Joint reconfiguration of feeders and allocation of capacitor banks in radial distribution systems considering voltage-dependent models

• Autor(a) principal: Home-Ortiz, Juan M. [UNESP]
• Data de Publicação: 2019
• Outros Autores: Vargas, Renzo [UNESP], Macedo, Leonardo H. [UNESP], Romero, Ruben [UNESP]
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.1016/j.ijepes.2018.11.035; http://hdl.handle.net/11449/186633
• topic: Allocation of capacitor banks; Mixed-integer second-order cone programming; Optimization; Reconfiguration of distribution systems; Voltage-dependent models

This paper presents a study about operation planning for radial distribution systems using a mixed-integer second-order cone programming formulation. The proposal consists of a complete model for solving the problems of the reconfiguration of feeders and allocation of capacitor banks simultaneously, with the objective of achieving a minimum overall investment, in both fixed and switchable capacitor banks, and operational costs, related to the cost of energy losses. The approach includes a voltage-dependent load representation, as well as a precise representation of the operation of capacitor banks, which is also voltage dependent. The use of a mixed integer second-order cone programming model guarantees convergence to optimality using existing optimization software. A strategy to reduce the number of candidate nodes of the system for capacitor banks allocation is also presented. Several tests are performed using the 69-node distribution system and a real 2313-node distribution system, demonstrating the effectiveness of the proposal. The results indicate that better solutions can be obtained by performing the simultaneous optimization of the reconfiguration and allocation of capacitor banks than by solving the two problems sequentially, which may even lead to infeasible solutions. It is also verified that both the load and capacitor banks models have an influence on the problem's solution.