Machine learning based-optimization of a power system

The optimal integration of Distributed Generation (DG) has become critical for enhancing performance and efficiency in modern distribution systems. However, improper siting and sizing of DG units can lead to voltage instability, increased power losses, and reduced network capacity. While traditional optimization methods have been widely employed, they often struggle to address the uncertainties and dynamic complexities of contemporary power networks. This comprehensive review examines machine learning-driven approaches for DG optimization, with particular focus on genetic algorithms for optimal sizing, artificial neural networks for load and generation forecasting, and reinforcement learning for adaptive system control. The paper systematically analyzes the superiority of these ML methods over conventional techniques in handling stochastic environments and improving overall system reliability. Through comparative assessment of current literature, this review identifies key trends, performance metrics, and future research directions for intelligent DG integration in smart distribution networks.

کلیدواژه ها:

Distributed Generation ، Machine Learning ، Power System Optimization ، DG Placement and Sizing ، Renewable Integration ، Smart Grid