Résumé:
The electrical network is a complex and interconnected system comprising transmission
lines, substations, and transformers. Its purpose is to transport electricity from power
plants to end users, such as households, businesses, and industries. However, energy
losses within the network pose challenges and lead to inefficiencies in resource utilization.
This research focuses on optimizing power flow to minimize active losses, using
mathematical models to represent network characteristics and power distribution. The
goal is to find optimal network configurations that ensure stability and reliability while
minimizing active losses. Specifically, the study explores the Modified Teaching Learning
Algorithm (MTLA) as a meta-heuristic optimization method and the K-means algorithm
for clustering in power flow optimization (OPF) of electrical networks. By leveraging
these approaches, the research aims to reduce active losses caused by resistance in network
components. This reduction not only enhances overall network efficiency but also has
significant financial implications for electricity companies. The study analyzes modern artificial
intelligence and machine learning techniques to mitigate active losses, contributing
to the development of more efficient and sustainable electrical systems.
