Abstract：Large quantity and ambiguity of oil atomic spectrometric information greatly affects the applicable efficiency and accuracy in fault diagnosis. A novel method for choosing less and effective spectrometric features is presented. Based on gearbox test bed，we simulated the normal wear state and two typical faults to acquire the lubricant samples. The three wear states are regarded as three vague sets，and spectrometric feature values are vague values on vague sets. Based on similarity between vague values，mean vague sensibility(MVS) is defined to describe the sensitive degree of spectrometric feature to wear state. Besides，the membership degrees of vague sets greatly depend on human experience. The probability density distribution of spectrometric data of three wear states was estimated with Parzen window. Combined with Bayesian formula, the range of vague sets membership was calculated. Experimental results verify that the proposed method is of efficient help in choosing high fault-sensitive features from so many spectrometric features.
Key words：Oil atomic spectrometry;Feature selection;Vague sets;Parzen window;Bayesian formula
[1] GAO Jing-wei，ZHANG Pei-lin，REN Guo-quan，et al(高经纬，张培林，任国全，等). Chinese Internal Combustion Engine Engineering(内燃机工程)，2004，25(6)：34. [2] LIU Yan，LI Shi-qi，DUAN Xue-yan(刘 燕，李世其，段学燕). Transactions of CSICE(内燃机学报)，2008，26(4)：374. [3] Gau W L，Buehrer D J. Vague Sets. IEEE Trans. on Systerns，Man，and Cybernetic，1993，23(2)：610. [4] ZHENG Chun-hui，LI Ti-fang，GUI Jian-zhou(郑春辉，李体方，归建洲). Aeronautical Computing Technique(航空计算技术)，2008，38(2)：34. [5] YE Jun(叶 军). Proceedings of the CSEE(中国电机工程学报)，2006，26(1)：17. [6] ZHANG Qing-hua(张清华). Journal of Electronics & Information Technology(电子与信息学报)，2007，29(8)：1855. [7] YAN De-qin，CHI Zhong-xian，LI Yan-hong(闫德勤，迟忠先，李艳红). Journal of Pattern Recognition and Artiftcial Intelligence(模式识别与人工智能)，2004，17(1)：22. [8] Sergios Theodoridis，Konstantions Koutroumbas. Patten Recognition，Third Edition(模式识别，第3版). Beijing：Publishing House of Electronics(北京：电子工业出版社)，2006. 26. [9] LIN Zhi-gui，XU Li-zhong，LIU Ying-ping(林志贵，徐立中，刘英平). Information and Control(信息与控制)，2005，34(1)：54. [10] SHENG Zhou，XIE Shi-qian，PAN Cheng-yi(盛 骤，谢式千，潘承毅). Probability Theory and Mathematical Statistics(概率论与数理统计). Beijing：Higher Education Press(北京：高等教育出版社)，2001. 22.
