论文部分内容阅读
变形监测数据序列是由不同频率成分组成的、具有多尺度特性的动态信号序列,通过多尺度分解和融合,可获得其本质特性和关键信息。在单测点多尺度分解的基础上,分别采用模极大值法和小波熵加权法对多测点高频信号和低频信号进行融合,并以小波熵为特征参数,建立了用于大坝变形性态识别的多尺度多测点诊断模型,工程实例表明该方法诊断大坝健康状态效果良好。
Deformation monitoring data sequence is composed of different frequency components, with multi-scale characteristics of the dynamic signal sequence, through multi-scale decomposition and fusion, can obtain its essential characteristics and key information. Based on the multi-scale decomposition of single measurement point, the maximum and minimum entropy weights are combined respectively to fuse the high-frequency signals and low-frequency signals of multi-measurement points. The wavelet entropy is used as the characteristic parameter to establish a large- Multi-scale and multi-measuring point diagnosis model of deformation character recognition, engineering examples show that this method is effective in diagnosing dam health status.