首页|基于IWOA-VMD算法的管道泄漏信号降噪方法

基于IWOA-VMD算法的管道泄漏信号降噪方法

扫码查看
管道泄漏信号淹没在背景噪声中,会导致定位检测困难.改进鲸鱼算法(IWOA),实现对变分模态分解(VMD)算法中分解尺度和惩罚因子的组合寻优,并计算本征模态分量概率密度值与原始信号概率密度值之间的欧式距离,确定有效信号和噪声信号的分界点,进而实现信号重构.结果表明,仿真信号降噪后,不仅可检测到1 Hz的低频谐波信号,且检测到的其余频率谐波信号的频率和幅值更加精确;通过放油实验模拟现场泄漏信号,与其他降噪算法相比,IWOA-VMD算法的重构信号具有较大的信噪比和较小的均方误差,对缓慢泄漏信号的检测和定位具有较强的适应性.研究结果可为管道完整性管理水平的提升提供理论依据.
IWOA-VMD Algorithm Based Pipeline Leakage Signal Denoising Method
The pipeline leakage signal is submerged in the background noise,which leads to the difficulty of location detection.By improving whale optimization algorithm(IWOA)to optimize the combination of decomposition scale and penalty factor in variational mode decomposition algorithm(VMD),and by calculating the Euclidian distance between the probability density value of intrinsic mode component and the probability density value of the original signal,the boundary point between the effective signal and the noise signal is determined,and then the signal is reconstructed.The results show that not only the 1 Hz low-frequency harmonic signal is detected,but also the frequency and amplitude of other frequency harmonic signals are more accurate.The field leakage signal is simulated by oil discharge experiment.Compared with other signal denoising algorithm,the reconstructed signal of IWOA-VMD algorithm has larger signal-to-noise ratio and smaller mean square error,and has stronger adaptability to the detection and location of slow leakage signal.The research results can provide theoretical basis for the improvement of pipeline integrity management.

whale optimization algrithmvariational mode decompositionEuclidian distancenoise reductionsignal-to-noise ratio

杨振

展开 >

中国石油新疆油田公司,新疆克拉玛依 834000

鲸鱼算法 变分模态分解 欧式距离 降噪 信噪比

2024

石油化工自动化
中国石化集团宁波工程有限公司 全国化工自控设计技术中心站 中国石化集团公司自控设计技术中心站

石油化工自动化

影响因子:0.527
ISSN:1007-7324
年,卷(期):2024.60(3)
  • 14