基于分形研究的受腐蚀钢筋混凝土构件承载力预测模型

Fractal-based model for prediction on load bearing capacity of corroded reinforced concrete members

受腐蚀钢筋混凝土构件的承载力受多种影响因素综合作用,影响因素的随机性导致受腐蚀钢筋混凝土构件承载力难以确定.提出将分形维数作为钢筋混凝土构件受腐蚀程度的定量衡量指标,以反映构件中钢筋锈蚀率及混凝土劣化状态;将此腐蚀指标与构件基本信息(如构件几何尺寸、配筋情况、加载信息等)相结合,建立了受腐蚀钢筋混凝土构件极限承载力人工神经网络预测模型,模型预测结果与试验结果能够很好地吻合.较好地解决了目前尚无法建立腐蚀指标的难题,为实际受腐蚀钢筋混凝土构件极限承载力的确定提供了一个简便可行的新方法.

Load bearing capacity of the corroded reinforced concrete (RC) member is a highly nonlinear problem and is difficult to be determined, which is influenced by many stochastic factors. Fractal dimension of surface cracks on the corroded reinforced concrete (CRC) beams, which was used to describe the corrosion state of the RC member, was taken as the corrosion index. Combining such an index and some basic information (including the geometry dimension, arrangement of steel, loading case, etc.) of the CRC beams, an artificial neural network model was built to predict the ultimate load bearing capacity of the CRC members. Good results are achieved. Therefore, the difficulties in the description of corrosion states of CRC members are overcome successfully. It is proved that the proposed method is a promising method for the determination of ultimate loading capacity of CRC members in practical engineering.