Partial Discharge Detection in Power Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods using acoustic emission sensors fail to detect partial discharge during lighting impulse withstand voltage tests due to electromagnetic noise and mechanical oscillations, making it impossible to determine if partial discharge occurs and the margin before dielectric breakdown in power devices.
Innovation Solution
A method employing an acoustic emission sensor to apply an impulse voltage, remove electromagnetic noise and mechanical oscillation components using a low-pass filter and mechanical oscillation removal unit, and determine partial discharge by correlating feature amounts of the acoustic signal with pre-acquired data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an acoustic emission sensor is used to detect partial discharge during lighting impulse withstand voltage tests, then the ability to detect partial discharge is improved, but electromagnetic noise and mechanical oscillations interfere with the detection accuracy
Solution Approach 1:
The patent extracts and removes the mechanical oscillation component from the acoustic signal by comparing it with a reference signal obtained under identical test conditions but without partial discharge. This isolation of the harmful mechanical oscillation component allows for its elimination, thereby improving the detection accuracy of partial discharge signals during lighting impulse withstand voltage tests
Solution Approach 2:
The patent introduces a reference signal as an intermediary element. This reference signal, obtained from a test without partial discharge but with identical mechanical oscillations, serves as a mediator to identify and remove the mechanical oscillation component from the test signal containing partial discharge, enabling accurate detection despite the presence of electromagnetic noise and mechanical oscillations
2Measurement precision
If signal processing techniques are applied to remove noise and mechanical oscillations, then the detection accuracy is improved, but the device complexity increases
Solution Approach 1:
The patent performs preliminary action by acquiring a reference signal under identical test conditions but without partial discharge before conducting the actual test. This pre-acquired reference signal contains the mechanical oscillation component that can be subsequently removed from the test signal, simplifying the signal processing required during the actual partial discharge detection
Solution Approach 2:
The patent applies the inversion principle by subtracting the reference signal (obtained without partial discharge) from the test signal (obtained with potential partial discharge). This reverse approach of removing what is present in the reference signal from the test signal effectively isolates the partial discharge component, improving detection accuracy without requiring complex signal processing systems
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
Enables accurate detection of partial discharge and determination of the margin before dielectric breakdown in power devices during lighting impulse withstand voltage tests, effectively distinguishing partial discharge signals from noise.
Implementation Method 1
a method using an acoustic emission (AE) sensor is known
Implementation Method 2
removing, through a low-pass filter, an electromagnetic noise superimposed on the acoustic signal
Data Source
AI summary
A partial discharge determination method and apparatus, and a partial discharge determination system capable of indicating how much margin is left before dielectric breakdown, in a lighting impulse withstand voltage test, and a power device for which whether partial discharge is caused is determined by them, and a method for manufacturing a power device including the partial discharge determination method. A low-pass filter receives an acoustic signal resulting from application of an impulse voltage and acquired by an acoustic emission sensor, and removes an electromagnetic noise superimposed on the acoustic signal. A mechanical oscillation removal unit removes a mechanical oscillation component of a device under test, from the acoustic signal resulting from application of the impulse voltage which is a high voltage, based on the mechanical oscillation component acquired in advance and included in the acoustic signal resulting from application of the impulse voltage which is a low voltage.