Planificación y Diseño Automático Óptimo de Grandes Redes Eléctricas de Distribución: Aplicación al Centro Histórico de Quito

Abstract

This paper introduces the computer program developed by CoDyPower (CDP) for automatic planning of large electrical distribution networks. The main objective is the design of electrical networks that minimize the economic cost of the installation and operation. The cost function to be optimized takes into account the number and type of transformers, the longitude and type of cables, the type and cost of overhead/underground lines, and the operating costs in a given time window. The optimization deals with the street geometry and the catalog of available transformers and cables. It considers distance constraints, voltage and current limits and maximum power. Due to the NP-hard character of the computational problem, the software applies artifi cial intelligent strategies and statistics. It splits the problem into macro and micro optimization sub-tasks, fi nding a feasible calculation technique for a practical computational time. Moreover, the proposed algorithms are able to fi nd multi-circuit solutions per transformer. The CDP software is applied to the automatic planning of the low and medium voltage electrical distribution networks for the Historic Center of Quito, with about 20,000 consumers. The achieved solution meets all the specifi cations and reduces by 40% approximately the initial estimated cost.