.png){kind=icon}
.png){kind=icon}
Return
Feature Extraction for Load Identification Using Long-Term Operating Waveforms
.png){kind=icon}
OA
AI
DOI：10.1109/TSG.2014.2373314.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper introduces a novel finite-state-machine (FSM) representation of long-term load operating waveforms for feature extraction and load identification. An operating waveform is first converted into a quantized sequence of states. Each state is assigned with 2-D numerical values: root mean square (RMS) current values and staying time values. A set of elemental states and events are defined to reduce the number of states and extract numerical features to represent electric loads for classification and identification. Three major categories of repeating patterns in waveforms that correspond to repeating operating actions are summarized and identification methods are proposed for each such category. Test results using a large dataset of real-world waveforms show that the different appliances have distinct ranges of features extracted from the proposed FSM representation, and thus can be identified with high accuracy.
Keywords:
Direct load control
energy management
feature extraction
load identification
mode extraction
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
9.8
Papers:
5.7K
Citations:
4.3W