基于模态应变能与神经网络的地下拱形结构损伤诊断

地下拱形结构是人防工程的重要结构形式，其损伤诊断是人防工程进行安全性评定的一个重要环节。基于地下拱形结构损伤前后动力特性的分析,证明模态应变能变化对单元损伤的敏感性。定义单元模态应变能变化率，并将其作为人防结构损伤诊断的标识量,通过标识量对损伤进行初步定位，然后采用BP神经网络来确定受损位置的损伤程度。以一典型的人防工程为算例，通过对其在不同损伤情况下计算结果的分析和讨论，说明本文所提出的方法是有效可行的，它可以准确完成地下拱形结构的损伤诊断。

Damage Diagnosis for Underground Arch Structure Based on Modal Strain Energy and Neural Networks

Underground arch structure is an important structure form of civil air defense engineering, and its damage detection is one important step for assessing the security of civil air defense engineering. Based on analysis of dynamic property of underground arch structure before and after damage, mode strain energy was proved to be sensitive to damage. The change rate of element mode strain energy was defined and taken as the damage signature. The damage location was confirmed by damage signature, and the damage extent was assured by the BP neural works. A typical air defense engineering was taken as an example, the method was verified through analyzing and discussing the calculating results of different damage. It is proved that the proposed method is correct and can detect damage in underground arch structures accurately.