摘要
提出一种基于Kolmogrov-Smirnov检验和LS-SVM的机械设备故障预测新方法。基于K-S检验计算参考样本与正常状态样本经验分布函数的相似度,确定2个样本是否属于同一分布,即设备是否处于相同的运行状态,实现对设备运行退化状态进行识别,并采用当前退化状态与正常状态的K-S距离作为性能评估量化指标,在此基础上给出基于K-S检验和LS-SVM的设备故障预测系统框架。研究结果表明:该方法可以有效地对设备进行退化评估和故障预测,计算效率高,具有较好的适用性。
A novel performance degradation assessment method based on K-S test was presented. The similarity between empirical distribution function of normal condition sample and that of test sample was calculated by using Kolmogrov-Smirnov test, and then whether the two samples came from the same distribution, i.e., whether the equipment in the same state or not could be judged, so the degradation statements could be identified. The K-S distance between current state and normal state was calculated as the performance index to assess the degradation. According to that result, the prognostic system framework based on K-S test and LS-SVM was given. The result shows that the proposed method can realize the performance degradation assessment and prognostics effectively and prove to be adaptive.
出处
《中南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2016年第6期1924-1929,共6页
Journal of Central South University:Science and Technology
基金
国家自然科学基金资助项目(51405246)
江苏省自然科学基金资助项目(BK2011391)~~
关键词
故障预测
退化评估
K-S检验
最小二乘支持向量机
prognostics
degradation assessment
K-S test
least squares support vector machine
作者简介
通信作者：花国然，博士，教授，从事先进制造技术研究：E-mail：huagr@ntu．edu．cn
