This paper proposes a new methodology for the evaluation of reliability in radial distribution networks through the identification of new investments in this kind of networks, in order to reduce the repair time and the failure rate, which leads to a reduction of the forced outage rate and, consequently, to an increase of reliability.
The novelty of this research work consists in proposing an ac optimization model based on mixed-integer nonlinear programming that is developed considering the Pareto front technique, in order to achieve a reduction of repair times and of failure rates of the distribution network components, while minimizing the costs of that reduction, the power losses cost, the cost of the optimal capacitor location and size, and the maximization of reliability, which is in the form of minimization of nonsupplied energy cost. In order to estimate the outage parameters, a fuzzy set approach is used. The optimization model considers the distribution network technical constraints. A case study using a 33-bus distribution network is presented to illustrate the application of the proposed methodology.