Details

Autor(en) / Beteiligte

• Oliveira, Augusto Cesar Laviola de
• Renato, Natalia dos Santos
• Martins, Marcio Arêdes
• Mendonça, Isabela Miranda de
• Moraes, Camile Arêdes
• Lago, Lucas Fernandes Rocha

Titel

Renewable energy solutions based on artificial intelligence for farms in the state of Minas Gerais, Brazil: Analysis and proposition

Ist Teil von

• Renewable energy, 2023-03, Vol.204, p.24-38

Ort / Verlag

Elsevier Ltd

Erscheinungsjahr

2023

Link zum Volltext

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• Rural areas have great renewable energy potential. With an introduction to sustainable development goals, the smart farm concept presents a novel idea for providing energy in rural areas using artificial intelligence and renewable energy management. We proposed the following topics in this research: (i) methodologies for sizing generation systems by integer linear programming, (ii) use of analytic hierarchy process (AHP) to select the alternative source by financial, environmental, social, and physical criteria, and (iii) training of an artificial neural network (ANN) for process optimization based on the agroeconomic profile of farms. We applied the proposed generic methodology to farms in São Francisco do Glória, Brazil. The biomass, solar, and wind systems were indicated by AHP for implementation in 62.16, 15.32, and 22.52% of farms, respectively. The best configuration of ANN presented a maximum precision of 81.80 ± 3.36%. If the systems were to be implemented, 1.975 GWh yr−1 would be generated, 435.23 tonnes of CO2 would no longer be emitted per year, and a CO2 credit of 366.62 tonnes yr−1 would be injected into the national electric system. Public policies are necessary for this scenario to become a reality in Brazil, such as research incentives and market development. [Display omitted] •Three power generation systems for farms in a small municipality were evaluated.•A decision-making tool was developed considering four indicators.•Economic, financial, social, and physical indicators were used.•Biomass was indicated as an alternative source in 62.16% of farms.•An artificial neural network was trained to optimize alternative source selection.