Utility Signal Processing for Simultaneous Device State Disaggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing utility consumption monitoring systems struggle to accurately disaggregate power consumption among individual devices in a building, especially when multiple devices switch states simultaneously, leading to inaccurate event classification and failed disaggregation.
Innovation Solution
A method and system that process utility consumption signals by extracting features from signal portions corresponding to device state changes, using plateaus to identify multiple device transitions, and employing a probabilistic model to refine device state probabilities, allowing for accurate classification of near-simultaneous events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional utility consumption monitoring systems are used to measure aggregate utility consumption signals, then the total utility consumption data for the building can be obtained, but the system fails to accurately determine utility consumption for individual devices when multiple devices switch states simultaneously
Solution Approach 1:
The patent segments the aggregate utility consumption signal into individual device-level signals by detecting state changes in the aggregate signal and identifying corresponding device events. The system divides the complex aggregate signal processing into manageable components: detecting magnitude changes, extracting signal portions, identifying device state changes, and calculating device-level consumption separately for each detected event.
Solution Approach 2:
The system performs preliminary detection of state changes in the aggregate utility consumption signal before attempting to disaggregate individual device consumption. By first detecting magnitude changes and extracting relevant signal portions, the system prepares the data structure needed for accurate device-level measurement, handling multiple simultaneous device transitions in advance.
2Measurement precision
If the system attempts to identify multiple device transitions using plateaus in the signal, then near-simultaneous events can be separated, but the processing complexity increases
Solution Approach 1:
The patent applies partial action by focusing detection efforts only on signal portions where magnitude changes occur, rather than analyzing the entire aggregate signal continuously. The system extracts and analyzes only the relevant signal portions corresponding to device state changes, reducing unnecessary processing while maintaining accurate event classification for near-simultaneous events.
3Measurement precision
If probabilistic models are used to refine device state probabilities, then event classification accuracy improves, but computational requirements increase
Solution Approach 1:
The probabilistic model is applied selectively only to signal portions where device state changes are detected, rather than continuously processing the entire aggregate signal. This partial application of the probabilistic model refines device state probabilities for relevant events while minimizing unnecessary computational energy consumption during periods without device transitions.
Data Source
Figure 1(a)
Figure 1(b)
Figure 1(c)
AI summary
A method of processing a utility consumption signal, comprising: monitoring an input utility consumption signal for a change in magnitude; if a change is detected: extracting a first portion of the input signal including the change; extracting a first set of features corresponding to the first portion; determining an event probability value for each of one or more utility consuming device state changes, the or each event probability value being determined from the first set of features and being the probability that the first portion of the signal corresponds to the change in state of the device; if none of the event probability values are above a first threshold value; searching for two or more plateaus in the first portion, wherein each pair of adjacent plateaus define a further change in magnitude of the signal between the plateaus; if two or more plateaus are detected: extracting a sub-portion of the signal corresponding to each of the further changes and each combination of the further changes; extracting a second set of features corresponding to each sub-portion; determining an event probability value for each of one or more utility consuming device state changes, the or each event probability value being determined from the second set of features and being the probability that the sub-portion of the signal corresponds to the change in state of the device.