Arc Fault Detection Using Spectral Density and Neural Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing arc fault circuit interrupter (AFCI) devices struggle to accurately detect arc faults due to variations in electrical characteristics, leading to potential safety risks and inefficiencies.
Innovation Solution
The integration of a wideband current sensor, zero cross detection circuit, and a microcontroller with digital filtering and spectral analysis capabilities to estimate arcing characteristics, allowing for precise detection of arc faults based on spectral density comparisons and machine learning models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional arc fault detection methods are used, then the device complexity is low, but the measurement precision and reliability of arc fault detection deteriorate due to variations in electrical characteristics
Solution Approach 1:
The patent applies dynamics by implementing adaptive threshold adjustment based on spectral density analysis. The system dynamically modifies detection thresholds according to the measured spectral characteristics of the circuit, allowing the detection parameters to change in response to varying electrical conditions. This enables accurate arc fault detection across different circuit configurations without requiring fixed, overly complex detection algorithms.
Solution Approach 2:
The patent transitions from traditional time-domain analysis to frequency-domain analysis by implementing spectral density estimation. This dimensional change from time to frequency domain allows the system to identify arc faults based on their characteristic spectral signatures, improving detection accuracy while maintaining manageable system complexity through well-established spectral analysis techniques.
2Measurement precision
If spectral density estimation and machine learning models are implemented, then the arc fault detection accuracy is improved, but the use of energy and computational resources increases
Solution Approach 1:
The patent implements partial action by applying spectral density estimation selectively to detected anomalies rather than continuously analyzing all signals. The system first performs initial screening with simpler methods, then applies the more computationally intensive spectral analysis only when potential arc faults are suspected. This reduces overall energy consumption while maintaining high detection accuracy for actual arc fault conditions.
3Reliability
If digital filtering and spectral analysis are applied to line current measurements, then the detection reliability is improved, but the processing time and complexity of the detection algorithm increases
Solution Approach 1:
The patent segments the signal processing into distinct stages: initial signal acquisition, digital filtering, spectral density estimation, and final detection decision. Each stage processes only the necessary portion of the signal with appropriate complexity, avoiding unnecessary computation. This segmented approach improves reliability through thorough analysis while minimizing overall processing time by eliminating redundant computations.
Data Source
AI summary
A system may include a line terminal. The system may further include a current sensor configured to measure a current flowing through the line terminal. The system may further include a communication circuit configured to transmit a signal including current measurements taken by the current sensor. The system may further include an external device including: a second communication circuit configured to receive the signal; and a controller including an electronic processor configured to: estimate a spectral density of the current measurements, develop a machine learning model based on the spectral density; and deploy the machine learning model to the circuit interrupting device.