Advancing Power Supply Resilience: Optimized Transmission Line Retrofitting Through Deep Q-Learning Algorithm
Year: 2025
Type: article
Abstract: This study explores practical approaches to improving the reliability of power supply systems through the expansion and optimization of substation power lines. As electricity demand steadily increases, ensuring a stable and efficient power delivery network has become essential to support industrial growth and socio-economic development. This study focuses on challenges such as vulnerability to single-line faults, limited transmission capacity, and complex coordination in system operation. To address these issues, the proposed strategy includes building redundant transmission lines, improving network configuration, and applying modern transmission technologies to enhance operational flexibility. Notably, a Deep Q-Learning algorithm is introduced during the planning and optimization process. Its ability to accelerate convergence and streamline decision making significantly reduces computation time while maintaining solution accuracy, thereby increasing overall efficiency in evaluating large-scale network configurations. Simulation results and case studies confirm that such improvements lead to shorter outage durations, enhanced fault tolerance, and better adaptability to future load demands. The findings highlight strong practical value for industrial applications, offering a scalable and cost-conscious solution for strengthening the reliability of modern power systems.
Source: Energies
DOI: 10.3390/en18164335
Authors: Lin Liu, Tianjian Wang, Xiuchao Zhu, Chenming Liu
Language: English
Subfield: Electrical and Electronic Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,137
OpenAlex ID: W4413416873