工程科学与技术   2021, Vol. 53 Issue (2): 118-124
基于基因表达式编程的NSM FRP–混凝土粘结强度预测模型
张芮椋, 薛新华     
四川大学 水利水电学院,四川 成都 610065
摘要: 纤维增强复合材料(fiber-reinforced polymer,FRP)已被广泛应用于混凝土加固工程中。FRP与混凝土界面间的粘结性能是影响加固效果的重要因素之一,为准确预测FRP嵌入式加固(near-surface-mounted,NSM)NSM FRP–混凝土的粘结强度,运用基因表达式编程(gene expression programming,GEP)方法,选取混凝土抗压强度、粘结长度、槽深宽比、FRP轴向刚度、FRP抗拉强度及环氧树脂抗拉强度等6个参数作为粘结强度的影响因素,建立了NSM FRP与混凝土粘结强度的预测模型,提出了具体的计算公式。通过比较粘结强度预测值与实验值,发现二者较为接近,说明该模型具有一定的可靠性。对该模型进行敏感性分析,发现其可以反映粘结强度与单因素之间的内在关系,即粘结强度随着粘结长度、混凝土抗压强度、槽深宽比及FRP轴向刚度等因素的增大而增大。将该GEP模型与经验模型及小波神经网络模型进行比较,并选取6个统计指标对模型进行评价。结果表明,GEP模型与小波神经网络模型的精度较高,各项误差指标均较小,决定系数分别为0.793和0.787。总体而言,GEP模型的精度略优于小波神经网络模型,二者的精度均远高于经验模型。
关键词: 粘结强度    基因表达式编程    纤维增强复合材料    混凝土    预测模型    
Bond Strength Prediction Model of the Near-surface-mounted Fiber-reinforced Polymer Concrete Based on Gene Expression Programming
ZHANG Ruiliang, XUE Xinhua     
College of Water Resource and Hydropower, Sichuan Univ., Chengdu 610065, China
Abstract: Fiber-reinforced polymer (FRP) has been widely used in concrete reinforcement projects. The bonding performance between the FRP and concrete influences the reinforcement effect well. Numerous studies were conducted on the bond strength of FRP externally bonded concrete, and a lot of empirical models were developed, while there were few empirical models of the bond strength of FRP near-surface-mounted (NSM) concrete. In order to accurately predict the bond strength between near-surface-mounted FRP and concrete, the gene expression programming (GEP) method was employed to develop a bond strength prediction model of NSM FRP bonded to concrete, and a specific calculation formula was established. The model was developed using six parameters including the concrete compressive strength, bond length, groove depth-to-width ratio, FRP axial rigidity, FRP tensile strength and epoxy tensile strength. The predicted values calculated by the proposed formula agreed well with the experimental values, indicating that the model was reliable. Through the sensitivity analysis of the model, it was found that the GEP model can reflect the internal relationship between bond strength and single factor, i.e., the bond strength increased with the increase of the bond length, concrete compressive strength, the groove depth-to-width ratio and FRP axial rigidity. The GEP model was compared with the empirical model and the wavelet neural network model. Six statistical indicators were selected to evaluate the prediction performance of these models. It can be found that the accuracy of the GEP model and the wavelet neural network model was high, and the errors were low where the coefficients of determination were 0.793 and 0.787, respectively. In general, the accuracy of the GEP model was slightly better than the wavelet neural network model, and the accuracy of the two models was much higher than the empirical models.
Key words: bond strength    gene expression programming    fiber-reinforced polymer    concrete    prediction model    

钢筋腐蚀问题一直是桥梁工程和混凝土加固工程中值得关注的问题。近年来,纤维增强复合材料(fiber-reinforced polymer,FRP)凭借其抗腐蚀、耐疲劳等优点,逐渐替代钢筋,并发挥越来越重要的作用。与钢筋混凝土构件相同,FRP与混凝土界面间的粘结性能是构件能否正常工作的基础。因此,国内外许多专家学者如Benmokrane[1]、高丹盈[2]、Achillides[3]、郝庆多[4]等都曾对两者间的粘结性能及其影响因素进行过研究。

