An intelligent computing technique to estimate the magnetic field generated by overhead transmission lines using a hybrid GA-Sx algorithm

摘要

The application of certain Artificial Intelligence techniques provides an efficient solution to the problem of characterizing the magnetic field of a high voltage overhead transmission line, and is an alternative to the expensive procedure of direct measurements, which requires equipment and time, to the use of complex numerical methods of a very specific scope, or of simply obtaining a theoretical value calculated using analytical procedures which forego the quality of the solution in favor of simplifying the calculations. This paper presents an implementation based on a hybrid algorithm in which the best solutions provided by a metaheuristics (a genetic algorithm which allows working with extensions) define the initial simplex for the application of the Nelder-Mead Method, which as a local search method permits a calculation-intensive search. In order to validate the quality of the results generated by this hybrid implementation, the estimates obtained are compared with measured values and with values obtained by means of analytical procedures.