Alocação Multi-objetivo de Medidores para Localização de Faltas utilizando Medições de Tensão

This paper presents a Binary Integer Linear Programming (BILP) formulation for the problem of optimal allocation of meters, considering two objectives: minimizing the number of meters and maximizing the number of identifiable faults. The problem was tested on a 15-bus European Medium Voltage System of CIGRÉ, considering the inclusion of a distributed generator and the possibility of operating the system as a weekly meshed network. The optimization problem was solved through the Algorithm for Bicriteria Discrete Optimization (ABCDO), to obtain an exact Pareto Front. Next, the Nondominated Sorting Genetic Algorithm (NSGA-II) was used to find different equivalent solutions for each Pareto Front point. The meter allocations obtained were used to feed a fault locating algorithm based on voltage measurements. The results presented allocations with one or two meters, reaching hit rates in the fault location close to 90%. Further, this work found optimal positions for the meters regardless of the topology variations, which is interesting for the application in modern distribution networks.