摘要
为通过经验模态分解提取桥梁监测动应变信号中具有周期趋势特点的温度效应成分(温度应变),需确定温度应变的多阶本征模态函数(IMF)构成。首先在时域与频域分别对信号各阶IMF的能量进行分析,再综合IMF的能量突变阶数和边际谱相关系数得到IMF阶数阈值,最后根据阈值选取多阶IMF重构信号中的周期趋势成分,达到提取应变信号中温度效应成分的目的。研究结果表明,动应变信号中的周期趋势成分使得IMF能量存在阶数突变,基于突变阶数可初步确定IMF阶数阈值的取值范围。进一步分析阈值范围内IMF频域的能量分布,发现IMF边际谱曲线的相关系数陡增至0.8以上,表明末阶IMF存在较一致的能量分布模式。通过实测数据的应用结果验证,采用首个超过0.8的IMF边际谱相关系数可较精确地确定IMF阶数阈值。基于阈值提取温度效应成分后,动应变中的活载信息保留较完整。
In order to extract the temperature effect component （temperature strain ）, which presents the cycle trend, of dynamic strain signal in bridge detection through EMD, the multi order intrinsic mode function （IMF） constitution of temperature strain should be ascertained. Each order＇s IMF energy of signal is analyzed in time domain and frequency domain, and the threshold of IMF order is acquired through synthetic analysis of energy catastrophe order and correlation coefficient of marginal spectrum. Finally, the temperature effect component of strain data is extracted through selecting the cycle trend of each order＇s IMF reconstruction signal according to the threshold. The result shows that the periodic trend component of dynamic strain signal leads to order catastrophe of IMF energy, with which the value range of IMF order threshold can be preliminarily determined. By analyzing the energy distribution in frequency domain of IMF located in preliminarily determined threshold range, the correlation coefficient of IMF marginal spectrum increases rapidly to more than O. 8, which shows that the energy distribution patterns of last order IMF are consistent. Via the application of the in-situ data, the result demonstrates that the IMF order threshold can be precisely obtained by the first correlation coefficient of IMF marginal spectrum more than 0. 8. After extracting the temperature effect component based on the threshold, the live load information in dynamic strain is retained properly.
出处
《公路交通科技》
CAS
CSCD
北大核心
2015年第7期62-68,共7页
Journal of Highway and Transportation Research and Development
基金
国家自然科学基金项目(51478472)
湖南省教育厅一般项目(14C0214)
湖南省教育厅青年项目(13B010)
湖南城市学院大学生科技创新项目(2014XS39)
关键词
桥梁工程
温度应变
经验模态分解
IMF能量
边际谱
bridge engineering
temperature strain
empirical mode decomposition （EMD）
IMF energy
marginal spectrum
