Fault detection and classification in electrical power transmission system using artificial neural network

Area of Science:

• Electrical Engineering
• Artificial Intelligence
• Power Systems

Background:

• Electrical power transmission lines are critical infrastructure.
• Faults in transmission lines can lead to widespread power outages.
• Accurate and timely fault detection and classification are essential for grid stability.

Purpose of the Study:

• To develop an efficient method for detecting and classifying faults in electrical power transmission lines.
• To utilize artificial neural networks for fault analysis.
• To validate the performance of the proposed neural network model.

Main Methods:

• Utilized three-phase currents and voltages from one end of the transmission line as input data.
• Employed a feedforward neural network with a backpropagation algorithm for fault detection and classification.
• Conducted detailed analysis with varying hidden layers to optimize the neural network architecture.
• Simulated various fault types with different parameters in the MATLAB environment.

Main Results:

• The proposed artificial neural network method demonstrated high efficiency in detecting and classifying transmission line faults.
• Satisfactory performance was achieved in identifying faults across all three phases.
• The model showed versatility in handling different fault scenarios and parameters.

Conclusions:

• The artificial neural network-based approach is effective for fault detection and classification in electrical power transmission lines.
• The method offers a reliable and efficient solution for enhancing power system reliability.
• The proposed technique has potential for extension to distribution networks.