Application of genetic algorithm for optimization on projects of public illumination

This paper presents a proposal for minimizing costs of public illumination projects (PI), using the optimization technique of genetic algorithm (GA). The solution is found from the characteristics of public roads (input parameters) and existing norms and standards, which sets the parameters for evaluating each project, ensuring the minimum requirements for safety and comfort efficiently. In order to evaluate the results of the proposed algorithm, tests were performed, based on the characteristics of several existing roads in the state of Sergipe, Brazil, with the distribution of masts in accordance with the following standards: Unilateral, Bilateral Opposite and Central. The implemented GA provides as solution the height of the mast to be used, in addition to the distance between the masts, the number of petals and their orientation, arm size, luminaries model and power of the bulbs used. The photometric features of the set formed by the luminaries and lamp are presented in a file of type IES, like those created from tests conducted by the manufacturers of the models. The results show a good performance of the proposed methodology.