Bhattacharyya距离
希尔伯特-黄变换
泄漏(经济)
降噪
信号(编程语言)
管道(软件)
算法
相似性(几何)
干扰(通信)
计算机科学
模式识别(心理学)
数学
人工智能
滤波器(信号处理)
计算机视觉
电信
图像(数学)
频道(广播)
宏观经济学
经济
程序设计语言
作者
Jingyi Lu,Jikang Yue,Lijuan Zhu,Gongfa Li
出处
期刊:Measurement
[Elsevier]
日期:2019-11-19
卷期号:151: 107283-107283
被引量:51
标识
DOI:10.1016/j.measurement.2019.107283
摘要
In order to better eliminate the noise interference in pipeline leakage detection, a method based on the combination of variational mode decomposition (VMD) and improved Bhattacharyya distance (BD) is proposed. The method first performs VMD decomposition on the input signal to obtain K intrinsic mode functions (IMFs), and calculates the variance of each IMF component and the original signal respectively. The improved Bhattacharyya distance is used to measure the similarity between the two variances to distinguish between the effective IMF components and the non-effective IMF components. Finally, the filtered signal is reconstructed by using the effective IMF components. The filtering results of simulation signals show that this method has better filtering effect than other filtering methods of EMD and VMD. The method is successfully applied to the filtering of pipeline leakage signals. The effectiveness and feasibility of this method in the filtering of pipeline leakage signal are verified by experimental comparison.
科研通智能强力驱动
Strongly Powered by AbleSci AI