Optimal Placement of Grid-Connected Wind Farms Based on Artificial Intelligence Techniques

Abstract

From ancient times to the present day, the wind has been consistently recognized as one of the most important, accessible, and environmentally friendly energy sources, which today has a significant share in global electricity generation. The limited resources of fossil fuels are running out day by day, and the exploitation of this type of energy produces harmful pollutants. Consequently, double attention is paid to wind power and similar free and sustainable resources. Attaining the optimal location for establishing wind farms and turbines preeminently reduces construction/transmission costs and energy-related losses. In light of the above, we employ several artificial intelligence (AI) algorithms to optimally identify and locate the grid-connected wind farms, significantly serving electricity energy manufacturers and distribution companies. In addition, this study aims to maximize the efficiency of the wind energy system by considering the critical factors influencing the production capacities, such as wind speed, air density, turbine size, and geographical location. This work leverages an image processing-based technique, density-based spatial clustering of applications with noise (DB-SCAN), k-means, fuzzy c-means (FCM), and k-medoids machine learning algorithms. Numerous experiments and simulations are designed and performed exploiting MATLAB/Simulink, all of which demonstrate the substantial superiority of the proposed systems.