Rotor Winding Short-Circuit Detection via Flux Probe Signal Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting short-circuit faults in rotor windings of rotating electrical machines are unreliable due to interference from load factor and network frequency variations, making it difficult to distinguish genuine faults from noise.
Innovation Solution
A method that calculates an algebraic mean value and standard deviation of the difference signal, determines a decision threshold based on these values, and identifies alert points outside a diagnostic range to adaptively detect short-circuit faults, allowing for reliable detection even with varying load factors and frequencies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the measurement duration is extended over a large number of rotor revolutions to reduce noise, then the measurement precision improves, but the reliability deteriorates due to load factor and network frequency variations causing false positives
Solution Approach 1:
The patent changes the parameter of measurement duration from extended multi-revolution sampling to a shorter duration of at least one revolution. This parameter change resolves the contradiction by achieving sufficient signal-to-noise ratio through a different approach (comparing multiple probes simultaneously) rather than extending time, thereby avoiding load factor and frequency variations that cause false positives.
Solution Approach 2:
The patent uses multiple probes (at least two probes) to simultaneously measure magnetic flux at different locations in the air gap. This partial action approach achieves noise reduction through spatial diversity rather than temporal averaging, allowing reliable detection within a shorter measurement window and avoiding the reliability issues of extended measurements.
2Reliability
If the measurement is performed over at least one revolution of the rotor, then the reliability of fault detection improves, but the productivity decreases due to longer detection time
Solution Approach 1:
The patent employs multiple probes (at least two probes positioned at different radial or tangential locations) to simultaneously capture magnetic flux variations. This spatial redundancy provides sufficient data for reliable fault detection within a single rotor revolution, eliminating the need for extended multi-revolution measurements and thereby improving detection speed without sacrificing reliability.
Solution Approach 2:
The patent combines the measurements from multiple probes to detect rotor winding faults. By merging the signals from at least two probes, the system achieves reliable fault detection within a shorter time frame (one revolution), thus improving productivity while maintaining the required reliability through combined spatial information.
3Reliability
If multiple probes are used to improve detection reliability, then the reliability improves, but the device complexity increases
Solution Approach 1:
The patent uses a minimal sufficient number of probes (at least two probes) rather than an excessive number. This partial action approach achieves the necessary reliability improvement through spatial diversity while avoiding the device complexity and cost associated with using many more probes. The minimum number of probes provides just enough spatial information to reliably distinguish faults from noise.
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 enhances the reliability of short-circuit fault detection by automatically adjusting the decision threshold to the maximum amplitude of the probe signal, reducing interference from load factor and network frequency variations, and enables detection within a shorter sampling period, such as one and a quarter revolutions of the rotor.
Implementation Method 1
When the machine is in operation, the variations of the magnetic field in the air gap generate an electromotive force at the terminals of each of the two coils.
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
The detection method uses flux measurements made by a probe disposed in a gap of the machine, and carries out the sampling of a probe signal, and the generating of at least one difference signal (G3) from the measurements of the probe signal in order to perform a pole-to-pole comparison of the signal. The processing of the difference signal performs the calculation of an algebraic average value (m) and a standard deviation of the difference signal, the determination of a maximum amplitude of the probe signal, the calculation of a decision threshold value (S) taking said standard deviation and said maximum amplitude into account, the determination of a diagnostic range (PM) centred on said algebraic average value (m) and having a width that is twice the value of the decision threshold value (S), and the identification of a possible alarm point corresponding to a value of the difference signal (G3) that is outside the diagnostic range (PM), in order to indicate a short-circuit fault.