ABNORMALITY DETERMINATION MODEL GENERATION DEVICE, ABNORMALITY DETERMINATION DEVICE, ABNORMALITY DETERMINATION MODEL GENERATION METHOD AND ABNORMALITY DETERMINATION METHOD
To provide an abnormality determination model generation device, an abnormality determination device, an abnormality determination model generation method and an abnormality determination method capable of being generally applied without being aware of the type or the like of a monitoring object and...
Gespeichert in:
Hauptverfasser: | , |
---|---|
Format: | Patent |
Sprache: | eng ; jpn |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | To provide an abnormality determination model generation device, an abnormality determination device, an abnormality determination model generation method and an abnormality determination method capable of being generally applied without being aware of the type or the like of a monitoring object and performing abnormality determination with high accuracy.SOLUTION: An abnormality determination model generation device is configured to: generate a first abnormality determination model by calculating the average and variance of respective variables if the maximum value of a correlation between respective variables is less than a prescribed value about K-pieces L-dimensional vectors on an L-dimensional variable space; generate a second abnormality determination model by performing main component analysis to calculate a conversion coefficient of a main component if the maximum value of a correlation between the respective variables is equal to or more than the prescribed value; and generate a third abnormality determination model by constituting M-dimensional vectors composed of M types of variables at the same time and performing a main component analysis of the plurality of M-dimensional vectors to calculate a conversion coefficient of the main component if a time series signal during a normal operation is M types (M≥2).SELECTED DRAWING: Figure 1
【課題】監視対象の種類等を意識することなく汎用的に適用でき、かつ高精度な異常判定を行うことができる異常判定モデル生成装置、異常判定装置、異常判定モデル生成方法および異常判定方法を提供する。【解決手段】異常判定モデル生成装置は、L次元の変数空間上におけるK個のL次元ベクトルについて、各変数間の相関の最大値が所定値未満である場合は、各変数の平均および分散を演算することにより、第一の異常判定モデルを生成し、各変数間の相関の最大値が所定値以上である場合は、主成分分析を行って主成分の変換係数を演算することにより、第二の異常判定モデルを生成し、正常動作時の時系列信号が、M種(M≧2)である場合は、同一時刻におけるM種の変数からなるM次元ベクトルを構成し、複数のM次元ベクトルに対して主成分分析を行って主成分の変換係数を演算することにより、第三の異常判定モデルを生成する。【選択図】図1 |
---|