INTELLIGENT DATA-ANALYZING METHOD FOR DETECTING ABNORMAL USE OF UTILITIES IN BUILDINGS

摘要

PROBLEM TO BE SOLVED: To detect the extraordinary use of utilities in a building.

SOLUTION: The use of the utilities is periodically measured all day long every day, and the measured values are grouped according to a day having a similar average utility consumption level. Data on each group are statistically analyzed by a generalized extreme studentized deviate (GESD) method. This method identifies outliers of data samples remarkably different from a large part of the data. A rate at which each outlier strays away from the remainder of the data represents the importance in the abnormal use of the utilities represented by the outliers. The resulting outlier information is readily descernable by the operators in a building when the necessity of further investigation is judged from the cause of a specific accident of the abnormal use of the utilities.

COPYRIGHT: (C)2003,JPO