ANOMALY DETECTING METHOD, AND APPARATUS FOR THE SAME

摘要

In an anomaly detecting method by using multi-dimensional time series sensor signals including, generating anomaly model by using data of a learning period including neither that period nor any exclusion candidate period, calculating anomaly measurements on the basis of the distance from the normal model and, making a period containing the maximum anomaly measurement value but no exclusion candidate period, learning exclusion periods and anomaly determining thresholds are determined as learned data on the basis of the result in each round, generates anomaly model data in a learning period except learning-exclusion periods regarding acquired data or data in a designated evaluation period, an anomaly measurement at each time point is calculated on the basis of the distance from the normal model, and data at each time point is determined to be anomaly or normal by comparing the anomaly measurements with anomaly determining thresholds.

主权项

1. A method of detecting anomaly in a facility or an apparatus by using multi-dimensional time series sensor signals outputted from sensors attached to the facility or apparatus during operation, comprising:
a step of generating anomaly model by using sensor signals in a pre-designated period out of the multi-dimensional time series sensor signals and calculating anomaly determining thresholds; a step of extracting feature vectors from the multi-dimensional time series sensor signals as observed vectors; a step of calculating anomaly measurements of the observed vectors by using the extracted observed vectors and the generated normal model; and a step of detecting any anomaly in the facility or apparatus by comparing the calculated anomaly measurements of the observed vectors and the anomaly determining thresholds, wherein: the normal model is generated by using multi-dimensional time series sensor signals cleared of signals of some periods in the pre-designated period out of all the multi-dimensional time series sensor signals.