吉林大学学报(信息科学版) ›› 2021, Vol. 39 ›› Issue (2): 142-151.

• • 上一篇    下一篇

PSO-VMD-SVD 的改进去噪方法研究

杜 颖1a, 李 宏1a, 刘庆强1a, 路敬祎1b,1c, 李 富2   

  1. 1. 东北石油大学 a. 电气信息工程学院; b. 人工智能能源研究院; c. 黑龙江省网络化重点实验室, 黑龙江 大庆 163318;2. 大庆钻探工程公司 钻井一公司, 黑龙江 大庆 163318
  • 收稿日期:2020-10-10 出版日期:2021-04-19 发布日期:2021-04-26
  • 通讯作者: 李宏(1969— ), 女, 黑龙江大庆人,东北石油大学教授,硕士生导师,主要从事深度学习研究,(Tel)86-15304893939 (E-mail)853386766@qq.com
  • 作者简介:杜颖(1996— ), 女, 山东潍坊人,东北石油大学硕士研究生,主要从事油气管道泄漏检测研究,(Tel)86-15265627781(E-mail)1123106925@qq.com
  • 基金资助:
    国家自然科学基金资助项目(61873058; NSFC51575407); 中国石油科技创新基金资助项目(2018D-5007-0302); 黑龙江省自然科学基金资助项目(LH2020F005); 东北石油大学青年科学基金资助项目(2018QNL-33); 武汉科技大学冶金装备与控制教育部重点实验室开放基金资助项目(MECOF2019B01)

Research on Improved Method for Denoising of PSO-VMD-SVD

DU Ying1a, LI Hong1a, LIU Qingqiang1a, LU Jingyi1b,1c, LI Fu2   

  1. 1a. School of Electrical Engineering and Information; 1b. Artificial Intelligence Energy Research Institute;1c. Heilongjiang Provincial Key Laboratory of Networking and Intelligent Control, Northeast Petroleum University, Daqing 163318, China;2. Drilling Company 1, Daqing Drilling Engineering Company, Daqing 163318, China
  • Received:2020-10-10 Online:2021-04-19 Published:2021-04-26

摘要: 针对 VMD(Variational Mode Decomposition)在处理信号时, 模态个数 K 和惩罚参数α的选择会影响信号分解的效果, 提出一种改进 PSO(Particle Swarm Optimization)优化 VMD(PSO-VMD)的算法, 通过改进阈值相关系数法选择有效模态分量, 根据信号分析对高频噪声进行奇异值去噪重构。 对仿真信号和管道泄露信号进行实验, 通过选择多个K 并结合相关系数法选择模态进行重构, 比较不同 K 对应的重构信号与原信号的信噪比(SNR: Signal to Noise Ratio)、相关系数(CC: Coefficient)、 平方绝对误差(SAE: Square Absolute Error)、 均方误差(MSE: Mean Square Error)证明改进 PSO 优化 VMD 的可靠性。 将该方法与改进的 PSO-VMD 结合豪斯特夫距离和奇异值(HD-SVD: HustoffDistance-Singular Value Decompositio)、 互信息和奇异值(MI-SVD: Mutual Information-Singular Value Decompositio)、相关系数和小波变换(CC-WT: Correlation Coefficient-Wavelet Transform)等方法对比, 效果更佳。

关键词: VMD 算法, 粒子群优化, 相关系数, 奇异值, 去噪

Abstract: According to VMD(Variational Mode Decomposition) in processing the signal, the selection of mode number K and parameter α will affect the signal decomposition results, an optimized VMD(Variational Mode Decomposition) algorithm improving PSO(Particle Swarm Optimization) (PSO-VMD) was proposed. Based on the analysis of signal, high frequency noise is reconstructed by singular value denoising. Experiments on simulation signal and pipeline leakage signal, by selecting multiple K and combined with proposed the correlation method to select mode and reconstruct signal, compare SNR ( Signal to Noise Ratio ), correlation CC (Coefficient), SAE(Square Absolute Error), MSE(Mean Square Error) of reconstructed signal of different K and original signal, proved the reliability of optimizing VMD by improved PSO. The proposed method is compared with PSO-VMD combined with HD-SVD(Haussteff Distance and Singular Value), MI-SVD(Mutual Information and Singular Values), CC-WT(Correlation Coefficient and Wavelet Transform) and other methods.The proposed denoising method has the best effect.

Key words: VMD algorithm, particle swarm optimization, correlation coefficient, singular value, denoising

中图分类号: 

  • TN911