基于改进小波阈值函数的桥梁应变监测数据降噪研究

Study on Noise Reduction of Bridge Strain Monitoring Data Based on Improved Wavelet Threshold Function

摘要:

为了减少噪声对桥梁应变监测信号的影响，进一步提高应变监测数据的可靠性，提出一种改进的小波阈值函数，通过改进软硬阈值折衷函数，构造了基于麻雀搜索（SSA）算法的适应度函数，并利用麻雀搜索算法对阈值函数中的调整因子进行寻优，克服了硬阈值函数的不连续以及软阈值函数存在恒定偏差的缺陷，同时也避免了传统试错法获取调整因子的缺陷；然后利用数值仿真软件建立了仿真信号，利用降噪效果评价指标，与其他改进阈值函数的去噪效果进行了比较，验证了该方法的降噪效果。研究表明：所提的桥梁应变监测信号降噪方法降噪效果更好，能较好地保留桥梁应变信息和基本特征。通过对兰家湾高墩钢栈桥应变监测数据降噪实例，验证了改进降噪方法的有效性。

Abstract:

In order to reduce the impact of noise on bridge strain monitoring signals and further improve the reliability of strain monitoring data, an improved wavelet threshold function is proposed. By improving the soft and hard threshold compromise function, a fitness function based on the Sparrow Search Algorithm (SSA) is constructed, and the Sparrow Search Algorithm is used to optimize and determine the adjustment factor in the threshold function, which overcomes the discontinuity of hard threshold functions and the constant deviation of soft threshold functions, and also avoid the shortcomings of traditional trial and error methods to obtain adjustment factors. Then, a simulation signal is established by using the numerical simulation software, and the denoising effect evaluation indicators are used to compare the denoising effect with the other improved threshold functions to verify the denoising effect of this method. The research shows that the denoising method proposed for bridge strain monitoring signals has a better denoising effect and can better preserve the bridge strain information and basic features. The effectiveness of the improved denoising method is verified by comparing the strain monitoring data of the high-pier steel trestle bridge in Lanjiawan.