Active distribution system planning considering non-utility-owned electric vehicle charging stations and network reconfiguration

• Main Author: Mejia, Mario A. [UNESP]
• Publication Date: 2023
• Other Authors: Macedo, Leonardo H. [UNESP], Muñoz-Delgado, Gregorio, Contreras, Javier, Padilha-Feltrin, Antonio [UNESP]
• Format: Article
• Language: eng
• Source: Repositório Institucional da UNESP
• Download full: http://dx.doi.org/10.1016/j.segan.2023.101101; https://hdl.handle.net/11449/302954
• topic: Active distribution networks planning; Electric vehicle charging stations; Multi-objective optimization; Network reconfiguration; Renewable distributed generation

This paper introduces a new mixed-integer linear programming formulation for planning active distribution networks and non-utility-owned electric vehicle charging stations (EVCS), using network reconfiguration as a planning option. The method employs multi-objective optimization to address the commercial interests of the utility and the EVCS owner separately. In this manner, the utility makes investment decisions for network assets, such as replacing overloaded conductors and installing capacitor banks and voltage regulators, whereas the EVCS owner decides on the purchase of land and infrastructure for the EVCS installation. Furthermore, as an outstanding feature, the proposed formulation incorporates network reconfiguration within the planning options. The proposed model aims to minimize the total expected cost for both owners. A travel simulation algorithm provides the EVCS load profiles. At the same time, scenario-based optimization addresses uncertainties related to substation energy prices, wind speed, solar irradiance, electricity demand, adoption rate, and loading profile of plug-in electric vehicles. The effectiveness of the proposed model has been demonstrated in a 69-node network. The findings show that when deciding where to install the EVCS (network node), the objectives of both owner's clash. This discrepancy, however, can be reduced through network reconfiguration. Finally, the scalability of the proposed method was evaluated using a real 134-node system.