计算机应用 ›› 2015, Vol. 35 ›› Issue (11): 3252-3255.DOI: 10.11772/j.issn.1001-9081.2015.11.3252

• 人工智能 • 上一篇    下一篇

基于健康度分析与和声蚁群算法-支持向量机的故障预测模型

邱文昊1, 黄考利2, 金赛赛3, 连光耀2   

  1. 1. 军械工程学院 四系, 石家庄 050003;
    2. 军械技术研究所 一室, 石家庄 050003;
    3. 航天飞行控制中心, 北京 100094
  • 收稿日期:2015-06-04 修回日期:2015-08-04 发布日期:2015-11-13
  • 通讯作者: 邱文昊(1990-),男,山东滕州人,硕士研究生,主要研究方向:故障预测、健康管理.
  • 作者简介:黄考利(1958-),男,山东潍坊人,研究员,博士,主要研究方向:故障预测、测试性验证; 金赛赛(1990-),男,山东济宁人,助理工程师,硕士,主要研究方向:故障预测与处理; 连光耀(1980-),男,河南宝丰人,高级工程师,博士,主要研究方向:装备测试与诊断.
  • 基金资助:
    国防预先研究科研项目(51327030104).

Fault prediction model based on health analysis and harmony search-ant colony algorithm-support vector machine

QIU Wenhao1, HUANG Kaoli2, JIN Saisai3, LIAN Guangyao2   

  1. 1. Department No.4, Ordnance Engineering College, Shijiazhuang Hebei 050003, China;
    2. Research Room No.1, Ordnance Technological Research Institute, Shijiazhuang Hebei 050003, China;
    3. Space Flight Control Center, Beijing 100094, China
  • Received:2015-06-04 Revised:2015-08-04 Published:2015-11-13

摘要: 针对现有的故障预测技术无法从整体上反映系统性能下降趋势等问题,提出一种基于健康度分析的故障预测方法.首先,在支持向量机回归算法基础上构造多输出支持向量机,以实现健康度的多步预测,并提出一种和声蚁群算法优化支持向量机参数,解决了蚁群算法易陷入局部最优的问题; 然后, 根据最优参数建立拟合监测数据和未来健康度下降过程非线性映射关系的和声蚁群算法-支持向量机(HSACA-SVM)故障预测模型; 最后,通过某装备电源系统监测数据验证了该模型的有效性.实例验证表明该模型能够较好地实现对健康度下降趋势的预测,预测准确率达到97%,进而实现故障预测.

关键词: 健康度分析, 和声算法, 蚁群算法, 多输出支持向量机, 故障预测

Abstract: A new method of fault prediction based on health analysis was proposed for the problem of the existing fault prediction technology could not response the declining trend of system property as whole. Firstly, in order to achieve multi-step prediction, multiple output Support Vector Machine (SVM) was formatted on the basis of support vector machine regression algorithm, while using the Harmony Search-Ant Colony Algorithm (HSACA) to optimize parameters of SVM to solve the local optimal problem. Then nonlinear mapping Harmony Search-Ant Colony Algorithm-Support Vector Machine (HSACA-SVM) model matching monitoring data and health degree was built with the optimal parameters. Finally, the proposed model was used to evaluate a power supply system. The results indicate that the HSACA-SVM model can predict the downward trend of health degree with 97% accuracy, and then realize fault prediction.

Key words: health analysis, Harmony Search Algorithm (HSA), Ant Colony Algorithm (ACA), multiple output Support Vector Machine (SVM), fault prediction

中图分类号: