航空学报 > 2019, Vol. 40 Issue (12): 323277-323277   doi: 10.7527/S1000-6893.2019.23277

基于隶属度和LMK-ELM的航空电子部件诊断方法

朱敏, 许爱强, 李睿峰, 戴金玲   

  1. 海军航空大学, 烟台 264001
  • 收稿日期:2019-07-09 修回日期:2019-08-23 出版日期:2019-12-15 发布日期:2019-09-16
  • 通讯作者: 许爱强 E-mail:hjhyautotest@sina.com
  • 基金资助:
    国家自然科学基金(11802338);山东省自然科学基金(ZR2017MF036)

Diagnosis method for avionics based on membership and LMK-ELM

ZHU Min, XU Aiqiang, LI Ruifeng, DAI Jinling   

  1. Naval Aviation University, Yantai 264001, China
  • Received:2019-07-09 Revised:2019-08-23 Online:2019-12-15 Published:2019-09-16
  • Supported by:
    National Natural Science Foundation of China (11802338); National Science Foundation of Shandong Province (ZR2017MF036)

摘要: 为提高航空电子部件模块级故障诊断精度,提出一种新的面向"软聚类"的局部多核学习(LMKL)-超限学习机(ELM)离线诊断方法。通过引入模糊C均值聚类对样本空间进行模糊划分,挖掘聚类内部多样性的同时,实现了对过学习的抑制;将模糊划分产生的隶属度信息融入LMKL-ELM的优化过程,运用基于初始-对偶混合优化问题的三步优化策略克服了局部核权重二次非凸的问题,在l1-范数与l2-范数约束下分别给出了相应的更新方法。将所提方法应用于某型机前端接收机,结果表明:与4种流行的多核诊断方法相比,该方法可有效避免漏警、抑制虚警,在l1-范数和l2-范数约束下,其诊断精度比其他方法的平均值分别提升了4.09%和5.13%。

关键词: 超限学习机, 局部多核学习, 模糊C均值聚类, 故障诊断, 航空电子

Abstract: To improve the accuracy of module-level fault diagnosis for avionics, a new off-line diagnosis method based on soft-clustering-sensitive Localized Multi-Kernel Learning (LMKL) and Extreme Learning Machine (ELM) is proposed. By introducing fuzzy C-means clustering to partition the sample space, the over-learning is suppressed while mining the diversity within the cluster. The membership information generated by the fuzzy partition is integrated into the optimization process of LMKL-ELM. A three-step optimization strategy based on the initial-dual hybrid optimization problem is used to overcome the quadratic non-convexity of the local kernel weights. The corresponding updating methods for these weights are given under l1-norm constraint and l2-norm constraint. The proposed method is applied to the front-end receiver. Compared with four popular multi-kernel diagnostic algorithms, the results show that the proposed method can effectively avoid missing alarm and suppress false alarm. The diagnostic accuracy is 4.09% higher in l1-norm and 5.13% higher in l2-norm than the average of other methods.

Key words: extreme learning machine, localized multiple kernel learning, fuzzy C-means clustering, fault diagnosis, avionic

中图分类号: