Arc Detection in DC Installations via Fourth-Order Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detecting electrical arcs in photovoltaic installations with DC current sources is challenging due to low arc current and the similarity in current spectra with and without arcs, making it difficult to apply existing detection methods effectively.
Innovation Solution
A method and device using a calculation processor to obtain digitized temporal signals, apply bandpass filtering, calculate statistical values of order four, and compare them with thresholds to detect electrical arcs, even when the arc current is low, involving the extraction of temporal sub-signals and analysis of fourth-order statistics across multiple time windows.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional arc detection methods are applied, then detection capability is sufficient for high current arcs, but detection effectiveness deteriorates when arc current is low
Solution Approach 1:
The patent transforms the detection parameter from direct current magnitude to statistical characteristics (kurtosis, skewness) of the current signal. By analyzing the fourth-order statistical moments of the current waveform, the system can distinguish arc signatures even when arc current is low and spectrally similar to normal operation. This parameter transformation enables reliable detection across the full range of arc currents.
2Power
If the connection cable length is increased, then power transmission capability is improved, but arc detection difficulty increases
Solution Approach 1:
The patent introduces statistical analysis (kurtosis and skewness calculations) as an intermediary processing step between current measurement and arc detection. This intermediary transformation extracts distinctive features from the current signal that are independent of cable length, enabling reliable arc detection even in long cable installations where arc current is attenuated and spectrally masked.
3Productivity
If computational complexity is reduced, then processing speed is improved, but detection accuracy may deteriorate
Solution Approach 1:
The patent extracts only the essential statistical features (kurtosis and skewness) from the current signal, discarding redundant information. By focusing computation on these two higher-order statistical moments rather than full spectral analysis, the system achieves both high processing speed and accurate arc detection, resolving the contradiction between computational efficiency and detection precision.
Data Source
AI summary
A method for detecting electrical arcs for electrical installations with DC current source, implemented by a detection device including a calculation processor, disposed between the DC current source and a load. A digitized temporal signal is obtained on the basis of the electric current provided by the DC current source. A digitized temporal sub-signal is extracted. A bandpass filtering is applied to the digitized temporal sub-signal. The method also includes calculating a statistical value of order four of the filtered digitized temporal sub-signal, and comparing the calculated statistical value of order four with a first statistical threshold.