Genetic Algorithm used for Improving the Preventive Maintenance Processes for Electricity Distribution Companies

The objective of this work is to improve the preventive maintenance processes used by the electricity distributions companies. This paper is focused on the maintenance of a transformation module in a mid-voltage electrical substation, specifically on the times and prices. Starting from actual data of a mid-voltage electrical substation for a specific transformation module, a Genetic Algorithm was implemented in order to optimize the use of the resources assigned by the company for the preventive maintenance of this module. This optimization was done theoretically by means of simulations based on real data of the substation of a complete year. After several simulations, the proposed algorithm always shows appropriated results, which never overpass the restrictions of available economic resources and out-of-service times of the transformation module, giving the highest priority to the activities unattended for long time. Using the proposed approach, the process of preforming a preventive maintenance process can be highly improved, due to it stops being relative to the expertise of the engineer in charge, which not only produces the best results.