文章摘要
高旭,黄丽华.冻融循环下FRP筋混凝土界面黏结强度预测[J].,2024,64(1):57-63
冻融循环下FRP筋混凝土界面黏结强度预测
Prediction of interfacial bond strength of FRP reinforced concrete under freeze-thaw cycles
  
DOI:10.7511/dllgxb202401007
中文关键词: FRP筋混凝土  黏结强度  冻融循环  反向传播神经网络(BPNN)  基因表达式编程(GEP)
英文关键词: FRP reinforced concrete  bond strength  freeze-thaw cycle  back propagation neural network (BPNN)  gene expression programming (GEP)
基金项目:国家自然科学基金资助项目(51678115).
作者单位
高旭,黄丽华  
摘要点击次数: 63
全文下载次数: 53
中文摘要:
      在冻融、腐蚀等恶劣服役环境下,用纤维增强复合材料(FRP)代替钢筋来提升混凝土结构的耐久性,已越来越多地应用在土木工程中.针对冻融循环下FRP筋混凝土界面黏结机理复杂,反映界面性能的理论模型难以构建问题,基于文献中110组冻融循环下FRP筋混凝土拉拔试验数据,采用遗传算法优化的反向传播神经网络(GA-BPNN)预测FRP筋混凝土界面黏结强度,通过分析权值矩阵的参数敏感性,筛选界面黏结强度的主要影响参数并以此为变量,运用基因表达式编程(GEP)方法建立界面黏结强度的计算公式.与目前文献中仅有的两个理论模型相比,所提公式在计算冻融循环下FRP筋混凝土界面黏结强度时精度更高、泛化性能更强.
英文摘要:
      In harsh service environments such as freeze-thaw and corrosion, fiber reinforced polymer (FRP) is used to take the place of the steel bars to improve the durability of concrete structures, which has been increasingly used in civil engineering. In view of the complex interfacial bond mechanism of FRP reinforced concrete under freeze-thaw cycles and the difficulty of constructing a theoretical model reflecting the interface performance, 110 groups of FRP reinforced concrete pull-out test data under freeze-thaw cycles in the literature are sorted out and the genetic algorithm optimized back propagation neural network (GA-BPNN) is applied to predict the interfacial bond strength of FRP reinforced concrete. By analyzing the parameter sensitivity of the weight matrix, the main parameters affecting the bond strength of the interface are assessed, and then the gene expression programming (GEP) method is used to establish the interfacial bond strength formula containing the main parameters. Compared with the two models presented in the literature, the proposed formula has higher accuracy and stronger generalization ability in calculating the interfacial bond strength of FRP reinforced concrete under freeze-thaw cycles.
查看全文   查看/发表评论  下载PDF阅读器
关闭