Study and development of a multiobjective algorithm applied to the restoration problem of electricity distribution systems

• Grant number: 20/07370-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2020
• End date: August 31, 2021
• Field of knowledge: Engineering - Electrical Engineering - Power Systems
• Principal Investigator: Antonio Marcos Cossi
• Grantee: Joao Pedro Trindade Carrit Delgado
• Host Institution: Faculdade de Engenharia (FEIS). Universidade Estadual Paulista (UNESP). Campus de Ilha Solteira. Ilha Solteira , SP, Brazil

Abstract

In this Scientific Initiation work, an algorithm will be developed, in a theoretical way, to be used in a multiobjective model applied to the problem of restoration of electricity distribution networks. The algorithm should contemplate the joining of two existing single-object optimization models, in order to create a multi-objective model, with two distinct objectives. The first model (objective 1), resolves the PRSDEE seeking to minimize the sections of the network that would be without power after the system is restored, subject to the technical, physical and operational restrictions of the distribution network. This model is solved using the solution technique known as Basic Variable Neighborhood Search (BVNS), which in Portuguese means Basic Search in Variable Neighborhood. The second model (objective 2), seeks to determine the amount of loads that can be disconnected by the electric utility, from customers who are part of a demand response program, based on incentives. By incorporating the demand-response model in the restoration model, the chances that during a restorative process, a larger set of system loads will be restored are increased. The multiobjective model algorithm will also be based on the BVNS algorithm, and therefore, it will be called BVNS-MO. Thus, the work consists of an adaptation of the BVNS algorithm of the single objective model incorporating the analysis of the results of the demand response model. It should be noted that the mathematical model itself will not be developed, but only the BVNS-MO algorithm, in theoretical form. In the future, it is expected to continue this work of Scientific Initiation through its implementation, simulations and analysis of results.