Device Power Estimation Using Operational State Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional technologies face challenges in accurately estimating the electric power consumption of individual devices when multiple devices have similar intrinsic waveforms, such as in factories with many identical devices.
Innovation Solution
An electric power consumption estimation apparatus that acquires total electric power consumption data and operating state data from devices, calculates an average electric power consumption, and uses this information to estimate device electric power consumption through an electric power consumption matrix calculated from a specific equation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional identification methods (NILM, HMM, FHMM) are used to identify home appliances, then devices with different intrinsic waveforms can be identified, but devices with similar intrinsic waveforms cannot be distinguished
Solution Approach 1:
The patent segments the identification approach by separating waveform analysis into two components: intrinsic waveform characteristics (for device type identification) and operational state characteristics (for individual device identification). This segmentation allows the system to handle both different and similar waveform devices effectively.
Solution Approach 2:
The patent adds a new dimension to the identification problem by incorporating operational state information (on/off states, operational modes) as a separate identifier alongside waveform characteristics. This dimensional expansion enables distinction between multiple devices with similar waveforms by tracking their unique operational patterns.
2Measurement precision
If matrix factorization is used to estimate device electric power consumption, then accurate estimation for individual devices can be achieved, but the system requires acquisition of operating state data from each device
Solution Approach 1:
The patent introduces an intermediary computational layer (matrix factorization process) that connects total power consumption measurements with individual device consumption estimates. This intermediary processing enables accurate estimation without requiring direct measurement from each device, reducing hardware complexity.
Solution Approach 2:
The patent replaces direct mechanical/electrical measurement systems for each device with a computational model (matrix factorization) that infers individual device consumption from aggregate data and operational states. This substitution reduces physical measurement complexity while maintaining accuracy.
Data Source
AI summary
An electric power consumption estimation apparatus includes: a total electric power consumption data acquisition unit that acquires total electric power consumption data indicative of a total electric power consumption from an electric power meter; an operating state data acquisition unit that acquires operating state data indicative of whether or not each of the plurality of devices is in an operating state, from each of the devices; an average electric power consumption analysis unit that estimates an average electric power consumption of each of the plurality of devices from the total electric power consumption and whether or not each of the devices is in the operating state; and an electric power consumption calculation unit that calculates a device electric power consumption, which is the electric power consumption of each of the plurality of devices for each predetermined period by using the average electric power consumption.