自20世纪末以来,对于FRP外贴式加固混凝土的研究颇为广泛,提出过许多FRP–混凝土粘结强度的经验模型。FRP嵌入式加固法(near-surface-mounted FRP reinforcement,NSM FRP)是一种较新的加固方法,它是将FRP筋嵌入混凝土保护层内的凹槽中,并将其表面用树脂填平[5]。目前对NSM FRP–混凝土粘结强度的研究还不够成熟,提出的粘结强度预测模型较少,比较典型的模型为Seracino[6]和Zhang[7]等提出的模型。近年来,一些学者将人工智能方法应用于该领域,如:Haddad等[8]将人工神经网络模型用于预测FRP–混凝土粘结强度;Nasrollahzadeh等[9]提出了两种用于确定NSM FRP与混凝土抗拔强度的模糊逻辑模型;Golafshani等[10]应用人工神经网络和遗传编程预测FRP与混凝土间的粘结强度;Yasmin等[11]将基因表达式编程应用于FRP外贴式加固混凝土的粘结强度预测中。

基因表达式编程(gene expression programming,GEP)是在遗传算法和遗传编程基础上提出的一种进化类算法,其效率远远高于遗传算法和遗传编程,且在函数发现方面具有独特优势[12-13]。因此,作者利用GEP方法建立了NSM FRP与混凝土间粘结强度的预测模型,并将预测结果与实验数据对比,验证了该预测模型的可靠性。

1 基因表达式编程

基因表达式编程GEP是Ferreiral[12]提出的一种借鉴生物学基因表达的算法,其计算步骤简述如下:首先创建初始种群,然后利用染色体进行表达,并对其进行适应度评估。然后选出最优个体,进行遗传操作形成新的种群,重复上述过程,直到发现优良染色体为止。

基因是构成染色体的基本单位,主要由头(可同时包含函数符号和终结符号)和尾(仅包含终结符号)两部分组成[14]。基因头长h可根据确定的问题预先选定,尾长t按下式计算:

$t = h \times (n - 1) + 1$ (1)

式中,n代表函数符集中的最大操作目数。

在GEP中有两种语言:基因语言和表达式树语言。每个基因对应一个表达式和一棵表达式树(expression tree, ET),分别代表基因型和表现型,两者之间可以相互转换。图1为染色体编码方法,即根据某一基因型,可得到对应的表达式,而该表达式又可由表达式树从上到下、从左至右遍历得到。

图1 染色体编码方法 Fig. 1 Chromosome coding method

在GEP中,Ferreira提出的适应度计算函数有3种[15]

1)基于绝对误差的适应度函数:

${f_i} = \sum\limits_{j = 1}^n {\left( {M - \left| {{C_{i,j}} - {T_j}} \right|} \right)} $ (2)

2)基于相对误差的适应度函数:

${f_i} = \sum\limits_{j = 1}^n {\left( {M - \left| {\frac{{{C_{i,j}} - {T_j}}}{{{T_j}}} \times 100} \right|} \right)} $ (3)

3)用于逻辑合成问题的适应度函数:

${\rm{If }}\,n \ge 1/2{\rm{ }}{C_t},\,{\rm{ then }}\,{f_i} = n;\,{\rm{ else }}\,{f_i} = 1$ (4)

式中,M为选择范围,Ci,j为染色体个体i对于适应度样本的返回值,Tj为适应度样本j的目标值,n为正确求得的适应度样本的个数,Ct为所有适应度样本的数目。

由于以上的适应度函数都有一定的局限性,这里采用均方根误差(root mean squared error,RMSE)作为适应度函数,其计算公式如下:

${f_i} = 1\;000 \times \frac{1}{{1 + {E_i}}}$ (5)

式中,Ei为预测序列与观测序列距离平方与样本总数之比的平方根, ${E_i} = \sqrt {\dfrac{1}{n}\displaystyle\sum\limits_{j = 1}^n {\left( {{y_j} - {y_{j,i}}} \right)} }$ ,适应度最大值为1 000。

2 基于GEP的NSM FRP–混凝土粘结强度预测模型 2.1 实验数据

从文献[16-34]中筛选条件参数各不相同的145组直拉实验数据用于建立NSM FRP–混凝土粘结强度预测模型。FRP在混凝土试件中均为单层单根布置,由于各文献中所用试件的尺寸不同,FRP长短及弯折程度也存在差异,考虑到对粘结强度产生影响的主要为锚固部分,故忽略FRP长度及弯折程度等因素的影响,重点考虑以下6个参数对粘结强度的影响,分别为:混凝土抗压强度fc′、粘结长度L、槽深宽比DgWg、FRP轴向刚度EfAf、FRP抗拉强度fu及环氧树脂抗拉强度fe,统计特征如表1所示。其中,FRP筋的粘结强度实验数据为86组,其余为FRP板条。从145组数据中随机选取100组数据作为训练数据,45组数据为测试数据。

