Partial-Discharge Diagnosis with φ-q Pattern Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to accurately distinguish between partial discharge signals and pseudo partial-discharge signals in electric devices, leading to inaccurate insulation condition assessments and potential electrical failures.
Innovation Solution
A partial-discharge diagnostic device and system that utilizes a processor and learning model generator to generate and analyze φ-q and φ-q-n pattern diagrams, apply fast Fourier transformation, and use machine learning to differentiate between partial and pseudo partial-discharge signals, thereby improving diagnostic accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If partial discharge signals are detected to assess insulation conditions, then diagnostic capability is provided, but pseudo partial-discharge signals cause inaccurate assessments
Solution Approach 1:
The patent extracts and removes pseudo partial-discharge signals from the detected signal data through pattern recognition and filtering processes. By identifying and excluding these false signals, the system maintains accurate insulation condition assessments while preserving genuine partial discharge information for diagnostic purposes.
Solution Approach 2:
The patent introduces an intermediary processing layer that includes pattern diagram generation and machine learning-based differentiation. This intermediary layer analyzes the raw signals, identifies characteristics of pseudo signals, and separates them from authentic partial discharge signals before final diagnostic assessment.
2Measurement precision
If signal analysis is performed to distinguish partial discharge from pseudo signals, then diagnostic accuracy improves, but processing complexity increases
Solution Approach 1:
The patent segments the signal analysis process into distinct stages: raw signal acquisition, pattern diagram generation (φ-q and φ-q-n), feature extraction, and machine learning-based classification. This segmentation allows complex analysis to be broken down into manageable steps, improving accuracy while maintaining systematic processing flow.
Solution Approach 2:
The patent creates pattern diagrams (φ-q and φ-q-n) as simplified representations or copies of the complex signal data. These pattern diagrams serve as intermediary data structures that capture essential signal characteristics in a more manageable format, facilitating accurate differentiation between partial and pseudo signals without processing the entire raw signal dataset directly.
3Reliability
If comprehensive signal processing is applied to all detected signals, then false positive rate decreases, but processing time increases
Solution Approach 1:
The patent performs preliminary pattern diagram generation and initial signal filtering before final machine learning classification. By pre-processing the signals to identify and remove obvious pseudo signals early in the process, the system reduces the computational burden on subsequent analysis stages while maintaining high reliability in final assessments.
Solution Approach 2:
The patent applies comprehensive processing selectively - using thorough pattern analysis and machine learning differentiation only for signals that pass initial filtering thresholds. This partial application of intensive processing to potentially problematic signals maintains low false positive rates while reducing overall processing time compared to applying full analysis to all signals uniformly.
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
Enhances the ability to accurately determine the presence and factors of partial discharge, reducing the influence of pseudo signals and improving the reliability of insulation condition assessments in electric devices.
Implementation Method 1
a learning model generator that generates a learning model for determining, based on the data, at least either a factor of the partial discharge or presence or absence of the partial discharge
Implementation Method 2
apply fast Fourier transformation, and use machine learning to differentiate between partial and pseudo partial-discharge signals
Data Source
AI summary
According to the present embodiment, a partial-discharge diagnostic device is a device that performs determination of a factor of partial discharge in an insulator and includes a processor and a learning model generator. The processor is configured to generate data in a predetermined format with a pseudo partial-discharge signal reduced in an electric signal varying with a phase. The learning model generator is configured to generate a learning model that determines, based on the data, at least either the factor of the partial discharge or whether the partial discharge is present.