Household Power Disaggregation Using HMM and External Factors
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing power consumption monitoring systems face challenges in accurately disaggregating total household power consumption into individual device consumption due to high costs of smart plugs and limitations in providing detailed user behavior information with existing non-intrusive load monitoring (NILM) methods, particularly with high-resolution data requiring additional installations and varying collection intervals.
Innovation Solution
A power consumption disaggregation system using a Hidden Markov Model (HMM) that considers external factors, extracting characteristic variables from high-resolution data to a reference resolution, and applying these variables to disaggregate power consumption based on a DB and analysis unit, including an extraction unit for dimension-reduced data and an analysis unit for HMM-based inference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If high-resolution power data is collected to provide detailed user behavior information, then measurement precision is improved, but device complexity and installation cost increase due to requiring additional measurement devices
Solution Approach 1:
The patent extracts characteristic variables from high-resolution power data that capture essential user behavior patterns, then uses these extracted features with a low-resolution smart meter for disaggregation analysis, eliminating the need for additional high-resolution measurement devices
Solution Approach 2:
The patent creates a virtual high-resolution dataset by converting low-resolution smart meter data into characteristic variables that replicate the informational content of high-resolution data, allowing NILM analysis without physical high-resolution sensors
2Measurement precision
If smart plugs are installed to monitor individual device power consumption, then measurement precision is improved, but cost increases due to the unit price of smart plugs
Solution Approach 1:
The patent enables the existing smart meter to perform disaggregation analysis by itself through software processing of its collected power data, eliminating the need for additional smart plugs or measurement devices at individual outlets
Solution Approach 2:
The patent makes the smart meter a multi-functional device that not only measures total power consumption but also performs NILM analysis to identify individual device usage patterns, replacing the need for multiple specialized measurement devices
3Ease of operation
If low-resolution power data with 15-minute collection intervals is used to reduce device complexity, then ease of operation is improved, but loss of information increases due to inability to provide detailed user behavior information
Solution Approach 1:
The patent pre-calculates and stores characteristic variables from high-resolution data that encode user behavior patterns, then uses these pre-processed features with low-resolution data to recover detailed behavior information without requiring high-resolution input during operation
Solution Approach 2:
The patent transforms the problem from time-resolution domain to feature-space domain by extracting characteristic variables that capture behavioral patterns, allowing detailed user behavior analysis through low-resolution temporal data combined with enriched feature representations
4Measurement precision
If additional measurement devices are installed to collect high-resolution power data, then measurement precision is improved, but device complexity increases due to varying collection intervals and formats
Solution Approach 1:
The patent merges multiple data sources (smart meter power data and external factor data from weather services) into a unified analysis framework, standardizing varying collection intervals and formats through the characteristic variable extraction process
Data Source
AI summary
A power consumption disaggregation system according to an embodiment of the present invention comprises: a DB for receiving power data from a smart meter installed in a house, receiving data according to external factors affecting the power data from an external server, and storing the data in time series; an extraction unit for extracting a characteristic variable of a reference resolution time interval from the power data and the data according to external factors stored in time series; and an analysis unit for disaggregating total power consumption in the house by inferring a hidden power consumption state by inputting the characteristic variable into a power consumption disaggregation model based on a Hidden Markov Model (HMM) considering dependency of external factors.