表1 实验数据的统计参数[16-34] Tab. 1 Statistical parameters of experimental data[16-34]

2.2 GEP模型的建立

利用GeneXproTools5.0软件,通过变化基因数目、染色体数及连接函数等各种参数,得出了较为准确的NSM FRP–混凝土粘结强度预测模型。通过测试得到的最优参数如表2所示,选用的函数集为F={+,–,*,/, exp(x), ln, x2, 1/x},变量集为T={d0,d1,d2,d3,d4,d5},其中d0d5分别为LfuEfAfDg/Wgfefc′,输出值为粘结强度Pu

表2 GEP参数设置 Tab. 2 GEP setting parameter

按照如上设置的参数运行程序可得到最优染色体。此时,染色体上各基因对应的子表达式树如图2所示。其中,c为随机常数。第1个基因中有3个常数,c0为–595.352,c2为8.587,c7为0.308;第2个基因中有3个常数,c0为44.492,c1为60.819,c8为4.557;第3个基因中有2个常数,c0为1.554,c3为–7.255。由于染色体上各基因的连接函数为“+”,可得NSM FRP–混凝土粘结强度的预测函数为:

图2 表达式树 Fig. 2 Expression tree

$\begin{aligned}[b] {P_{\rm{u}}} =& \frac{{{A_{\rm{f}}}{E_{\rm{f}}}}}{{0.308 \left( {{f_{\rm{u}}} + 595.352} \right) - 8.587 f_{\rm{c}}'}} + f_{\rm{c}}'+ \\ &\frac{{\left( {{f_{\rm{e}}} + 44.492} \right) L}}{{60.819 f_{\rm{c}}'}} + \frac{{{{\rm{e}}^{\left( {{{{D_{\rm{g}}}} / {{W_{\rm{g}}}}} - 4.557} \right)}}}}{{f_{\rm{c}}'}} + 1.554 \times \\ &\left( {{{{D_{\rm{g}}}} / {{W_{\rm{g}}}}} - f_{\rm{c}}'} \right) \left( {7.255 + f_{\rm{c}}'} \right) \frac{1}{{L + f_{\rm{c}}'}} \end{aligned} $ (6)
2.3 模型的检验分析

采用决定系数(R2)、均方根误差(RMSE)、平均绝对误差(MAE)、相对平方根误差(RRSE)、平均绝对百分比误差(MAPE)及绝对误差积分(IAE)等6个指标来检验该预测模型的可行性及准确性,分别按式(7)~(12)计算。不难看出,决定系数越高,均方根误差、平均绝对误差等值越小,模型的预测效果越好。分别计算训练数据、测试数据及全部数据的各指标,列于表3

表3 GEP模型的统计指标[16-34] Tab. 3 Statistical indexes of the GEP model[16-34]

${\;\;\;\;\;\;\;\;\;\;\;\;\;\;{{R}}^{\rm{2}}} = \frac{{{{\left[ {\displaystyle\sum\limits_{i = 1}^n {{{\left( {{x_i} - \overline x} \right)}^2}{{\left( {{y_i} - \overline y} \right)}^2}} } \right]}^2}}}{{\displaystyle\sum\limits_{i = 1}^n {{{\left( {{x_i} - \overline x} \right)}^2}\displaystyle\sum\limits_{i = 1}^n {{{\left( {{y_i} - \overline y} \right)}^2}} } }}$ (7)
${\rm{RMSE}} = \sqrt {\frac{1}{n}\sum\limits_{i = 1}^n {{{\left( {{x_i} - {y_i}} \right)}^2}} } $ (8)
${\rm{MAE}} = \frac{1}{n}\sum\limits_{i = 1}^n {\left| {{x_i} - {y_i}} \right|} $ (9)
${\rm{RRSE}} = \sqrt {\frac{{\displaystyle\sum\limits_{i = 1}^n {{{\left( {{x_i} - {y_i}} \right)}^2}} }}{{\displaystyle\sum\limits_{i = 1}^n {{{\left( {{x_i} - \overline x} \right)}^2}} }}} $ (10)
${\rm{MAPE}} = \frac{1}{n}\sum\limits_{i = 1}^n {\left| {\frac{{{x_i} - {y_i}}}{{{x_i}}}} \right|} $ (11)
${\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\;\; \rm{IAE}} = \frac{{\displaystyle\sum\limits_{i = 1}^n {{{\left[ {{{\left( {{x_i} - {y_i}} \right)}^2}} \right]}^{\frac{1}{2}}}} }}{{\displaystyle\sum\limits_{i = 1}^n {{x_i}} }} \times 100{\text{%}} $ (12)

