Synchronous Machine Health Monitoring Using Negative Sequence EMF

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional synchronous machine health monitoring methods fail to accurately detect inter-turn faults due to assumptions of sinusoidal winding distribution, leading to erroneous fault detection and potential unplanned outages, especially in critical applications like aviation engines.

Innovation Solution

A system and method for monitoring synchronous machine health by estimating the negative sequence back electromotive force (EMF) using sequence components of phase voltages and currents, which includes a data store for lookup values and a prognostic module to compute a machine health indicator, raising alarms for deviations from healthy states.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If conventional synchronous machine health monitoring methods assume sinusoidal winding distribution, then the monitoring system is simple to implement, but fault detection accuracy deteriorates under inter-turn fault conditions

Engineering Contradiction:
Improvemonitoring system complexityVSAvoidfault detection accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent changes the monitoring approach from assuming sinusoidal distribution to calculating actual winding distribution parameters using sequence components. By computing negative sequence voltage and current components and deriving the actual winding distribution function, the system adapts to fault conditions without requiring complex hardware modifications, thus improving detection accuracy while maintaining reasonable system complexity

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent replaces the mechanical assumption of sinusoidal winding distribution with an electrical calculation approach using sequence components. Instead of physically measuring or mechanically verifying winding distribution, the system uses electrical measurements (voltages and currents) and mathematical transformations to determine actual winding distribution, simplifying the physical system while improving accuracy

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

2Ease of manufacture

If synchronous machine models assume sinusoidal winding distribution, then model creation is straightforward, but the model becomes inaccurate under fault conditions leading to erroneous fault detection

Engineering Contradiction:
Improvemodel creation easeVSAvoidfault detection reliability
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The patent makes the model dynamic by allowing it to adapt to fault conditions. Instead of using a fixed sinusoidal assumption, the model dynamically calculates the actual winding distribution based on measured sequence components. This dynamic adjustment enables the model to remain accurate whether the machine is healthy or experiencing an inter-turn fault, improving reliability without significantly complicating model creation

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The monitoring system performs self-verification by using its own measurements to detect faults. The system calculates negative sequence components from its measurements and uses these to determine actual winding distribution, enabling the model to self-adjust and self-verify its accuracy without external intervention, thereby improving reliability while maintaining ease of implementation

Inventive Principle:
Principle #25Self-service

3Ease of operation

If conventional monitoring methods are used, then the system operates with simple assumptions, but inter-turn faults remain undetected until they cause unplanned outages

Engineering Contradiction:
Improvemonitoring operation simplicityVSAvoidcontinuous operation reliability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The patent implements preliminary fault detection by continuously monitoring negative sequence components and comparing actual winding distribution against healthy baseline values. This preliminary detection occurs before faults escalate to cause unplanned outages, allowing maintenance to be scheduled proactively. The system performs this monitoring with straightforward operations using standard electrical measurements and calculations

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach enables early detection of inter-turn faults in synchronous machines, reducing the risk of unplanned outages and allowing for timely maintenance by accurately assessing the health of the armature winding, even under asymmetrical fault conditions.

Implementation Method 1

estimating the negative sequence back electromotive force (EMF) using sequence components of phase voltages and currents

Methodology Applied
Scientific EffectElectromagnetic Induction: Electromagnetic Induction

Data Source

PatentEP2469703B1System and method for synchronous machine health monitoring
Publication Date: 2020.09.23 GENERAL ELECTRIC CO
  • EP2469703B1 patent drawingFigure 1
  • EP2469703B1 patent drawingFigure 2A~2B
  • EP2469703B1 patent drawingFigure 3

AI summary

A method, system and computer program product for monitoring health of a synchronous machine is provided. The method includes receiving (402) a plurality of phase voltage values and a plurality of phase current values. The method then computes (404) a negative sequence voltage (Vn) based on the plurality of phase voltage values. The method also computes (406) one or more operating parameters based on at least one of the plurality of phase voltage values and the plurality of phase current values. The method retrieves (408) from a data store, one or more known Vn based on the one or more operating parameters. The method then computes (410) a machine health indicator based on the computed Vn and the one or more known Vn, and raises (412) an alarm based on the machine health indicator.