非正态变异下的非线性轮廓异常点识别方法研究

doi:10.12005/orms.2019.0013

摘要/Abstract

摘要： 在质量控制领域,非线性轮廓异常点识别问题是重点研究问题之一。本文综合运用了小波分析、数据深度、聚类分析等数据分析处理技术,提出了一种新的非正态变异的异常点识别方法。文章通过仿真分析技术,将新方法χ²与控制图方法进行性能对比,结果证实新方法能够以更高的准确率和稳定性识别异常点,表现出更好的异常点识别性能。最后将新方法应用于木板垂直密度轮廓实例对新方法进行验证,分析结果表明本方法能够有效识别出异常轮廓数据。

关键词: 异常点识别, 小波降噪, 马氏深度, 聚类分析

Abstract: Outliers detection on nonlinear profiles is one of the key problems in the field of quality control. This paper proposes a new outlier detection method for abnormal variation based on data process technologies which are wavelet analysis, Mahalanobis depth and cluster analysis. By simulation analysis, the performances of the new method and control chart method are compared and it is proved that the new method can identify the outliers with high accuracy and stability which performs outlier detection better. Finally, the new method is applied to a real data set consisting of vertical density profile. The results show that the new method can effectively identify the abnormal profiles data.

Key words: outliers detection, wavelet analysis, mahalanobis depth, cluster analysis

中图分类号:

C931.1

聂斌, 王曦, 胡雪. 非正态变异下的非线性轮廓异常点识别方法研究[J]. 运筹与管理, 2019, 28(1): 101-107.

NIE Bin, WANG Xi, HU Xue. Nonlinear Profile Outlier Detection for Non-Normal Variation[J]. Operations Research and Management Science, 2019, 28(1): 101-107.

参考文献

[1] Woodall W H, Spitzner D J, Montgomery D C, et al. Using control charts to monitor process and product quality profiles[J]. Journal of Quality Technology, 2004, 36(3): 309-320.
[2] Hajji H. Statistical analysis of network traffic for adaptive faults detection[J]. IEEE Trans. Neural Networks, 2005, 16(5): 1053-1063.
[3] Manikopoulos C, Papavassiliou S. Network intrusion and fault detection: a statistical anomaly approach[J]. IEEE Comm. Magazine, 2002, 40(10): 76-83.
[4] Song D, Heywood M I, Zincir-Heywood A N. Training genetic programming on half a million patterns: an example from anomaly detection[J]. IEEE Trans. Evolutionary Computation, 2005, 9(3): 225-239.
[5] Thottan N,Ji C. Anomaly Detection in IP Networks[J]. IEEE Trans. Signal Processing, 2003, 51(8): 2191-2204.
[6] Hu W, Xiao X, Fu Z, et al. A system for learning statistical motion patterns[J]. IEEE Trans. Pattern Analysis and Machine Intelligence, 2006, 28(9): 1450-1464.
[7] Stauffer C, Grimson W E. Learning patterns of activity using real-time tracking, 2000, 22(8): 747-757.
[8] Carlotto M J. A cluster-based approach for detection man-made objects and changes in imagery[J]. IEEE Trans. Geoscience and Remote Sensing, 2005, 43(2): 374-387.
[9] Kwon H, Nasrabadi N M. Kernel RX-algorithm: a nonlinear anomaly detector for hyperspectral Imagery[J]. IEEE Trans. Geoscience and Remote Sensing, 2005, 43(2): 388-397.
[10] Frakt A B, Karl W C, Willsky A S. A Multiscale hypothesis testing approach to anomaly detection and localization from noisy tomographic data[J]. IEEE Image Processing, 1998, 7(6): 825-837.
[11] Kang L, Albin S L. On-line monitoring when the process yields a linear profile[J]. Journal of Quality Technology, 2000, 32(4): 418-426.
[12] Mahmoud M A, Parker P A, Woodall W H. A change point method for nonlinear profile data[J]. Quality and Reliability Engineering International, 2007, 23(2): 247-268.
[13] Febrero M, Galeano P, Gonzalez M W. Outlier detection in functional data by depth mesures, with application to identify abnormal levels[J]. Environmetrics, 2008, 19(4): 331-345.
[14] Martinez J, Saavedra A, Garcia-Nieto P J, et al. Air quality parameters outliers detection using functional data analysis in the langreo urban area(northern spain)[J]. Applied Mathematics & Computation, 2014, 241(3): 1-10.
[15] Carlo S, Pedro G, Rosa E L. Functional outlier detection by a local depth with application to levels[J]. Stochastic Environmental Research and Risk Assessment, 2016, 30(4): 1-16.
[16] Diaz M C, Garcia Nieto P J, Alonso Fernandez J R, et al. Detection of outliers in water quality monitoring samples using functional data analysis in San Esteban estuary(Northern Spain)[J]. Science of the Total Environment, 2012, 439(1): 54-61.
[17] 郝文广,丁常富,梁娜.小波降噪与FFT降噪比较[J].电力科学与工程,2011,27(3):59-61.
[18] 葛哲学,陈仲生.Matlab时频分析技术及其应用[M].北京:人民邮电出版社,2006:243.
[19] Canan Hamurkaroglu, Mehmet Mert, Yasemin Saykan. Nonparametric control charts based on mahalanobic depth[J]. Journal of Mathematics and Statistics, 2004, 33: 57-67.
[20] Zhang Hang, Susan Albin. Detecting outliers in complex profiles using a control chart method[J]. IIE Transactions, 2009, 41(4): 335-345.
[21] Walker E, Wright S. Comparing curves using additive models[J]. Journal of Quality Technology, 2002, 34(1): 118-129.