Avaliação estocástica da viabilidade econômica de projetos para produção de hidrogênio verde mediante energia solar fotovoltaica em Minas Gerais

Abstract

The production of green hydrogen has been considered a promising alternative in the energy transition due to its decarbonization potential. However, challenges associated with high production costs, technological uncertainties, and macroeconomic fluctuations still limit its economic forecasts in different contexts. Given this, this study conducted a stochastic scenario analysis using the Monte Carlo method to evaluate green hydrogen production projects through photovoltaic systems in the state of Minas Gerais, Brazil. The modeling was based on seven critical variables: CAPEX of photovoltaic systems and electrolyzers, solar irradiation, hydrogen (H₂) selling price, exchange rate, interest rate, and type of electrolyzer (PEM, AEL, SOE). In the simulation, parameters based on the literature were used, which resulted in only 6.16% of scenarios with a positive net present value. The results showed that viable projects were mostly associated with an electrolyzer CAPEX below R$ 1,000/kW and a photovoltaic system CAPEX between R$ 2,800/kWp and R$ 3,800/kWp. In non-viable scenarios, on the other hand, these values exceeded R$ 2,400/kW and R$ 4,900/kWp, with the SOE electrolyzer standing out as particularly costly. The hydrogen selling price was set between R$ 30/kg and R$ 40/kg, and it was observed that prices below R$ 30/kg had low economic probability. The modeling indicated that the interest rate should be below 10% per year. Lower rates, between 1% and 8% per year, increased project profitability. Conversely, higher rates, such as 14%, resulted in a negative NPV. Solar irradiation was highlighted as a determining factor, with values between 5.0 and 6.5 kWh/m²/day being necessary for viability. Values below this range rendered the project economically unfeasible. Additionally, certain micro-regions in Minas Gerais were found to have suitable conditions for green hydrogen production. Using a computational tool developed in this study, a sensitivity analysis revealed that the NPV of the projects is strongly influenced by the CAPEX of electrolyzers, followed by the hydrogen selling price and the interest rate, respectively.