Revolutionizing Power Distribution: Leveraging TLBO and Non-Linear Regression for Sustainable Energy Management

This paper presents a comprehensive investigation into the optimization of power losses in a distribution system, focusing on the IEEE 15 bus system as the testbed. Initially, power losses are assessed through power load modelling, with subsequent refinement utilizing voltage load modelling to mitigate losses. A comparative analysis between these models underscores their efficacy in loss reduction. The introduction of Electric Vehicle (EV) charging stations introduces load uncertainty and exacerbates power losses within the system. To address this challenge, Distributed Generation (DG) systems, including Photovoltaic (PV) and wind systems, are strategically deployed in the distribution network to minimize losses. Furthermore, the study explores the pivotal role of nonlinear regression as a post-processing technique to refine solutions obtained from optimization algorithms. Leveraging nonlinear regression, particularly Gaussian function modelling, enhances the accuracy and precision of the optimization process, leading to improved loss reduction strategies. Moreover, the optimization process extends to the adjustment of transformer tap settings using an advanced algorithm inspired by Teaching-Learning-Based Optimization (TLBO). This iterative refinement of tap settings aims to minimize power losses and enhance system efficiency. Through comprehensive simulations and analyses, this study underscores the significance of integrating advanced optimization techniques, renewable energy sources, and nonlinear regression methodologies in reducing power losses within distribution systems. The findings offer valuable insights for enhancing the operational efficiency and sustainability of distribution networks in the face of evolving energy demands.