PE-VMD与小波阈值的干涉型光纤联合去噪方法  

Joint denoising method for interferic fibers with PE-VMD and wavelet thresholds

作　　者：徐隆 杨军 周龙 龙彪 Xu Long;Yang Jun;Zhou Long;Long Biao(College of Electronic Engineering,Guangxi Normal University,Guilin 541004,China)

机构地区：[1]广西师范大学电子工程学院,桂林541004

出　　处：《国外电子测量技术》2022年第10期39-46,共8页Foreign Electronic Measurement Technology

基　　金：桂林市重点研发计划项目(2020010329);广西师范大学重大科技成果转化培育项目(2020PY002)资助。

摘　　要：干涉型分布式光纤传感系统测量精度高、范围广,适用于能源和周界安防等重要领域,但存在有用信号易被噪声污染影响测量准确度的问题,基于此,提出了一种排列熵(PE)优化变分模态分解(VMD)与小波阈值(WT)的联合去噪方法。首先,根据排列熵相关参数确定信号的变分模态分解层数k,做VMD分解得到k个固有模态函数(IMF)分量;再对k个IMF分量做小波阈值去噪,最后对去噪后的IMF分量进行重构;得到原始信号。通过仿真实验证明该方法相比于经验模态分解-小波阈值去噪(EMD-WT)、集合经验模态分解-小波阈值去噪(EEMD-WT),变分模态分解-小波阈值去噪(VMD-WT)等方法信噪比更高、均方误差更小,平均信噪比提升了2 dB,最后通过搭建的系统采集实际信号进行去噪处理,引入降噪误差比指标,通过实验结果验证了方法的有效性。Interferometric distributed optical fiber sensing system has high measurement accuracy and wide range, and is suitable for important fields such as energy and perimeter security, but there is a problem that the useful signal is easily affected by noise pollution, based on this, this paper proposes a joint denoising method of permutation entropy optimized variational modal decomposition and wavelet threshold. Firstly, according to the permutation entropy-related parameters, the number of variational modal decomposition layers k of the signal is determined, the VMD is decomposed to obtain k IMF components, and then the wavelet threshold denoising is done for k IMF components, and finally the IMF components after denoising are reconstructed to obtain the original signal. Simulation experiments show that compared with the empirical modal decomposition-wavelet threshold denoising(EMD-WT), collective empirical modal decomposition-wavelet threshold denoising(EEMD-WT), variational modal decomposition-wavelet threshold denoising(VMD-WT) etc., the signal-to-noise ratio is higher, the mean square error is smaller, and the average signal-to-noise ratio is increased by 2 dB,finally, the actual signal is collected by the system for denoising, and the noise reduction error ratio index is introduced, and the effectiveness of the proposed method is verified by experimental results.

关 键 词：干涉型分布式光纤 变分模态分解 排列熵 小波阈值 去噪 

分 类 号：TN247[电子电信—物理电子学] TN713