Study on the Damage Assessment method of Reinforced Concrete Structure by Nondestructive Impact Type Inspection

近年来，日本既有钢筋混凝土结构的维修问题日益严重，对结构全寿命周期费用进行估算具有重要意义。为了达到这一目的，有必要对既有结构的耐久性进行评估，并准确预测结构的劣化程度。在本研究中，1)对既有钢筋混凝土桥梁进行了多次无损冲击式检测，以考察其可靠性和准确性。2)为了建立能够估计化学和力学效应引起的损伤演化过程的分析方法，建立了基于连续损伤力学的数值计算程序。结果表明，该方法可以评估既有钢筋混凝土梁的耐久性，以及在不同暴露条件下其极限强度的变化。3)本文还提出了多模式微分神经网络在混凝土材料疲劳损伤监测中的应用。该系统利用输入数据和输出数据之间的关系作为监测信号来检测结构特性的变化，能够很好地检测输入数据和输出数据之间具有复数模式的混凝土的疲劳损伤过程

Recently, the maintenance problem of the existing RC structures has increased in Japan, and it is important to estimate the total life cycle cost of structure. In order to attain this purpose, it is necessary to evaluate the durability and predict the deterioration of the existing structures accurately. In this study, the following considerations are presented.1)Several Nondestructive Impact Type Inspections were performed for the existing reinforced concrete bridges in order to investigate its reliability and the accuracy.2)To develop the analytical method that could estimate the damage evolution process caused by chemical and mechanical effect, numerical program based on continuum damage mechanics was constructed. It is found that the proposed method could evaluate the durability of the existing RC beam, and the change of its ultimate strength under the various exposition conditions.3)This study also presents an application of multi-pattern differentiation neural network for monitoring of fatigue damage in concrete material. This system uses the relationship between input data and output data for monitoring signal to detect the change of structural characteristic, and it can detect the progress of fatigue damage of concrete very well which has plural pattern between input data and output data