Automatic Electrical Phase Identification Using Voltage Clustering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing electrical distribution networks often lack accurate documentation of phase connections for network components, leading to inefficiencies and infrastructure stress due to undocumented alterations over time.

Innovation Solution

A method using voltage measurement data and clustering algorithms, such as k-means, to identify the electrical phase used by network devices, allowing for better load balancing and infrastructure management by segregating devices based on shared voltage changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If voltage measurement data and clustering algorithms are used to identify electrical phases, then phase identification accuracy is improved, but system complexity and computational requirements increase

Engineering Contradiction:
Improvephase identification accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs self-identification of electrical phases by automatically analyzing voltage measurement data from connected devices. The clustering algorithm autonomously groups devices by phase without requiring manual intervention or external reference equipment, enabling the network to self-diagnose and self-document its topology configuration.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual physical inspection and documentation methods with automated computational analysis. Instead of workers physically tracing connections and recording phases, the system uses voltage data analysis and clustering algorithms to automatically determine phase assignments, eliminating the need for manual mechanical processes.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If manual documentation methods are used for phase connections, then system complexity is reduced, but documentation accuracy and reliability deteriorate over time due to undocumented alterations

Engineering Contradiction:
Improvedocumentation reliabilityVSAvoiddocumentation system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The system continuously monitors voltage measurements from network devices and automatically updates phase documentation based on real-time data. This creates a feedback loop where the system constantly verifies and corrects its understanding of network topology, ensuring documentation remains accurate even as physical alterations occur in the field.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system proactively identifies and documents phase connections before problems occur by continuously analyzing voltage data patterns. Rather than waiting for issues to arise or relying on periodic manual updates, the system maintains current documentation through ongoing automated analysis of electrical characteristics.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3304687B1Automatic network device electrical phase identification
Publication Date: 2020.10.07 ITRON GLOBAL SARL
  • EP3304687B1 patent drawingFigure 1
  • EP3304687B1 patent drawingFigure 2~3
  • EP3304687B1 patent drawingFigure 4

AI summary

Techniques detect an electrical phase used by electrical network devices (e.g., a transformer, electrical meter, etc.). Voltage measurement data is obtained, such as from electrical meters. The voltage measurement data may be associated with a timestamp, and may be made at intervals over a period of time. Voltage changes may be calculated using the voltage measurement data. In an example, the voltage change is a difference determined between sequential voltage measurements. In some instances, voltage changes data is removed if it exceeds a threshold. An initial classification of network devices (e.g., randomly or by assumed electrical phase) is determined. A clustering technique (e.g., k-means) is applied, wherein the classification is updated in a manner that segregates the network devices according to actual electrical phase.