Arc Detection via Dual Wavelet Current Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting serial arcs in electrical systems, particularly in high-voltage on-board networks of motor vehicles, face challenges due to the difficulty in distinguishing the voltage drop caused by arcs from normal operational changes, requiring complex hardware and multiple measurements, and often result in low detection rates.
Innovation Solution
A method utilizing two wavelet transformations with different mother wavelets applied to chronological sequences of current measurements to determine the presence of an arc, allowing for accurate detection with reduced hardware requirements and improved detection rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If voltage measurements are performed at different points to detect serial arcs, then arc detection capability is improved, but device complexity and hardware requirements increase
Solution Approach 1:
The patent extracts the arc detection function from complex multi-point voltage measurement systems and implements it through single-point current measurement combined with wavelet transformation. The wavelet transformation extracts characteristic frequency components from the current signal that indicate arc presence, eliminating the need for multiple voltage sensors and complex differential measurements.
Solution Approach 2:
The patent replaces the electrical measurement approach (voltage measurements at multiple points) with a signal processing approach (wavelet transformation of current measurements). This substitution transforms the problem from electrical domain measurement to mathematical signal analysis, reducing hardware complexity while maintaining detection reliability.
2Speed
If conventional frequency analysis is used to detect arcs, then detection speed is improved, but detection precision decreases
Solution Approach 1:
The patent changes the analysis parameter from conventional frequency domain analysis to wavelet domain analysis. Wavelet transformation provides time-frequency localization that allows precise identification of arc events by analyzing both temporal and spectral characteristics simultaneously, improving detection precision while maintaining high speed through efficient algorithms.
Solution Approach 2:
The patent transitions from one-dimensional frequency analysis to two-dimensional time-frequency analysis using wavelet transformation. This additional temporal dimension allows precise localization of arc events in time while maintaining frequency resolution, thereby improving detection precision without sacrificing speed.
3Ease of operation
If current-based detection is used for serial arcs, then ease of operation is improved, but detection precision decreases due to normal operational variations
Solution Approach 1:
The patent introduces wavelet transformation as an intermediary processing step between current measurement and arc detection. This intermediary extracts characteristic frequency components that are specific to arc events, filtering out normal operational variations. The wavelet coefficients serve as an intermediate representation that enhances arc-related signals while suppressing background noise from load changes.
Data Source
AI summary
Methods and devices for detecting electrical arcs are provided. A first wavelet transformation with a first mother wavelet is applied to a chronological sequence of current measurements (80) of a current through a lead, to obtain first wavelet coefficients. In addition, a second wavelet transformation with a second mother wavelet different from the first mother wavelet is applied to the chronological sequence in order to obtain second wavelet coefficients. On the basis of the first wavelet coefficients and the second wavelet coefficients, it is then determined whether an arc (10; 11; 24) is present.