实验数据的真实值与预测值对比见图3。从图3可见,基于基因表达式编程建立的模型精度较高,粘结强度的预测值与真实值之间拟合程度较高,相关性较好。根据表1计算可知,粘结强度的最大值与最小值相差87.03 kN。而在表3中,均方根误差和平均绝对误差分别为7.85 kN和6.34 kN,且相对平方根误差、平均绝对百分比误差及绝对误差积分都较小,这说明基于基因表达式编程模型得到的预测值与真实值之间误差相对较小。由此可见,该预测模型具有一定的可行性和有效性。

图3 GEP模型预测结果与实验结果对比 Fig. 3 Comparison of prediction results of GEP model with experimental results

为了检验该模型是否能够正确反映各因素与粘结强度之间的关系,对该模型进行敏感性分析。限于篇幅,仅对粘结长度、混凝土抗压强度、槽深宽比及FRP轴向刚度与粘结强度的关系进行分析(图4)。在研究某一因素对粘结强度的影响时,保持其他因素不变且等于其平均值,即L=207.8 mm,fu=2041.9 MPa,EfAf = 3865.3 kN,Dg/Wg= 2.04,fe=40.4 MPa,fc′=36.4 MPa。由图4可见,随着粘结长度、混凝土抗压强度、槽深宽比及FRP轴向刚度的增加,粘结强度均增大。这与文献中的研究结果是一致的。在图4(a)中:当粘结长度较小时,随着粘结长度的增加,粘结强度增大较快;当粘结长度继续增加,粘结强度增大的速度变缓,在一定程度上体现了“有效粘结长度”的概念。值得说明的是,在研究槽深宽比与粘结强度的关系时,控制槽深度或宽度一致才有意义。一般来说,环氧树脂的强度大于混凝土的强度,而槽的尺寸越大,填充所用的环氧树脂越多,这样粘结强度也会提高。从以上分析可知,基于基因表达式编程建立的预测模型能够反映粘结强度与各影响因素之间的内在机理。

图4 GEP模型敏感性分析 Fig. 4 Sensitivity analysis of GEP model

3 GEP模型与其他模型对比分析

Seracino[6]和Zhang[7]等都曾提出过NSM FRP与混凝土粘结强度的经验模型。Seracino等提出的模型既适用于外贴式加固法,也适用于嵌入式加固法,但是其局限性在于该模型不能应用于FRP筋,因此本文不与该模型进行比较。Zhang等提出的NSM FRP与混凝土粘结强度的经验模型对FRP筋及FRP板条均适用,其计算公式如下所示:

${P_{\rm{u}}} = \left\{ \begin{array}{l} \sqrt {2{G_{\rm{f}}}{A_{\rm{f}}}{E_{\rm{f}}}{L_{{\rm{per}}}}} \le {P_{\rm{t}}},\;L \ge {L_{\rm{e}}} ; \\ {\beta _{\rm{L}}}\sqrt {2{G_{\rm{f}}}{A_{\rm{f}}}{E_{\rm{f}}}{L_{{\rm{per}}}}} \le {P_{\rm{t}}},\;L < {L_{\rm{e}}} \end{array} \right.$ (13)

