled aggregate concrete under axial compression[J]. Construction and Building Materials,2016,111:85-97. DOI: 10.1016/j.conbuildmat.2016.01.054

    [15]

    JIANG T, TENG J G. Analysis-oriented models for FRP confined concrete: A comparative assessment[J]. Engineering Structures,2007,29(11):2968-2986. DOI: 10.1016/j.engstruct.2007.01.010

    [16]

    CUI C, SHEIKH S A. Experimental study of normal-and high-strength concrete confined with fiber-reinforced polymers[J]. Journal of Composites Construction,2010,14(5):553-561. DOI: 10.1061/(ASCE)CC.1943-5614.0000116

    [17] 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), 2021, 48(9):88-97.

    MA Gao, LIU Kang. Prediction of compressive strength of CFRP-confined concrete columns based on BP neural network[J]. Journal of Hunan University (Natural Sciences),2021,48(9):88-97(in Chinese).

    [18]

    FERREIRA C. Gene expression programming: A new adaptive algorithm for solving problems[J]. Complex Systems,2001,13(2):87-129. DOI: 10.48550/arXiv.cs/0102027

    [19]

    FERREIRA C. Gene expression programming[M]. Portugal: Angora do Heroismo, 2002.

    [20]

    MITCHELL M. An introduction to genetic algorithms[M]. London: MIT Press, 1996.

    [21]

    FERREIRA C. Mutation, transposition, and recombination: An analysis of the evolutionary dynamics[C]. 4th International Work Shop on Frontiers in Evolutionary Algorithm. North Carolin: Research Triangle Park, 2001: 614-617.

    [22]

    WANG Y C. A review of gene expression programming algorithms and their applications[J]. Computer Applications and Software,2010,27(6):23-26.

    [23]

    AHMAD A, PLEVRIS V, KHAN Q U Z. Prediction of properties of FRP-confined concrete cylinders based on artificial neural networks[J]. Crystals,2020,10(9):811. DOI: 10.3390/cryst10090811

    [24]

    YU T, TENG J G. Design of concrete-filled FRP tubular columns: Provisions in the Chinese technical code for infrastructure application of FRP composites[J]. Journal of Composites Construction,2011,15(3):451-461.

    [25]

    BENZAID R, MESBAH H, CHIKH N E. FRP-confined concrete cylinders: Axial compression experiments and strength model[J]. Journal Reinforced Plastics and Composites,2010,29(16):2469-2488. DOI: 10.1177/0731684409355199

    [26]

    TENG J G, JIANG T, LAM L, et al. Refinement of a design-oriented stress-strain model for FRP-confined concrete[J]. Journal of Composites Construction,2009,13(4):269-278. DOI: 10.1061/(ASCE)CC.1943-5614.0000012

    [27]

    Al-TERSAWY S H, HODHOD O A, HEFNAWY A A. Reliability and code calibration of RC short columns confined with CFRP wraps[C]. In Proc. 8th Int. Symp. on Fiber Reinforced Polymer Reinforcement for Concrete Structures. Patras, Greece: University of Patras, 2007: 560-563.

    [28]

    ILKI A, KUMBASAR N, KOC V. Low strength concrete members externally confined with FRP sheets[J]. Structural Engineering and Mechanics,2004,18(2):167-194. DOI: 10.12989/sem.2004.18.2.167

    [29]

    XIAO Y, WU H. Compressive behavior of concrete confined by carbon fiber composite jackets[J]. Journal of Materials in Civil Engineering,2000,12(2):139-146. DOI: 10.1061/(ASCE)0899-1561(2000)12:2(139)

    [30]

    SAMAAN M, MIRMIRAN A, SHAHAWY M. Model of concrete confined by fiber composites[J]. Journal of Structural Engineering,1998,124(9):1025-1031. DOI: 10.1061/(ASCE)0733-9445(1998)124:9(1025)

    [31]

    MANDER J B, PRIESTLEY M J N, PARK R. Theoretical stress-strain model for confined concrete[J]. Journal of Structural Engineering,1988,114(8):1804-1826. DOI: 10.1061/(ASCE)0733-9445(1988)114:8(1804)

    [32]

    YASMIN M, AHMED A, ROZEN H. Predictive model to the bond strength of FRP-to-concrete under direct pullout using gene expression programming[J]. Journal of Civil Engineering and Management,2019,25(8):773-784. DOI: 10.3846/jcem.2019.10798

    [33]

    AIRE C, GETTU R, CASAS J R. Study of the compressive behavior of concrete confined by fiber reinforced composites[C]. Composites in Constructions, Proceedings of the International Conference. Lisse, Netherlands: A. A. Balkema, 2001: 239-243.

  • 期刊类型引用(12)

    1. 龙国文,曾开华,谢帮华,田海,邱智坚. 相变混凝土复合材料的性能研究综述及展望. 功能材料. 2024(10): 10038-10046 . 百度学术
    2. 陈春鸣,张燕,彭建山. 轻质相变混凝土物理力学性能研究进展. 化工新型材料. 2024(S2): 294-298 . 百度学术
    3. 肖桐,刘庆祎,张家豪,赵佳腾,刘昌会. 热固性树脂基复合相变材料的制备及其储能强化研究进展. 复合材料学报. 2023(03): 1311-1327 . 本站查看
    4. 张宇,恽定康,邱菊. 冻融循环下相变混凝土剪力墙抗震性能研究. 徐州工程学院学报(自然科学版). 2023(01): 42-49 . 百度学术
    5. 张家玮,黄玮,黄大建,李旭辉,马建虎,郑永. 基于微孔漂珠的相变微胶囊制备及其对砂浆力学和热性能影响. 复合材料学报. 2023(08): 4703-4719 . 本站查看
    6. 于文艳,殷凯. 复合相变蓄热涂层的蓄放热调温特性. 科学技术与工程. 2023(31): 13492-13498 . 百度学术
    7. 蔡昕辰,刘志彬,张云,白梅,刘锋. 相变材料在道路工程中的应用研究进展. 功能材料. 2021(12): 12013-12021 . 百度学术
    8. 周建庭,聂志新,郭增伟,杨娟,郑忠. 相变混凝土的制备与性能研究综述. 江苏大学学报(自然科学版). 2020(05): 588-595 . 百度学术
    9. 刘朋,缪正坤,田国华,刘伟,王东. 不同工况下陶粒吸附相变石蜡试验研究. 江苏科技信息. 2019(04): 44-46 . 百度学术
    10. 司亚余,马芹永,罗小宝,顾皖庆,白梅. 癸酸-正辛酸与活性炭复合相变材料的制备. 安徽理工大学学报(自然科学版). 2019(03): 66-72 . 百度学术
    11. 马芹永,白梅. 珍珠岩基相变骨料混凝土断裂特性试验与分析. 建筑材料学报. 2018(03): 365-369 . 百度学术
    12. 顾皖庆,马芹永,白梅,罗小宝,司亚余. 月桂醇膨胀珍珠岩复合相变材料吸附试验与分析. 科学技术与工程. 2018(25): 230-235 . 百度学术

    其他类型引用(21)

图(7)  /  表(8)
计量
  • 文章访问数:  1056
  • HTML全文浏览量:  486
  • PDF下载量:  62
  • 被引次数: 33
出版历程
  • 收稿日期:  2021-11-21
  • 修回日期:  2022-01-12
  • 录用日期:  2022-01-14
  • 网络出版日期:  2022-01-26
  • 刊出日期:  2023-01-14

目录

    /

    返回文章
    返回