Application of Machine Learning for Optimal Wind Farm Location

Abstract

Presence of renewable sources of energy in power systems is vital to cope the negative impacts of environmental climate change. The drift in autonomous power network situations offers a strong possibility of wind generation to become one of the key contributors in sustainable energy. This paper presents a methodology for determining the optimal location of a wind farm in a power transmission network based on security assessment. The optimization problem is devised with an objective of minimizing mean system operating cost, considering both (N-1) line and (N-1) bus outages. Moreover, machine learning algorithm is applied to predict the optimal wind farm location in a computationally efficient manner. The IEEE 39-bus test system is used to test and validate the effectiveness of the proposed approach. DIgSILENT PowerFactory and MATLAB were utilized for optimal power flow simulations and machine learning prediction algorithm, respectively. The results give a unique solution for optimal location of wind generation, along with a priority order list, which is useful when integrating multiple wind farms in the power transmission system.