Arc Fault Detection in DC PV Systems Using Frequency Domain Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Arc faults in DC photovoltaic power systems are difficult to detect due to rapid current fluctuations caused by switching noise, leading to potential system failures, shock hazards, and fires.
Innovation Solution
A method involving frequency domain analysis to remove switching spurs and narrow band influences from current flow data, using notch filters and curve fitting to identify and attenuate these interferences, allowing for effective arc fault detection by comparing processed data with a threshold value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequency domain analysis is performed on current flow to detect arc faults, then arc detection capability is improved, but switching spurs and noise interfere with accurate detection
Solution Approach 1:
The patent extracts and removes switching spurs from the frequency domain representation by identifying their characteristic patterns (narrow bandwidth, specific frequency locations) and subtracting them from the total spectrum. This isolation technique separates the harmful switching artifacts from the arc fault signals, enabling accurate arc detection despite the presence of converter noise.
Solution Approach 2:
The patent applies different processing treatments to different frequency regions. Narrow band spurs are identified and removed using specific algorithms tailored to their characteristics (narrow bandwidth, high amplitude), while the remaining frequency spectrum is analyzed for arc faults. This localized processing approach optimizes detection accuracy for each specific interference type.
2Productivity
If switched mode inverters are used in photovoltaic systems, then power conversion efficiency is improved, but rapid current fluctuations and noise are generated that complicate arc fault detection
Solution Approach 1:
The patent performs preliminary processing of the current signal by transforming it to the frequency domain and removing switching spurs before conducting arc fault analysis. By pre-processing the signal to eliminate known interference patterns (switching frequency harmonics, narrow band spurs), the system prepares a cleaner signal for subsequent arc detection, reducing the complexity of the detection task.
Solution Approach 2:
The patent introduces an intermediary processing stage that transforms the time-domain current signal into the frequency domain, where switching artifacts and arc faults can be distinguished based on their spectral characteristics. This frequency domain representation serves as an intermediary that facilitates the separation and identification of different signal components.
3Reliability
If narrow band spurs are present in the frequency domain representation, then false arc detections may occur, but removing them requires additional processing complexity
Solution Approach 1:
The patent changes the analysis parameters by examining specific characteristics of frequency components (bandwidth, amplitude, frequency location) to distinguish narrow band spurs from arc faults. By monitoring changes in these parameters over time and comparing them against known patterns, the system identifies and removes spurs without requiring complex additional hardware or algorithms.
Data Source
AI summary
Power systems having a DC content, such as photovoltaic (solar) panels present a problem if an arc fault appears because of a small break in a cable. The present disclosure describes an arc fault detection system that captures data in segments, examines the frequency spectrum to remove ‘false arc’ signatures and interference from a power converter of the power system, and then examines the cleaned frequency spectrum for arc events.