AI-based hybrid power quality control system for electrical railway using single phase PV-UPQC with Lyapunov...

Area of Science:

• Electrical Engineering
• Artificial Intelligence
• Renewable Energy Systems

Background:

• 25 kV AC traction networks face significant power quality issues like voltage unbalance and high Total Harmonic Distortion (THD).
• Existing power quality management systems struggle with dynamic load changes and integrating renewable energy sources effectively.

Purpose of the Study:

• To develop and validate an advanced AI-driven hybrid power quality management system for 25 kV AC railway networks.
• To address critical power quality challenges including voltage unbalance, THD, voltage variations, and power factor.
• To integrate a single-phase Photovoltaic Unified Power Quality Conditioner (PV-UPQC) with an Artificial Neural Network (ANN)-Lyapunov control architecture.

Main Methods:

• Implementation of a hybrid system combining a single-phase PV-UPQC with an ANN-Lyapunov control architecture.
• Utilizing a dual-approach methodology: ANN-based reference signal generation and Lyapunov optimization for dynamic parameter tuning.
• Employing MATLAB/Simulink for system simulation and performance validation.

Main Results:

• Achieved significant reduction in voltage unbalance from 1.5% to 0.8%.
• Reduced Total Harmonic Distortion (THD) to below 1% and corrected power factor to unity.
• Demonstrated a 40% faster dynamic response and DC link voltage regulation within ±2%.
• Maintained an overall system efficiency of 95%.

Conclusions:

• The proposed AI-driven hybrid system effectively enhances power quality and energy efficiency in electrical railways.
• The integration of ANN-based control, Lyapunov optimization, and PV technology offers a robust solution for modern traction networks.
• The system's validated performance confirms its capability to manage complex power quality challenges in 25 kV AC railway systems.