Partial Discharge Monitoring Using Density-Based Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing partial discharge monitoring systems struggle to accurately determine defects in high-voltage power devices in real time, especially when the voltage phase of the power device is unknown, leading to inaccurate determination results.
Innovation Solution
A partial discharge monitoring system and method that utilize a machine learning algorithm to recognize patterns in partial discharge signals, generating PRPD data, and clustering feature dot data based on density and distance to improve determination accuracy without requiring information on the voltage phase of the high-voltage power device.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If PRPD analysis is performed using current phase or AC voltage phase instead of actual power device voltage phase, then the analysis can proceed without accurate voltage phase measurement, but the determination accuracy deteriorates due to phase difference
Solution Approach 1:
The patent introduces an intermediary approach by using current phase or AC voltage phase as a substitute mediator for the actual power device voltage phase. This allows the PRPD analysis to proceed when accurate voltage phase measurement is unavailable, while the system attempts to compensate for the phase difference to maintain determination accuracy.
Solution Approach 2:
The patent applies parameter changes by modifying the phase reference parameter from the actual voltage phase to the current phase or AC voltage phase. This parameter substitution enables the system to operate under conditions where accurate voltage phase measurement is not possible, while adjusting the analysis parameters to minimize the impact of phase difference on determination accuracy.
2Device complexity
If manual operator measurement is used for partial discharge detection, then the system can be simple, but real-time monitoring capability and safety in dangerous areas are limited
Solution Approach 1:
The patent implements self-service by enabling the system to automatically perform partial discharge detection and monitoring without requiring manual operator intervention. The automated system can safely operate in dangerous areas and provide real-time monitoring, while the system uses its own resources (sensors, processors) to complete the measurement and analysis tasks independently.
Solution Approach 2:
The patent replaces the mechanical/manual measurement system with an automated electronic system. Instead of operators physically measuring partial discharge in real-time, the system uses electronic sensors and processing units to automatically detect, measure, and analyze partial discharge signals, thereby eliminating the need for manual intervention and enabling continuous real-time monitoring.
3Device complexity
If feature dot data clustering is performed only in high-density areas, then the clustering process is simple, but signals in low-density areas may be missed
Solution Approach 1:
The patent applies segmentation by dividing the feature dot data clustering process into multiple stages or regions. The system first identifies high-density areas for initial clustering, then separately processes low-density areas to ensure no signals are missed. This segmentation allows the system to handle different data density regions with appropriate methods while maintaining overall reliability.
Solution Approach 2:
The patent implements partial or excessive action by performing clustering not only in high-density areas but also extending to low-density areas where signals may be sparse. This excessive action ensures that all potential signals are captured, including those in low-density regions that might otherwise be overlooked, thereby improving detection completeness at the cost of increased processing scope.
Data Source
AI summary
The present disclosure relates to a partial discharge monitoring system and a partial discharge monitoring method that are capable of monitoring and determining a defect generated in a high-voltage power device in real time by classifying signals generated from the high-voltage power device with a machine learning algorithm being applied, and capable of easily forming feature dot data clusters in both a high-density area and a low-density area of two-dimensional feature dot data by performing a process of clustering feature dot points generated from the signals of the power device on the basis of density and distance in parallel, respectively, thereby generating PRPD data for each cluster without missing a signal to improve partial discharge determination accuracy.