斜拉桥应变监测数据预处理方法及应用

Data Preprocessing Method and Its Application for Strain Monitoring of Cable-stayed Bridge

摘要:

为解决大跨度斜拉桥监测数据存在的高频噪声、局部缺失及温致干扰等问题，采用数据重采样、缺失数据填补与温致响应分离的预处理流程。首先，通过时间窗口平均法将原始高频数据重采样至2s间隔以实现数据量压缩与关键趋势特征保留。其次，基于时间戳对齐自动识别缺失点并采用五点加权平滑法完成缺失数据稳健填补。最后，利用经验模态分解技术自适应分离信号中的温致应变分量以消除环境温度波动影响。经预处理的应变数据成功应用于该桥梁结构状态预警分析，验证了所提方法的工程适用性，有效提升了监测数据质量，为桥梁长期安全评估与维护决策提供了可靠的数据支撑。

Abstract:

To address the issues of high-frequency noise, local missing and temperature-induced interference in the monitoring data of long-span cable-stayed bridges, a preprocessing workflow integrating data resampling, missing data filling and temperature-induced response separation is adopted. Firstly, the original high-frequency data are resampled to a 2 s interval via the time-window averaging method, realizing both data volume compression and retention of key trend features. Secondly, the missing points are automatically identified through timestamp alignment, and the robust filling of missing data is accomplished using the five-point weighted smoothing method. Finally, the empirical mode decomposition (EMD) technique is adopted to adaptively separate the temperature-induced strain components from the signals, thus eliminating the impact of ambient temperature fluctuations. The preprocessed strain data have been successfully applied to the structural condition early-warning analysis of the bridge, which verifies the engineering applicability of the proposed method, effectively improves the quality of monitoring data and provides the reliable data support for the long-term safety assessment and maintenance decision-making of bridges.