Appliance Identification via Power Signal Harmonic Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing non-intrusive load monitoring (NILM) technologies face challenges in identifying individual appliances, especially when appliances with similar power consumption are present, and require separate hardware and signal processing for high-speed sampling.
Innovation Solution
A method and device for identifying operating appliances by analyzing harmonic components of a power signal, which involves collecting a mixed load signal, generating candidate appliance lists based on harmonic component ratios and magnitudes, and filtering to identify the operating appliance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If low-speed sampling method is used for NILM, then existing power meters can be used and separate hardware is not required, but it is difficult to identify operating appliances when appliances with similar power consumption exist
Solution Approach 1:
The patent changes the parameter being analyzed from basic power consumption to harmonic component ratios. By extracting and comparing harmonic ratios (e.g., H3/H1, H5/H1) of the mixed signal with pre-stored appliance harmonic profiles, the system achieves accurate appliance identification using low-speed sampling without requiring additional hardware, thus resolving the contradiction between device simplicity and identification accuracy.
2Measurement precision
If high-speed sampling method is used for NILM, then operating appliances can be identified by extracting specific signals, but separate hardware and signal processing functions are required
Solution Approach 1:
The patent extracts only the essential harmonic ratio features from the power signal rather than performing comprehensive high-speed signal processing. By calculating and comparing harmonic ratios (such as third-to-fundamental and fifth-to-fundamental ratios) with pre-stored appliance profiles, the system achieves accurate identification without requiring separate high-speed sampling hardware or complex signal processing functions, thus resolving the contradiction between identification accuracy and device complexity.
3Measurement precision
If harmonic component analysis is performed to identify appliances with similar power consumption, then identification accuracy is improved, but the system must handle noise and similar harmonic components
Solution Approach 1:
The patent uses a feedback mechanism by comparing the calculated harmonic ratios of the mixed signal with pre-stored harmonic ratio profiles of individual appliances. The system iteratively identifies which appliance combination best matches the observed harmonic ratios, allowing it to distinguish between appliances with similar power consumption and reject noise or similar harmonic components through this comparative validation process.
Data Source
AI summary
Provided are a method and device for identifying an individual appliance through analysis of a harmonic component of a power signal. An operating appliance identification method includes collecting a mixed load signal from a home power distribution network, generating a first candidate appliance list by selecting first candidate appliances among individual appliances registered in an individual appliance harmonic analysis information table based on a ratio of a harmonic component value included in the mixed load signal, generating a second candidate appliance list including second candidate appliances by filtering the first candidate appliances based on a magnitude of the harmonic component value included in the mixed load signal, and identifying an operating appliance among a plurality of appliances connected to the home power distribution network by using the second candidate appliance list.