桥梁长期健康监测大数据温度与应变特征及关联性分析

针对在珠江黄埔大桥北汊斜拉桥稳定运营阶段采集到的温度和应变大数据时间序列,进行特征分析,获取了温度和应变数据的基本特征,初步探索了二者的关联性。进一步利用傅里叶变换和小波变换对监测数据进行分析。结果表明温度与应变数据存在周期为24 h和12 h的频域成分,并且温度与应变之间有明显的正相关性。最后,建立了斜拉桥的整体和关键局部梁段有限元模型;结合小波变换的小波细节层信息,验证了24 h周期的温度与应变信号之间的关联性,证明了日温对应变有显著的影响。对桥梁长期健康监测温度和应变大数据所采用的数据分析方法及有限元验证方法可为桥梁长期健康监测数据的分析和数据挖掘提供有价值的参考。

Analysis of Characteristics and Correlation for Temperature and Strain Based on Long-Term Bridge Health Monitoring Big Data

In this paper, analysis of time series of temperature and strain data collected in the stable operation stage of the north span, which is a cable-stayed bridge, of Huangpu Bridge of Pearl River is accomplished to obtain the basic characteristics of the temperature and the strain data. Preliminary exploration is also carried out on the relevance between temperature and strain. Subsequently, both the Fourier transform and the wavelet transform are used to analyze the monitoring data, indicating that the temperature and the strain data have two components with a period of 24 h and 12 h, respectively in frequency domain, and there exists a remarkable positive correlation between temperature and strain. Consequently, finite-element models are established for the whole bridge and a key part of the bridge as well. Simulations for the finite-element model of the key part of the bridge, in conjunction with the wavelet detail information extracted from the wavelet transform, verify the correlation between the temperature and the strain components with a period of 24 h. It is demonstrated that daily temperature has a significant effect on strain. The methods for coping with the temperature and the strain data in this paper provide a valuable reference for data analysis and finite-element verification in long-term bridge health monitoring.