Logo do repositório

Percorrer por autor "Kagan, Nelson"

A mostrar 1 - 2 de 2
  • Intelligent Energy Management using CBR: Brazilian Residential Consumption Scenario
    Publication . Fernandes, Filipe; Alves, Douglas; Pinto, Tiago; Takigawa, Fabrício; Fernandes, Rubipiara; Morais, Hugo; Vale, Zita; Kagan, Nelson
    This paper proposes a novel case-based reasoning (CBR) approach to support the intelligent management of energy resources in a residential context. The proposed approach analyzes previous cases of consumption reduction in houses, and determines the amount that should be reduced in each moment and in each context, in order to meet the users’ needs in terms of comfort while minimizing the energy bill. The actual energy resources management is executed using the SCADA House Intelligent Management (SHIM) system, which schedules the use of the different resources, taking into account the suggested reduction amount. A case study is presented, using data from Brazilian consumers. Several scenarios are considered, representing different combinations concerning the type of house/inhabitants, the season, the type of used energy tariff, the use of Photovoltaic system (PV) generation, and the maximum amount of allowed reduction. Results show that the proposed CBR approach is able to suggest appropriate amounts of energy reduction, which result in significant reductions of the energy bill, while, with the use of SHIM, minimizing the reduction of users’ comfort.
    2016Artigo científico Acesso restrito Ver mais
  • Multi-objective robust optimization to solve energy scheduling in buildings under uncertainty
    Publication . Soares, João; Vale, Zita; Borges, Nuno; Lezama, Fernando; Kagan, Nelson
    With the high penetration of renewable generation in Smart Grids (SG), the uncertainty behavior associated with the forecast of weather conditions possesses a new degree of complexity in the Energy Resource Management (ERM) problem. In this paper, a Multi-Objective Particle Swarm Optimization (MOPSO) methodology is proposed to solve ERM problem in buildings with penetration of Distributed Generation (DG) and Electric Vehicles (EVs) and considering the uncertainty of photovoltaic (PV) generation. The proposed methodology aims to maximize profits while minimizing CO 2 emissions. The uncertainty of PV generation is modeled with the use of Monte Carlo simulation in the evaluation process of the MOPSO core. Also, a robust optimization approach is adopted to select the best solution for the worst-case scenario of PV generation. A case study is presented using a real building facility from Brazil, to verify the effectiveness of the implemented robust MOPSO.
    2017Documento de conferência Acesso aberto Ver mais