摘要
传统的机械设备故障率预测方法正确率低,已不能适应现代化设备的检修需求。本文在探讨ACO和LSSVM算法的基础上,提出一种新的PHM算法。利用时间序列预测法计算出季节因子并结合ACO-LSSVM算法对航空某设备的故障率进行建模,得到较好的实验结果,并给出预测结果和实际结果的对比分析。
The traditional prediction method of mechanical equipment failure rate is of lower accuracy,it is unable to adapt the demand of modern equipment maintenance.This paper proposes a novel PHM algorithm based on ACO and LSSVM algorithms.Using time series analysis prediction method to calculate seasonal factor and combining with ACO-LSSVM algorithm to model the failure rate of an aviation device,a good experimental result is obtained,and the comparative analysis of predicted result and actual result is given.
引文
[1]张宝珍.预测与健康管理技术的发展及应用[J].测控技术,2008,27(2):5-7.
[2]孙博,康锐,谢劲松.故障预测与健康管理系统研究和应用现状综述[J].系统工程与电子技术,2007,29(10):1762-1767.
[3]彭宇,刘大同,彭喜元.故障预测与健康管理技术综述[J].电子测量与仪器学报,2010,24(1):1-9.
[4]曾声奎,Michael G Pecht,吴际.故障预测与健康管理(PHM)技术的现状与发展[J].航空学报,2005,26(5):626-632.
[5]黄飞.时间序列法预测某型飞机故障率[J].燃气涡轮试验与研究,2001,14(1)30-32.
[6]Billinton R,Chen H,Ghajar R.Time-series models for re-liability evaluation of power systems including wind energy[J].Microelectronics and Reliability,1996,36(9):1253-1261.
[7]丁明,张立军,吴义纯.基于时间序列分析的风电场风速预测模型[J].电力自动化设备,2005,25(8):32-34.
[8]陈栋梁.支持向量机训练算法研究[D].合肥:合肥工业大学,2007.
[9]Suykens J A K,Vandewalle J.Least squares support vec-tor machine classifiers[J].Neural Processing Letters,1999,9(3):293-300.
[10]段海滨,王道波,朱家强,等.蚁群算法理论及应用研究的进展[J].控制与决策,2004,19(12):1321-1326,1340.
[11]Marco Dorigo,Gianni Di Caro.The ant colony optimization meta-heuristic[J].New Ideas in Optimization,1999,26(1):29-41.
[12]段海滨.蚁群算法原理及应用[M].北京:科学出版社,2005.
[13]倪丽萍,倪志伟,李锋刚,等.基于蚁群算法的SVM模型选择研究[J].计算机技术与发展,2007,17(9):95-98.
