Electric Arc Detection via Independent Filter Correlation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electric arc detection methods face challenges in accurately distinguishing between arc signals and noise generated by loads like light dimmers, leading to false detections and distortion in arc detection.
Innovation Solution
The method involves filtering an electrical signal with three distinct frequencies, determining correlations between samples from each filter, and detecting the presence of an electric arc based on these correlations, rather than relying on the sum of signals from all filters, allowing for independent analysis of each filter's output and reducing false positives.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If the sum of signals from all filters is used for correlation calculation, then the detection process is simplified, but the accuracy of electric arc detection deteriorates due to signal distortion
Solution Approach 1:
The patent divides the detection process into separate correlation calculations for each filter output rather than using a summed signal. Each filter's output is correlated independently with its previous cycle values, preserving the unique characteristics of each frequency band and preventing signal distortion that would occur with summation.
2Reliability
If noise from loads like light dimmers is present, then false detections of electric arcs increase, but filtering out all frequencies may remove actual arc signals
Solution Approach 1:
The patent applies different correlation thresholds or criteria for different filter outputs based on their specific characteristics. Each filter's correlation result is evaluated independently, allowing the system to account for local noise characteristics at each frequency band while maintaining sensitivity to arc signals that may appear in specific frequency ranges.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
An electric arc detection method comprising: - samples from a first, second and third filtering, in parallel, of a signal are determined for a current time window and for another time window; - a correlation is determined among: - a first correlation between said samples from the first filtering determined for the current window and samples from the first filtering determined for the other window; - a second correlation between said samples from the second filtering determined for the current window and samples from the second filtering determined for the other window; - a third correlation between said samples from the third filtering determined for the current window and samples from the third filtering determined for the other window; and - an electric arc is detected as a function of at least one of the determined correlations.