Akademik Veri Yönetim Sistemi
International Symposium on Advanced Engineering Technologies 2, Kahramanmaraş, Türkiye, 16 - 18 Haziran 2022, ss.82-83
It is an indisputable fact that the need for electrical energy and the energy consumption per capita are increasing dayby day. According to World Bank data, the energy consumed in Turkey in 1960 was 92,311 kWh; In 1992, it exceeded1000 kWh for the first time and became 1044.36 kWh; In 2014, it reached 2847.22 kWh. Similarly, energy consumptionis increasing worldwide. For this reason, studies on reducing energy losses in electrical transmission and distributionsystems always remain popular in the literature. To this end, the optimal reconfiguration of distribution networks ishandled in different ways. Considering that the energy losses in our country and in similar developing countries arearound 14%, it is clear that reducing energy losses through reconfiguration studies will be of great benefit. Numerousstudies have been conducted on this subject in the literature. The studies include an algorithmic method, objectivefunction and problem constraints. Optimization algorithms such as artificial neural networks, probabilistic approaches,fuzzy logic, genetic algorithms, evolutionary programming, and recently particle swarm optimization (PSO) have beenused as a method. As the objective function, it is focused on problems such as reducing energy losses, reliabilityindices, voltage stability, voltage profile, economic issues, load balancing, power quality and optimum capacitorplacement. Limits of current and voltage and constraints such as reliability also play a crucial role in the results ofalgorithms. Today, it is seen that the particle swarm optimization method has started to be used in reconfigurationstudies in addition to other algorithms. The particle swarm optimization method was first proposed by James Kennedyand Russell Eberhart in 1995. Afterwards, many academic studies have been carried out in various fields related to thismethod. Particle swarm optimization is similar to genetic algorithms, but instead of using genetic operators, it enablesthe particles to evolve through iterations through cooperation and competition among themselves. Each particledetermines its motion in the next iteration according to its own flight data and the flight data of other particles. Particleswarm optimization is an algorithm that relies on non-differential, simple, few variable parameters and produces quitegood results, thus increasing its popularity. In this study, a comparison is made between the methods used for thereconfiguration problem in the electricity distribution network and the PSO algorithm. Also, a literature review of thestudies in which this algorithm is proposed has been conducted.