Transformer Winding Deformation Diagnosis via SVM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for diagnosing winding deformation in power transformers are off-line, requiring power outages, are labor-intensive, and have long test intervals, making it difficult to detect faults in a timely manner and increasing the risk of accidents.
Innovation Solution
An intelligent on-line diagnosis and positioning method using voltage and current monitoring data, which involves dividing monitoring indicators into subsamples, calculating entropy features, and employing an SVM model for classification to determine deformation and its location without disrupting grid operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If off-line diagnosis methods are used, then measurement precision can be achieved, but productivity deteriorates due to power outages and long test intervals
Solution Approach 1:
The patent replaces mechanical off-line testing systems with an on-line electrical monitoring system using SVM algorithms. The system continuously monitors voltage and current signals, extracts features using signal processing, and diagnoses winding deformation in real-time without requiring power outages or manual testing, thus resolving the contradiction between measurement precision and productivity
Solution Approach 2:
The patent implements continuous on-line monitoring of transformer windings by continuously acquiring voltage and current signals, processing them through feature extraction algorithms, and providing real-time diagnosis. This eliminates the discontinuous nature of off-line testing and enables uninterrupted diagnostic surveillance, improving both productivity and timely fault detection
2Reliability
If off-line diagnosis methods are used, then comprehensive testing can be performed, but loss of time increases due to long test intervals
Solution Approach 1:
The patent performs preliminary continuous monitoring and analysis of transformer winding conditions before actual faults occur. By continuously acquiring and analyzing voltage and current signals with SVM-based algorithms, the system detects early signs of winding deformation and issues warnings before catastrophic failures happen, eliminating the time delay inherent in periodic off-line testing
Solution Approach 2:
The system provides continuous real-time monitoring instead of periodic off-line testing, enabling immediate detection of winding deformation as it develops. This continuous surveillance eliminates the time intervals between tests and ensures faults are detected at their earliest stages, improving both reliability and reducing time loss
3Measurement precision
If off-line diagnosis methods are used, then thorough testing can be conducted, but device complexity increases due to intensive test works
Solution Approach 1:
The patent creates a universal on-line monitoring system that can diagnose all types of winding deformation (axial displacement, radial deformation, bending) using a single SVM-based platform. The system processes voltage and current signals through unified feature extraction algorithms, eliminating the need for multiple specialized off-line testing devices and procedures, thus reducing device complexity while maintaining comprehensive diagnostic capability
Solution Approach 2:
The patent replaces complex mechanical off-line testing equipment with an electrical on-line monitoring system using signal processing and SVM algorithms. The system continuously analyzes voltage and current waveforms to detect winding deformation, eliminating the need for physical disassembly, special test equipment, and manual operations, thereby simplifying the overall diagnostic system while maintaining thorough testing capability
Data Source
AI summary
Disclosed is an intelligent on-line diagnosis method for winding deformation of power transformer. When a transformer is subjected to short-circuit shock or transportation collision, transformer windings may undergo local twisting, swelling or the like under the action of an electric power or mechanical force, which is called winding deformation and will cause a huge hidden danger to the safe operation of the power network. Commonly used diagnosis methods for winding deformation are all off-line diagnosis methods, which have the disadvantages that transformers need to be shut down and highly skilled operators are required. The present invention provide an intelligent on-line diagnosis method for winding deformation on the basis of combination of information entropy and support vector machine. By carrying out feature extraction of current and voltage signals based on permutation entropy and wavelet entropy, integrating the variation of the monitoring indicators of the power transformers in complexity, time-frequency domain and the like and automatically learning the diagnostic logic from fault features through the machine learning algorithm, intelligent diagnosis of winding deformation is realized, thereby reducing labor costs and improving diagnosis efficiency.