A non-intrusive load detection and decomposition method based on a state matrix decision tree

• Main Authors: SU LUMEI, ZHANG BAOQIONG, ZHU WENTING, ZHENG XIAOLONG, DENG GUANSEN, ZHENG RUIJIE
• Format: Patent
• Language: Chinese; English
• Published: 26.02.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; HANDLING RECORD CARRIERS; MEASURING; MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; TESTING

The invention relates to a non-intrusive load detection and decomposition method based on a state matrix decision tree, which comprises the following steps: S1, preprocessing sample data, including data cleaning, data integration and data reduction, to obtain effective sample data; S2, determining the data sample period by using the spectrum analysis; S3, selecting load features based on the sequential forward feature selection algorithm and the K-means clustering algorithm, and extracting load features with high discrimination by utilizing the sequential feature selection algorithm accordingto the sample period; S4, based on load identification and decomposition of the improved sliding window bilateral CUSUM event detection algorithm and a decision tree, establishing a single equipment working state model for automatic identification. On this basis, the state matrix decision tree is introduced to establish a load time series feature probability model, so as to realize the automatic identification of overlaid