Optimal Reactive Power Compensation in Distribution Systems using Genetic Algorithms: A Case Study on the IEEE 33-Bus Network

This paper examines the reduction of reactive power in distribution systems through the strategic placement and sizing of capacitor banks utilizing Genetic Algorithms (GAs). The main goal is to cut down on power losses and improve the network's voltage profiles. The GA-based methods offer a straightforward and resilient solution for practical application, in contrast to more complex hybrid algorithms or techniques that necessitate meticulous parameter optimization. The innovation arises from the incorporation of realistic constraints, including a finite number of capacitor banks and dimensional limits, alongside a versatile chromosome encoding tailored for radial systems. The IEEE 33-bus test system is an example of the proposed method in action. The simulation results show that voltage regulation has gotten a lot better and power loss has gone down. We compare recent methods, like the Flower Pollination Algorithm and other metaheuristic techniques, and show how well they work, how fast they are, and how well they can be used in the real world.