摘要
提出一种基于时间序列数据挖掘的故障预报新方法。把故障前兆因子作为一种暂态,根据旋转机械轴承振动的实验数据建立时间序列．利用时延嵌入的方法重构状态空间,在状态空间中使用遗传算法搜寻最优暂态束．组成暂态集。用暂态集对旋转机械轴承振动的测试数据进行分析．判断是否为故障前兆因子．从而实现故障预报。
A new fault prediction method based on time series data mining is proposed. Fault symptoms are regarded as a sort of temporal patterns hidden in the time series formed by rotating machinery bearing vibration data. The time series is embedded into a reconstructed phase space with time-delay. In this phase space; genetic algorithms are used to search optimal temporal pattern clusters. The optimal collection is comprised of temporal pattern clusters and then is used to test the other bearing vibration data of the rotating machinery. Once the symptom is detected, the fault is forecasted.
出处
《南京航空航天大学学报》
EI
CAS
CSCD
北大核心
2006年第B07期120-123,共4页
Journal of Nanjing University of Aeronautics & Astronautics
基金
江苏省自然科学基金(BK2004021)资助项目
教育部科学技术研究(105088)资助项目。
关键词
时间序列
数据挖掘
旋转机械
故障预报
time series
data mining
rotating machinery
fault prediction
