AI Fault Localization for Power Transmission Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional power supply systems face challenges in accurately and reliably localizing faults in power transmission lines due to the dependence on parameter choices like optimal measurement points and window sizes, which can lead to inaccurate fault detection and propagation of faults within the grid.
Innovation Solution
An in-field apparatus equipped with a preprocessing unit for z-score normalization and an artificial intelligence module that performs feature-based or direct fault classification using neural networks, including random forests, decision trees, and convolutional or recurrent neural networks, to determine optimal evaluation times and window sizes for precise fault localization, trained on historical and simulated fault records.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional impedance calculation methods are used for fault localization, then the system can identify fault locations, but the accuracy strongly depends on parameter choices (optimal measurement point and window size) which reduces reliability
Solution Approach 1:
The patent transforms the fixed parameter selection problem into a dynamic optimization problem by using AI algorithms to automatically determine optimal measurement points and window sizes based on the specific fault conditions and signal characteristics, eliminating the need for manual parameter tuning and improving both accuracy and reliability
Solution Approach 2:
The patent replaces the conventional mechanical impedance calculation method with an AI-based signal processing system that uses neural networks and pattern recognition to analyze fault signals, substituting mathematical calculations with intelligent algorithms that can adapt to varying fault conditions
2Measurement precision
If AI-based fault classification is implemented, then prediction accuracy is improved, but computational resource requirements increase
Solution Approach 1:
The patent applies partial AI processing by using lightweight neural network models and selective feature extraction that processes only the most relevant signal characteristics, achieving sufficient accuracy without requiring full computational power of complex AI systems
Solution Approach 2:
The patent segments the fault detection process into multiple stages: preliminary signal filtering, feature extraction, and targeted AI classification, allowing computational resources to be distributed efficiently across different processing steps rather than requiring intensive computation throughout
Data Source
Figure 1~2
Figure 3
AI summary
An in-field apparatus and method for automatic localization of a fault having occurred at power transmission lines of a power supply system, the in-field apparatus (1) comprises a preprocessing unit (2) adapted to process measured voltage and/or current raw time series data of the power transmission lines to provide a normalized raw data and/or feature representation of the measured raw time series data; and an artificial intelligence module (3) configured to predict an optimal evaluation time used for evaluation of the measured voltage and/or current raw time series data to localize the fault based on the normalized raw data and/or feature representation.