ANOMALY AGGREGATION METHOD

PROBLEM TO BE SOLVED: To provide a method for identifying anomalous values. SOLUTION: A method for aggregating anomalous values comprises obtaining operational data from at least one machine and calculating at least one exceptional anomaly score from the acquired operational data. The exceptional an...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SENTURK-DOGANAKSOY DENIZ, TRAVALY ANDREW J, RUCIGAY RICHARD J, SKOWRONEK PETER T, LACOMB CHRISTINA A
Format: Patent
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:PROBLEM TO BE SOLVED: To provide a method for identifying anomalous values. SOLUTION: A method for aggregating anomalous values comprises obtaining operational data from at least one machine and calculating at least one exceptional anomaly score from the acquired operational data. The exceptional anomaly scores can be aggregated to identify acute or chronic anomalous values. The exceptional anomaly scores can be aggregated by calculating at least one magnitude anomaly measure. The magnitude anomaly measure indicates the average value of the exceptional anomaly scores over a predetermined time width. The exceptional anomaly scores can also be aggregated by calculating at least one frequency anomaly measure. The frequency anomaly measure indicates a percentage of time width having anomalous values out of the predetermined time width. COPYRIGHT: (C)2009,JPO&INPIT