DC Arc Detection Using Neural Networks for PV Fault Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting direct-current electric arcs in photovoltaic systems suffer from low accuracy in real-time detection, posing a risk of fire and damage due to the persistent nature of direct-current arcs.
Innovation Solution
A method utilizing neural network models for burning and starting arc detection, combined with Fourier transform and per unit normalization, to improve the accuracy of identifying direct-current electric arcs by analyzing electrical quantities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual searching for electric arc characteristics is used, then the detection method is simple, but the detection accuracy is low
Solution Approach 1:
The patent replaces manual characteristic searching (mechanical/systematic approach) with a neural network model (intelligent system). The neural network automatically learns and extracts electric arc characteristics from current waveforms, eliminating the need for manual feature engineering and threshold setting, thereby significantly improving detection accuracy while maintaining reasonable system complexity
Solution Approach 2:
The patent transforms the detection approach by changing from fixed manual thresholds to dynamic neural network-based parameter extraction. The neural network adaptively determines detection parameters based on learned patterns from training data, allowing the system to adjust to varying operating conditions and improve accuracy without requiring complex manual parameter tuning
2Reliability
If conventional detection methods are used, then the system is simple, but misidentification rate is high
Solution Approach 1:
The patent replaces conventional threshold-based detection mechanisms with a neural network-based intelligent detection system. The neural network learns complex patterns and relationships in the data that are difficult to capture with simple threshold comparisons, thereby reducing misidentification rates and improving reliability
Solution Approach 2:
The neural network model incorporates feedback mechanisms through its learning process, where detection results and performance metrics are used to refine and improve the model over time. This feedback loop enables the system to reduce misidentification rates by continuously learning from past performance, enhancing reliability without requiring proportional increases in system complexity
Data Source
AI summary
A direct-current electric arc detection method and apparatus, a device, a system, and a storage medium. The method includes: obtaining an electrical quantity of a direct current circuit; inputting the electrical quantity into a burning arc neural network model to perform burning arc detection on the direct current circuit, to obtain a burning arc detection result; and if the burning arc detection result is that a burning arc is detected, determining that a direct-current electric arc fault exists in the direct current circuit. Because the electrical quantity is input into the burning arc neural network model to perform burning arc detection on the direct current circuit, a direct-current electric arc fault is detected, and accuracy of detecting a direct-current electric arc fault is further improved.