式中:βL为折减系数, ${\;\beta _{\rm{L}}} = \dfrac{L}{{{L_{\rm{e}}}}}\left( {2.08 - 1.08\dfrac{L}{{{L_{\rm{e}}}}}} \right)$ Le为有效粘结长度, ${L_{\rm{e}}} = \dfrac{{1.66}}{\eta }$ ${\eta ^2} = \dfrac{{\tau _{\max }^2{L_{{\rm{per}}}}}}{{2{G_{\rm{f}}}{A_{\rm{f}}}{E_{\rm{f}}}}}$ Gf为界面断裂能, ${G_{\rm{f}}} = 0.4{\left( {{{{D_{\rm{g}}}} / {{W_{\rm{g}}}}}} \right)^{0.422}}f_{\rm{c}}^{'0.619}$ τmax为最大粘结应力, ${\tau _{\max }} = 1.15{\left( {{{{D_{\rm{g}}}} / {{W_{\rm{g}}}}}} \right)^{0.138}}f_{\rm{c}}^{'0.613}$ Lper为凹槽三边长度之和, ${L_{{\rm{per}}}} = 2{D_{\rm{g}}} + {W_{\rm{g}}}$

将GEP模型与Zhang模型及小波网络(wavelet neural network,WNN)建立的模型进行对比,分别计算R2、RMSE等6个统计指标列于表4图56分别为Zhang模型及小波神经网络建立的模型预测结果与实验结果的对比图。

表4 各模型的统计指标 Tab. 4 Statistical indexes of each model

图5 Zhang模型预测结果与实验结果对比 Fig. 5 Comparison of prediction results of Zhang model with experimental results

图6 WNN模型预测结果与实验结果对比 Fig. 6 Comparison of prediction results of WNN model with experimental results

图56图3进行比较,可以发现,GEP模型与WNN模型的预测结果与实验结果较为接近,拟合效果较好,二者的精度远高于Zhang模型。吴以莉等[35]的研究表明,粘结剂对粘结性能有较大影响。Zhang模型中仅考虑了粘结长度、混凝土抗压强度、槽尺寸、轴向刚度及FRP抗拉强度等因素的影响,没有考虑粘结剂的影响。而本文中用于建立模型的数据来源于不同文献中的实验结果,尽管粘结剂均为环氧树脂,但仍存在差别。因此,在未考虑粘结剂这一因素前提下,Zhang模型的预测效果较差。

表4可以看出,GEP模型的统计指标与WNN模型较为接近,预测效果也大体相同,这说明建立的GEP模型是比较可靠的。而GEP模型的决定系数、均方根误差、相对平方根误差、平均绝对百分比误差值均略优于WNN模型,绝对误差积分值相同,平均绝对误差值略高于WNN模型。综合来看,GEP模型的精度略高于WNN模型。

4 结 论

利用GeneXproTools5.0软件建立了NSM FRP–混凝土粘结强度的GEP预测模型,并将其与其他模型进行比较,得出了如下结论:

1)利用搜集整理的145组数据,建立了可以预测NSM FRP–混凝土粘结强度的GEP模型。该模型可以反映粘结长度、FRP抗拉强度、FRP轴向刚度、槽深宽比、环氧树脂抗拉强度及混凝土抗压强度等6个参数对粘结强度的影响。

2)利用6个统计指标对建立的GEP模型进行评价。结果表明,决定系数较大,均方根误差、平均绝对误差等较小,验证了该模型的可行性及准确性。通过单因素变量法对该模型进行了敏感性分析,证明该模型可以反映粘结强度与各影响因素之间的内在机理。

3)将GEP模型与Zhang模型、WNN模型进行比较,发现GEP模型与WNN模型的预测效果较好,精度均高于Zhang模型。而GEP模型与WNN模型的预测结果较为接近,总体上来说,GEP模型的精度略高于WNN模型。

