2023.09.22

USER ELECTRIC USAGE PATTERN CLASSIFICATION SYSTEM AND METHOD

Taiwan, Province of China

Overview

An electricity consumption pattern classification method in the present invention involves, initially, reading multiple training electricity usage consumption data sets. After which, perform a first round of classification with a first machine learning clustering algorithm according to the electricity consumption characteristics of the training electricity consumption data sets, generating multiple first-order electricity consumption data groups. Then, second-order feature values of the training electricity consumption data sets are generated with a feature extraction algorithm, and a second round of classification with a second machine learning clustering algorithm is performed according to the generating second-order electricity consumption data groups under the first-order electricity consumption data groups. When receiving a tobe-classified electricity consumption data set, the electricity consumption pattern classification result of the electricity consumption data set is determined according to the average similarity between the to-be-classified electricity consumption data set and each second-order electricity consumption data group. The second-order electricity consumption data groups produced by the two-stage classification accurately represent the different electricity consumption patterns in the data sets, thereby providing precise electricity consumption pattern classification results.

Category

環境感知與控制
用戶端能源管理
Appl. Type

發明

Status

Patented

Appl. No.

112136359

Patent No

發明第I850119號

Filing Date

2023.09.22

Expired Date

2043.09.21

Notification

2024.11.22