Electric Device Model Identification via Machine Learning Waveform Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to accurately identify electric device models and manufacturers based on current waveform data, as there are limited differences in instantaneous current waveforms between devices, making it difficult to distinguish between similar appliances.
Innovation Solution
A model identification system that acquires device information, extracts specific operation data, and uses machine learning to extract parameters for model identification, comparing these parameters to identify the electric device's model and manufacturer by analyzing time series data of current waveforms and electric power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional current waveform analysis is used to identify electrical devices, then the identification process is simple, but the identification accuracy is insufficient because there is no large difference in instantaneous current waveform between manufacturers
Solution Approach 1:
The patent segments the identification process into multiple stages: device information acquisition, operation section extraction, feature quantity extraction through machine learning, and model identification. This segmentation allows complex analysis to be performed systematically on specific operation sections rather than attempting to analyze the entire current waveform at once, thereby improving identification accuracy while managing system complexity.
Solution Approach 2:
The patent transitions from analyzing instantaneous current waveform characteristics (one-dimensional time domain) to extracting feature quantities through machine learning that capture subtle operational patterns. This dimensional transformation enables the system to identify devices based on nuanced operational differences rather than relying solely on gross waveform characteristics, significantly improving identification accuracy.
2Measurement precision
If machine learning process with multiple samplings is implemented, then the model identification accuracy is improved, but the processing time and computational load increase
Solution Approach 1:
The patent performs preliminary extraction of operation sections that are most informative for identification before applying machine learning. By pre-identifying and isolating critical operation sections, the system reduces the amount of data requiring intensive machine learning processing, thereby maintaining high identification accuracy while reducing overall processing time.
Solution Approach 2:
The patent applies machine learning with multiple samplings selectively to extracted operation sections rather than processing entire current waveforms. This partial application of computationally intensive methods to only the most relevant data segments achieves high identification accuracy while minimizing unnecessary computational overhead and processing time.
Data Source
AI summary
A model identification system includes a device information acquiring unit that acquires device information used to identify a model of an electric device, an operation extracting unit that extracts data of a predetermined operation section, a feature quantity extracting unit that extracts a parameter used to identify the electric device, and a model identifying unit that identifies a model of an electric device, wherein the feature quantity extracting unit performs a machine learning process by sampling the data of the predetermined operation section extracted from the operation extracting unit a plurality of times, extracts a parameter corresponding to each sampling, and extracts a parameter appropriate to identify a model among a plurality of sampled parameters.