Control Rod Drive Motor Current Analysis for Abnormality Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting abnormalities in control rod drive devices of nuclear power facilities are inadequate, particularly when the ratio between D-axis and Q-axis currents in electric motors is imbalanced, or when mechanical resonance is unlikely to occur, leading to undetectable abnormalities due to reduced current fluctuations.
Innovation Solution
An abnormality detector that utilizes both electrical and mechanical parameters to diagnose abnormalities by extracting specific section data from the phase current, calculating natural frequencies, and applying machine learning techniques to detect pulse-like load fluctuations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If conventional current-based diagnosis methods are used, then abnormality detection is simple and easy to implement, but abnormality cannot be detected when D-axis current is much larger than Q-axis current
Solution Approach 1:
The invention changes the detection parameter from simple current magnitude to current frequency characteristics. By analyzing the frequency spectrum of the current and detecting resonance frequency shifts, the system can identify abnormalities even when the Q-axis current is small relative to the D-axis current, thus maintaining ease of implementation while improving detection reliability.
Solution Approach 2:
The invention utilizes mechanical vibration principles by detecting resonance frequencies of the drive device. When abnormalities occur, the resonance frequency characteristics change, which can be detected through current analysis. This approach enables reliable abnormality detection independent of the D-axis/Q-axis current ratio.
2Reliability
If diagnosis based on natural vibration frequency amplitude increase is used, then abnormality can be detected through resonance, but abnormality cannot be detected in devices with non-resonant structures
Solution Approach 1:
The invention creates a universal diagnosis method that works for both resonant and non-resonant structures. By analyzing frequency spectrum characteristics and detecting changes in resonance behavior (or lack thereof), the system can identify abnormalities across different device structures, making the diagnosis method universally applicable.
Solution Approach 2:
The invention shifts from relying on amplitude increase at resonance to analyzing frequency spectrum characteristics and resonance frequency shifts. This parameter change enables the method to detect abnormalities in non-resonant structures by identifying deviations from expected frequency characteristics, thus improving versatility.
3Device complexity
If effective value or average value of phase current is calculated, then simple processing is achieved, but small current changes are buried in measurement noise
Solution Approach 1:
The invention applies frequency analysis techniques to the current signal, treating it similarly to vibration analysis. By transforming the time-domain current signal into the frequency domain and examining spectral characteristics, the system can detect small periodic variations that would be obscured in simple average or effective value calculations, thereby improving measurement precision without excessive complexity.
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
An abnormality detector comprises a specific section extractor, a feature amount calculator, and an abnormality diagnoser. The specific section extractor divides, into specific section data, phase current caused to flow through an electric motor used in a control rod drive device, and extracts the specific section data. The feature amount calculator calculates a feature amount for use in diagnosis in the abnormality diagnoser. The abnormality diagnoser diagnoses a presence or absence of abnormality of the control rod drive device based on the feature amount. The feature amount calculator calculates a natural frequency of the entire control rod drive device based on an electrical parameter of the electric motor and a mechanical parameter of the control rod drive device, and calculates the feature amount by using the natural frequency and an applied voltage frequency to the electric motor.