首页> 外国专利> TIME SERIES DATA ANALYSIS METHOD AND TIME SERIES DATA ABNORMALITY MONITOR METHOD

TIME SERIES DATA ANALYSIS METHOD AND TIME SERIES DATA ABNORMALITY MONITOR METHOD

机译:时序数据分析方法和时序数据异常监测方法

摘要

PROBLEM TO BE SOLVED: To detect recovery of abnormality of time series data quickly.SOLUTION: The time series data analysis method for analyzing time series data from an object to be monitored and controlled by a successive trajectory parallel measurement method is configured to embed the time series data into a n-dimensional state space, and calculate a data vector selected from the embedded time series data and a tangential direction of the trajectory of the embedded time series data in the vicinity vector of the data vector. Then parallel degree between the tangential direction of the trajectory on the data vector and the tangential direction of the trajectory on the vicinity vector is calculated, and the time series data is analyzed by using any one of a differential value (index 1) for every predetermined time of the calculated parallel degree, an increase amount (index 2) for every predetermined time of the calculated parallel degree, and an accumulated value (index 3) of the increase amount for every predetermined time of the calculated parallel degree, as an index.
机译:解决的问题:快速检测时间序列数据的异常恢复解决方案:将时间序列数据分析方法配置为嵌入时间,该方法用于通过连续轨迹并行测量方法分析要监视和控制的对象的时间序列数据将序列数据放入n维状态空间,并计算从嵌入的时间序列数据和嵌入的时间序列数据的轨迹的切线方向在数据向量的邻近向量中选择的数据向量。然后计算数据向量上轨迹的切线方向与邻近向量上轨迹的切线方向之间的平行度,并对每个预定时间使用微分值(索引1)中的任何一个分析时间序列数据将计算出的平行度的时间,每个计算出的平行度的预定时间的增加量(指标2),以及计算出的平行度的每预定时间的增加量的累加值(指标3)作为指标。

著录项

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号