Otimização por colônia de formigas da agenda de manutenção de unidades geradoras em usinas hidrelétricas

Abstract

The efficient scheduling of maintenance for power generating units of a hydroelectric power plant is crucial to ensure the reliability and continuous availability of the electrical power generation system. However, the complexities of these tasks, coupled with the specific regulatory constraints of the Brazilian electrical sector, pose challenges to optimizing this process. In this context, this research proposes an approach based on Ant Colony Optimization (ACO) to optimize the maintenance schedule of generating units in a hydroelectric power plant. It is noteworthy that the proposed methodology incorporates detailed considerations of the regulatory aspects related to the Availability Factor (AF), a vital indicator in the context of the operation and regulation of Brazilian hydroelectric plants. The AF is directly impacted by scheduled shutdowns that the generating units of a hydroelectric power plant need to undergo throughout the year. Thus, the main scope of this work is to determine the optimal time to initiate maintenance for each generating unit, framing this optimization within an annual time horizon with daily discretization. The proposed methodology aims to optimize maintenance requirements, aiming to maximize operational efficiency and minimize penalties associated with non-compliance with the Availability Factor defined in the plant’s contract. It is emphasized that the proposed methodology has been applied and validated using real data sets from the Santo Antônio Hydroelectric Plant (SAHP). The SAHP, facing significant financial penalties due to non-compliance with regulatory standards set by the Availability Factor, provides a robust and relevant test environment for evaluating the effectiveness of the proposed approach.