Method for generating a classifier of a domestic hot water system
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-controlled electric domestic hot water systems contribute to peak electrical energy consumption, leading to network congestion and the need for oversized equipment, as they do not respond to off-peak control signals effectively, making it difficult to identify and maintain these systems.
Innovation Solution
A method for generating a classifier that identifies non-controlled domestic hot water systems by analyzing time series of electrical power consumption data using an encoding function, classification function, and decision trees, allowing for targeted maintenance and enslavement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If electric domestic hot water systems are activated during off-peak hours using control signals, then electricity distribution network peak demand is reduced, but uncontrolled systems fail to respond and continue to generate peak consumption
Solution Approach 1:
The patent implements a feedback mechanism by continuously monitoring the electrical power consumption patterns of domestic hot water systems and comparing them against expected patterns for controlled systems. The system provides feedback to identify which systems are not responding to control signals, enabling targeted maintenance or reconfiguration to restore control effectiveness.
Solution Approach 2:
The system enables self-service by automatically detecting and classifying uncontrolled domestic hot water systems without requiring manual intervention. The classification algorithm autonomously identifies systems that are not responding to control signals, allowing the distribution network operator to automatically generate maintenance alerts or reconfiguration instructions.
2Reliability
If targeted maintenance operations are performed on identified uncontrolled systems, then control reliability is improved, but identification of these systems is currently difficult
Solution Approach 1:
The patent replaces manual detection methods with an automated electronic classification system. Instead of physically inspecting each system or requiring manual reporting, the system uses electrical consumption data processed through classification algorithms to automatically identify uncontrolled systems, significantly reducing the difficulty of detection.
Solution Approach 2:
The patent introduces an intermediary classification system that acts as a mediator between the control signals sent by the distribution network operator and the domestic hot water systems. This intermediary analyzes consumption patterns to infer the control status of systems, providing indirect but reliable identification of uncontrolled systems without requiring direct access to the systems themselves.
3Loss of time
If decision trees are used for classification, then computation time is reduced, but classification accuracy must be maintained
Solution Approach 1:
The patent segments the classification process into distinct stages: data preprocessing, feature extraction, and classification using decision trees. This segmentation allows each stage to be optimized independently, with the decision tree classifier receiving pre-processed data that reduces computation time while maintaining the necessary information for accurate classification.
Solution Approach 2:
The patent performs preliminary actions by pre-processing the electrical consumption data and extracting relevant features before applying the decision tree classification. This preliminary processing organizes the data in a way that accelerates the classification computation while ensuring that all necessary information for accurate classification is preserved and enhanced.
Data Source
Figure 1
Figure 2A~2C
Figure 3
AI summary
The invention relates to a method for generating a classifier for classifying a domestic hot water system connected to a delivery point of an electrical distribution network, the classifier (200) being configured to: - implement an encoding function (102) associating with a time series of electrical power consumption values (X) measured at the delivery point, a compressed representation (Z) of the time series of electrical power values, and - assign a class to the domestic hot water system according to the compressed representation of the time series of electrical power consumption values using a decision tree or a combination of decision trees (109).