参考文献
[1]
Benmokrane B,Tighiouart B,Chaallal O. Bond strength and load distribution of composite GFRP reinforcing bars in concrete[J]. ACI Materials Journal, 1996, 93(3): 246-253.
[2]
Gao Danying,Brahim B. Influential factors of bond properties between fiber reinforced polymer(FRP) rebars and concrete[J]. Industrial Construction, 2001, 31(2): 9-14. [高丹盈,Brahim B. 纤维聚合物筋与混凝土粘结性能的影响因素[J]. 工业建筑, 2001, 31(2): 9-14. DOI:10.13204/j.gyjz2001.02.00]
[3]
Achillides Z,Pilakoutas K. Bond behavior of fiber reinforced polymer bars under direct pullout conditions[J]. Journal of Composites for Construction, 2004, 8(2): 173-181. DOI:10.1061/(ASCE)1090-0268(2004)8:2(173)
[4]
Hao Qingduo,Wang Yanlei,Hou Jilin,et al. Experimental study on bond behavior of GFRP ribbed rebars[J]. Engineering Mechanics, 2008, 25(10): 158-165. [郝庆多,王言磊,侯吉林,等. GFRP带肋筋粘结性能试验研究[J]. 工程力学, 2008, 25(10): 158-165.]
[5]
Guo Weitong,Su Jianyao. Research on bonding performance of FRP embedded reinforcement concrete[J]. Journal of Hebei Institute of Architecture and Civil Engineering, 2017, 35(4): 46-50. [郭卫彤,苏建遥. FRP嵌入式加固混凝土粘结性能研究[J]. 河北建筑工程学院学报, 2017, 35(4): 46-50. DOI:10.3969/j.issn.1008-4185.2017.04.00]
[6]
Seracino R,Saifulnaz M R R,Oehlers D J. Generic debonding resistance of EB and NSM plate-to-concrete joints[J]. Journal of Composites for Construction, 2007, 11(1): 62-70. DOI:10.1061/(ASCE)1090-0268(2007)11:1(62)
[7]
Zhang S S,Teng J G,Yu T. Bond strength model for CFRP strips near-surface mounted to concrete[J]. Journal of Composites for Construction, 2013, 18(3): 1-46. DOI:10.1061/(ASCE)CC.1943-5614.0000402
[8]
Haddad R,Haddad M. Predicting fiber-reinforced polymer–concrete bond strength using artificial neural networks:A comparative analysis study[J]. Structural Concrete, 2020, 1-12. DOI:10.1002/suco.201900298
[9]
Nasrollahzadeh K,Afzali S. Fuzzy logic model for pullout capacity of near-surface-mounted FRP reinforcement bonded to concrete[J]. Neural Computing and Applications, 2018, 31(11): 7837-7865. DOI:10.1007/s00521-018-3590-2
[10]
Golafshani E M,Rahai A,Sebt M H. Artificial neural network and genetic programming for predicting the bond strength of GFRP bars in concrete[J]. Materials and Structures, 2015, 48(5): 1581-1602. DOI:10.1617/s11527-014-0256-0
[11]
Yasmin M,Ahmed A,Rozan 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
[12]
Ferreiral C. Gene expression programming:A new adaptive algorithm for solving problems[J]. Complex Systems, 2001, 13(2): 87-129.
[13]
Ferreiral C.Gene expression programming[M].Portugal:Angora do Heroismo,2002.
[14]
Peng Yuzhong,Yuan Changan,Mai Xiongfa,et al. Survey on theoretical research of gene expression programming[J]. Application Research of Computers, 2011, 28(2): 413-419,438. [彭昱忠,元昌安,麦雄发,等. 基因表达式编程的理论研究综述[J]. 计算机应用研究, 2011, 28(2): 413-419,438. DOI:10.3969/j.issn.1001-3695.2011.02.00]
[15]
Wang Yanchun. On gene expression programming algorithm and its applications[J]. Computer Application and Software, 2010, 27(6): 23-26. [王艳春. 基因表达式编程算法及其应用综述[J]. 计算机应用与软件, 2010, 27(6): 23-26. DOI:10.3969/j.issn.1000-386X.2010.06.00]
[16]
Sharaky I A,Torres L,Baena M,et al. An experimental study of different factors affecting the bond of NSM FRP bars in concrete[J]. Composite Structures, 2013, 99: 350-365. DOI:10.1016/j.compstruct.2012.12.014
[17]
De Lorenzis L,Rizzo A,La Tegola A. A modified pull-out test for bond of near-surface mounted FRP rods in concrete[J]. Composites (Part B:Engineering), 2002, 33(8): 589-603. DOI:10.1016/S1359-8368(02)00052-5
[18]
De Lorenzis L,Nanni A. Bond between near-surface mounted fiber-reinforced polymer rods and concrete in structural strengthening[J]. ACI Structural Journal, 2002, 99(2): 123-132.
