Planning of low-voltage distribution systems with uncertainty on load demand in urban areas

In a long-term planning procedure of distribution networks, it is essential to design an appropriate topology in order to satisfy economic and technical aspects. This paper presents a novel algorithm to optimize the topology and phase balancing to tackle the challenge of load demand uncertainty (i.e., growth rate and new load). The paper aims at developing a longterm planning tool of low-voltage (LV) distribution systems to find which load connection phase induces the lowest costs (investment and power losses) and balancing system improvement while satisfying the constraints over the planning horizon. A mixed integer quadratically constrained programming (MIQCP)-arborescence flow and shortest path in parallel with first-fit bin packing are developed to realize this work. In this study, an example of LV distribution system with 33 buses is applied to be a case study of the initial planning year. To evaluate the results, Monte Carlo (MC) simulation method is employed to determine the statistical actualized costs of different strategies. The simulation results support the validity of the methodology proposed in this article.