ALGORITMO PARA ESTUDO DO FLUXO DE POTÊNCIA: análise para sistemas com até dez barras

Abstract

Considering the great importance of the load flow analysis for the electric power system, the present work made a study on the flow and presented the methods for solving the problems of the load flow, where greater attention was given to the method of Newton Raphson. Therefore, the objective of this work was to develop an algorithm for calculating the power flow with systems of up to ten bars without the presence of transformers, using as a solution basis the Newton Raphson iterative method. Knowing the difficulty and complexity in the manual calculations for the study of power flow, the idea of the development of this algorithm is to use it later, as an academic tool, assisting in the Electrical Power System discipline of the Electrical Engineering course when studying flux of load. In this way, students will be able to manipulate and analyze the behavior of systems with a larger number of bars, since manually it would be something very complex and would require a lot of time. The development of such a program was based on researches and studies of bibliographic technical materials for the theoretical basis and was later created with the Matlab computational tool. It can be concluded that the results presented by the program, reached satisfactory levels, and, in this way, can be used as a tool for academic assistance. With the project completed and the algorithm developed, a validation was made so that the results obtained with the execution of the program were of reliable values. Therefore, a test example was used for validation, in which the results presented by the developed algorithm reached satisfactory values. The values obtained by the source code present a small numerical difference due to the book development method (manual method) and the algorithm (computational method). Besides this reason, numerical rounding should be taken into account, considering that the computational program used of all decimal values for calculations. However, it was shown that the small divergence in the values did not affect the reliability of the values generated by the algorithm.