Multi-Objective Optimization of Electric Distribution Systems with integrated distributed Generation using Deep Reinforcement Learning

This paper proposes a method for optimizing the placement and capacity of Distributed Generators (DGs) in distribution systems based on Deep Reinforcement Learning (DRL). The objective of the method is to minimize power losses, investment costs, voltage deviations, and CO2 emissions while ensuring strict compliance with system operating constraints. The proposed approach leverages the robust capabilities of DRL to handle nonlinear and complex-constrained problems, making it highly adaptable to various operational scenarios. Experimental results on standard distribution systems demonstrate that the proposed method outperforms traditional algorithms, significantly improving operational efficiency and enhancing the integration of renewable energy sources. This contributes to the development of smart grid systems and promotes sustainable energy solutions.