基于D-S证据理论的结构多精度损伤识别结果融合方法

A RESULT-MERGING ALGORITHM OF STRUCTURE DAMAGE IDENTIFICATIONS WITH MULTI-RESOLUTION BASED ON D-S EVIDENTIAL THEORY

在结构的损伤检测中,由于传感器数目、测试噪声等因素,限制了结构损伤检测与识别的能力。不同的损伤识别方法在检测过程中识别的精度不同,使用单一的损伤检测识别方法所获得的信息是不全面的。将Dempster-Shafer证据理论应用于结构的损伤识别中,将各种识别不同精度的检测方法融合,使最终识别的结果更为可靠、精确。通过对钢梁结构损伤识别的模拟计算,验证了D-S证据理论在融合多种不同精度损伤识别方法中的有效性。

Because of effect of sensor numbers and test noise,detectability and recognition capability are restricted in structure damage detection.As resolutions of different damage detection algorithms are quite not alike, the overall-information of a structure can't be obtained by single algorithm. Here,D-S evidential theory is implemented in the damage detection of a structure and merging of damage detection algorithms with different resolutions is achieved.Through the simulation of the damage detection about a girder structure, the effectivity of D-S evidential theory used for the merging of many structure damage detection results with multi-resolution is proved.