Nonintrusive Load Monitoring Classification for Unknown Operation Modes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing nonintrusive load monitoring (NILM) techniques struggle to accurately classify current waveforms for the same electric instrument when the operation mode is unknown, as they require pre-known operation modes to generate necessary conditions for estimation.
Innovation Solution
A classification device with a processing unit that includes a waveform information classification unit, an on/off information classification unit, and a set information classification unit, which classify set information based on similarity degrees of waveform and on/off information, allowing for accurate classification of current waveforms without requiring information about the operation mode.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If the NILM technique separates current waveform by characteristic of current waveform, then the working status of each electric instrument can be monitored without attaching current measurement device, but the same electric instrument may be separated into different current waveforms when having different operation modes
Solution Approach 1:
The patent segments the classification process into multiple stages: first separating current waveforms by NILM technique, then performing secondary classification by matching waveform characteristics and timing information. This multi-stage segmentation allows the system to maintain the installation simplicity of NILM while improving classification accuracy through progressive refinement.
Solution Approach 2:
The patent changes the classification parameters from simple waveform characteristics to a combination of waveform characteristics, timing information, and operational patterns. By analyzing multiple parameters including start/stop timings and operational sequences, the system can accurately identify the same electric instrument across different operation modes without requiring pre-registration of all possible modes.
2Measurement precision
If the method uses pre-derived necessary conditions and clue conditions based on known operation modes, then the same electric instrument can be estimated accurately, but the method cannot estimate when operation mode information is not known in advance
Solution Approach 1:
The patent enables the system to automatically learn and adapt to electric instrument operation patterns through self-service mechanisms. By analyzing timing information and operational sequences from multiple current waveforms, the system automatically builds classification models without requiring external input of operation mode definitions, allowing it to handle both known and unknown operation modes effectively.
Solution Approach 2:
The patent performs preliminary analysis of timing information and operational patterns from multiple current waveforms to establish classification criteria before final identification. By pre-processing the data to extract temporal characteristics and operational sequences, the system prepares the necessary information structures that enable accurate classification even when operation modes were not previously known.
3Quantity of substance
If multiple current waveforms are separated for the same electric instrument, then comprehensive monitoring data is obtained, but the complexity of determining which combination represents the same electric instrument increases
Solution Approach 1:
The patent implements feedback mechanisms where classification results from preliminary analysis are used to guide the final identification process. By using timing information and operational patterns as feedback criteria, the system can efficiently evaluate multiple waveform combinations and identify the correct groupings without exhaustive analysis, reducing computational complexity while maintaining comprehensive monitoring coverage.
Solution Approach 2:
The patent adds temporal and operational dimensions to the classification process by analyzing timing information and operational sequences. This dimensional expansion transforms the problem from simple waveform matching to multi-dimensional pattern recognition, enabling the system to distinguish the same electric instrument across different operation modes while reducing the search space for valid waveform combinations.
Data Source
AI summary
In order to classify a current waveform of current estimated to be supplied to the same electric instrument, even when an operation mode of an operating electric instrument is unknown, a classification computer includes: a first classification unit to perform first classification of each piece of set information by information being included in each piece of the set information being a combination of waveform information and on/off information, and representing a similarity degree of the waveform information; a second classification unit to perform second classification of each piece of the set information by information being included in each piece of the set information and representing a similarity degree of the on/off information; and a third classification unit to classify the set information by a classification result related to the first classification and the second classification.