Partial Discharge Detection Using Internal External UHF Sensor Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting partial discharges in high-voltage equipment, such as UHF detectors, often generate nuisance alarms due to the detection of external electromagnetic waves, and current systems either fail to reliably distinguish between internal and external signals or miss discharges occurring on all three phases.
Innovation Solution
A method using comparative frequency analysis between internal and external UHF sensors, calculating the signal-to-noise ratio density, and generating an alarm when the density exceeds a threshold, while also determining the type of partial discharge by identifying the frequency with the maximum ratio and superimposing temporal signals on a reference voltage curve.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If UHF detectors are used to detect partial discharges inside high voltage equipment, then partial discharge detection capability is improved, but external electromagnetic waves are also detected causing false alarms
Solution Approach 1:
The detection system is segmented into multiple independent UHF detectors positioned at different locations (inside and outside the high voltage equipment). Each detector independently monitors its local electromagnetic environment, allowing the system to distinguish between internal partial discharge signals and external electromagnetic interference by comparing readings from multiple segments.
Solution Approach 2:
A signal processing unit acts as an intermediary that receives signals from multiple UHF detectors and performs correlation analysis. This intermediary processes the raw electromagnetic signals, compares them against stored reference patterns of partial discharges, and determines whether detected signals represent actual partial discharges or external interference, thereby reducing false alarms.
2Object-affected harmful factors
If signal subtraction method is used to reject external noises, then external noise rejection is improved, but useful signal is also removed masking partial discharge signals
Solution Approach 1:
The system implements feedback by continuously monitoring signals from multiple detectors and dynamically adjusting the correlation analysis. The signal processing unit compares real-time detector readings with stored reference patterns of partial discharges, and only generates alarms when the correlation exceeds a threshold, ensuring that useful partial discharge signals are not mistakenly removed as noise.
Solution Approach 2:
The system changes the parameter of signal comparison from simple amplitude-based subtraction to correlation-based pattern recognition. By comparing the temporal and spectral characteristics of detected signals against reference partial discharge patterns, the system can distinguish between external noise and internal partial discharges without removing useful signal information.
3Measurement precision
If coincidence analysis is used to distinguish internal from external signals, then external signal discrimination is improved, but partial discharges on all three phases are missed
Solution Approach 1:
The signal processing unit serves as an intermediary that performs correlation analysis between detector signals and stored reference patterns of partial discharges. This intermediary evaluates each detected signal independently against the reference patterns, allowing the system to identify partial discharges on any phase regardless of whether other phases are also discharging, thus maintaining detection completeness.
Solution Approach 2:
The system changes the discrimination parameter from phase-coincidence timing to signal-pattern correlation. By comparing the temporal and spectral characteristics of detected signals against reference partial discharge patterns, the system can reliably identify partial discharges on individual phases or multiple phases without requiring simultaneous detection across all phases.
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 reliably discriminates between external noise and actual partial discharges without losing information, effectively reducing false alarms and accurately identifying partial discharge events, including those occurring on all three phases.
Implementation Method 1
at least one sensor internal to said high voltage equipment detects electromagnetic signals and at least one sensor external to said high voltage equipment detects electromagnetic signals
Data Source
Figure 1
Figure 2a~2c
Figure 3
AI summary
Method of detecting partial discharge in a gas insulated electrical substation, in which: -a frequency spectrum of the signals detected by a UHF sensor internal to said substation and a frequency spectrum of the signals detected by a UHF sensor external to said substation are measured (1, 2), -an average and a variance of the frequency spectrum of the signals detected by the internal sensor are calculated (3), -said average and/or said variance are compared (4) respectively with a threshold of average and/or of variance so that, if said average and/or variance is greater than or equal to the threshold of average and/or of variance: -a density of the frequency spectrum of the signal-to-noise ratio formed (5) from the spectra of the signals detected by the internal and external sensors is calculated (6), and -an alarm signal (A) is generated if said density is greater than or equal to a density threshold.