[19]
Lee D,Cheng L,Yan–Gee H J. Bond characteristics of various NSM FRP reinforcements in concrete[J]. Journal of Composites for Construction, 2012, 17(1): 117-129. DOI:10.1061/(ASCE)CC.1943-5614.0000318
[20]
Ceroni F,Pecce M,Bilotta A,et al. Bond behavior of FRP NSM systems in concrete elements[J]. Composites (Part B:Engineering), 2012, 43(2): 99-109. DOI:10.1016/j.compositesb.2011.10.017
[21]
Kalupahana W,Ibell T J,Darby A P. Bond characteristics of near surface mounted CFRP bars[J]. Construction and Building Materials, 2013, 43: 58-68. DOI:10.1016/j.conbuildmat.2013.01.021
[22]
Bilotta A,Ceroni F,Di Ludovico M,el al. Bond efficiency of EBR and NSM FRP systems for strengthening concrete members[J]. Journal of Composites for Construction, 2011, 15(5): 757-772. DOI:10.1061/(ASCE)CC.1943-5614.0000204
[23]
Bilotta A,Ceroni F,Nigro E,et al. Strain assessment for the design of NSM FRP systems for the strengthening of RC members[J]. Construction and Building Materials, 2014, 69: 143-158. DOI:10.1016/j.conbuildmat.2014.07.024
[24]
Palmieri A,Matthys S,Barros J A,et al.Bond of NSM FRP strengthened concrete:round robin test initiative[C]//6th International Conference on FRP Composites in Civil Engineering.Rome:CICE,2012.
[25]
Seracino R,Jones N M,Ali M S,et al. Bond strength of near-surface mounted FRP strip-to-concrete joints[J]. Journal of Composites for Construction, 2007, 11(4): 401-409. DOI:10.1061/(ASCE)1090-0268(2007)11:4(401)
[26]
Soliman S M,El–Salakawy E,Benmokrane B. Bond performance of near-surface-mounted FRP bars[J]. Journal of Composites for Construction, 2010, 15(1): 103-111. DOI:10.1061/(ASCE)CC.1943-5614.0000150
[27]
Novidis D G,Pantazopoulou S J. Bond tests of short NSM FRP and steel bar anchorages[J]. Journal of Composites for Construction, 2008, 12(3): 323-333. DOI:10.1061/(ASCE)1090-0268(2008)12:3(323)
[28]
Shield C,French C,Milde E.The effect of adhesive type on the bond of NSM tape to concrete[C]//7th International Symposium on Fiber-Reinforced Polymer (FRP) Reinforcement for Concrete Structures.Kansas:ACI,2005.
[29]
Lee D,Hui J,Cheng L.Bond characteristics of NSM reinforcement in concrete due to adhesive type and surface configuration[C]//6th International Conference on FRP Composites in Civil Engineering.Rome:CICE,2012.
[30]
Palmieri A,Matthys S,Taerwe L.Double bond shear tests on NSM FRP strengthened members[C]//6th International Conference on FRP Composites in Civil Engineering.Rome:CICE,2012.
[31]
Novidis D,Pantazopoulou S,Tentolouris E. Experimental study of bond of NSM-FRP reinforcement[J]. Construction and Building Materials, 2007, 21(8): 1760-1770. DOI:10.1016/j.conbuildmat.2006.05.054
[32]
De Lorenzis L,Lundgren K,Rizzo A. Anchorage length of near-surface mounted fiber-reinforced polymer bars for concrete strengthening-experimental investigation and numerical modeling[J]. ACI Structural Journal, 2004, 101(2): 269-278.
[33]
ElGamal S,AlSalloum Y,Alsayed S,et al. Performance of near surface mounted glass fiber reinforced polymer bars in concrete[J]. Journal of Reinforced Plastics Composites, 2012, 31(22): 1501-1515. DOI:10.1177/0731684412464088
[34]
Peng H,Hao H,Zhang J,et al. Experimental investigation of the bond behavior of the interface between near-surface-mounted CFRP strips and concrete[J]. Construction and Building Materials, 2015, 96: 11-19. DOI:10.1016/j.conbuildmat.2015.07.156
[35]
Wu Yili,Yao Jian,Zhu Xiaoxu. Analysis on bonding performance test of mounted CFRP strengthened concrete beam[J]. Architecture Technology, 2010, 41(5): 454-456. [吴以莉,姚谏,朱晓旭. 内嵌CFRP加固混凝土梁粘结性能试验分析[J]. 建筑技术, 2010, 41(5): 454-456. DOI:10.3969/j.issn.1000-4726.2010.05.020]