Gateway electric energy meter data anomaly detection method and system based on pseudo abnormal point identification

• Main Authors: ZHANG RAN, XU YANJUN, WANG JIA, XUE LISI, WU MENG, ZHONG LI, XIE ZHI, WANG JIAJU, LIU CHEN, BAI TAI
• Format: Patent
• Language: Chinese; English
• Published: 12.01.2021
• Subjects: MEASURING; MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES; PHYSICS; TESTING

The invention discloses a gateway electric energy meter data exception detection method and system based on pseudo exception point identification, and the method comprises the steps: dividing a data exception mode into two types: a random exception point mode and a continuous exception point mode; for a random abnormal point mode, according to the characteristic that the power load data has the regularity, performing time sequence decomposition and autocorrelation analysis on the data curve to realize the random abnormal point detection, thereby avoiding abnormal detection interference data caused by periodic power consumption behaviors of a user; for a continuous abnormal point mode, according to abnormal power utilization characteristics of a power user, extracting two characteristics ofa line loss rate and an electric quantity continuous descending trend based on historical data, training a user electricity stealing behavior detection model, thereby avoiding the abnormal detectioninterference data brought