近似熵在混凝土结构损伤识别中的应用  

Application of approximate entropy in concrete structures damage identification

作　　者：谢中凯[1] 刘国华[1] 

机构地区：[1]浙江大学水工结构与水环境研究所,浙江杭州310058

出　　处：《浙江大学学报（工学版）》2013年第3期456-464,共9页Journal of Zhejiang University：Engineering Science

基　　金：国家自然科学基金资助项目(50579081)

摘　　要：基于近似熵理论,利用自由振动信号对钢筋混凝土梁结构进行损伤动力识别研究.通过自功率谱密度函数对衰减信号进行分类,针对不同的信号模式提出不同的近似熵计算方法进行结构的损伤识别.使用互相关函数将不具有规则性的衰减信号转化成具有规则性的互相关-时间序列,采用数值模拟结合试验的方法论证了对衰减信号引入近似熵理论的可行性.对由试验测得的不同损伤工况下的自由振动信号引入互相关近似熵理论发现,互相关近似熵能够识别混凝土结构的小损伤,体现了良好的敏感性;对试验数据添加不同强度的噪声,再进行互相关近似熵计算,发现互相关近似熵在信噪比很低的情况下仍然能够对测试的梁结构进行损伤识别,体现了很强的抗噪能力.Approximate entropy（ApEn） was introduced to detect damage in reinforced concrete beam structures based on free vibration signal.Decay signal was divided into various categories via auto spectral density function,and different ApEn methods were proposed according to different signal patterns.Cross-correlation function was applied to transform irregular decay signal into regular cross-correlation time series.The feasibility of introducing ApEn into calculating decay signal was validated by numerical simulation and experiment method.Tiny damage in concrete structures could be sensitively detected by applying cross-correlation ApEn method to free vibration signal tested from the experiment under various damage conditions.Various-intensity noise was added to the original tested signal,and the cross-correlation ApEn method was found to be able to recognize the damage under a very low signal-to-noise ratio condition,which confirms that the cross-correlation ApEn method has strong antinoise ability.

关 键 词：衰减信号 自功率谱密度 近似熵 互相关近似熵 损伤识别 

分 类 号：TU317[建筑科学—结构工程] TU528