Non-Invasive Power Device Identification via Motif Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional non-invasive monitoring techniques face challenges in identifying individual power devices in a circuit due to differences in step responses and AC power characteristics, leading to increased costs in Smart Grid implementations.
Innovation Solution
A method and device for non-invasively analyzing power devices by acquiring power pattern information based on ON/OFF states, classifying similar patterns, and determining motif information to identify specific power devices through symmetrical average power variations, allowing for efficient data selection and storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a smart tag is provided for each outlet or power device in a home, then power consumption measurement accuracy is improved, but manufacturing and installation costs increase considerably
Solution Approach 1:
The patent merges multiple smart tags into a single smart tag that monitors power consumption for multiple outlets simultaneously. The smart tag is installed at the service panel and measures total power consumption, then uses signal processing to decompose and identify individual device power patterns, eliminating the need for multiple separate smart tags while maintaining measurement accuracy
Solution Approach 2:
The smart tag is designed with multi-functionality to serve multiple outlets and power devices through a single installation. It performs total power measurement, individual device pattern recognition, and power consumption allocation across multiple outlets, replacing what would traditionally require multiple dedicated smart tags
2Device complexity
If conventional non-invasive monitoring techniques are used to identify power devices, then device complexity is reduced, but identification accuracy deteriorates due to differences in step responses and AC power characteristics
Solution Approach 1:
The system performs preliminary action by collecting and storing power consumption patterns from each power device during normal operation before actual identification is needed. These baseline patterns are saved in memory and used for comparison during monitoring, enabling accurate identification without complex real-time analysis
Solution Approach 2:
The system uses feedback by continuously comparing real-time power consumption measurements against stored baseline patterns and adjusting identification decisions based on pattern matching results. The signal processing algorithm refines identification accuracy through iterative comparison and analysis of power pattern characteristics
Data Source
AI summary
A method and apparatus for analyzing power devices in a circuit are disclosed. In one embodiment, a power analysis apparatus analyzes information on power supplied to the circuit and classifies the power patterns into groups of their own similar power patterns by making reference to the information on the power patterns, so as to acquire at least one piece of motif information which is information on at least one fingerprint. The apparatus further counts the frequency of occurrence of each motif and determines a pair of specific motifs having the difference between the counted frequencies of occurrence, which is within a predetermined value range, and average power variation values symmetrical to each other, whereby an accurate determination can be made as to the individual power devices in the circuit.