Stochastic Modeling of Irradiation, Applied to Investment Decision in a Photovoltaic Project Subject to Regulatory Shortfall Penalty Rules

Photovoltaic energy generation has expanded its participation in Brazil and presents itself as a promising alternative as a way of meeting future demand. The standard approaches to investment decision taking are the so-called Net Present Value (NPV) of Cash Flows forecasted over a time horizon, and Discounted at a risk-adjusted rate (DCF). Although this approach considers both the expected return and the risk of the analyzed project, it overlooks several uncertainties involved that may constitute risk factors for the project. One of these is the variability of solar irradiation over the years as well as the penalty rules imposed by the regulatory agency. In this paper, monthly average daily measurements of the solar resource were taken at a given location over five years. The statistical distribution that best fit the variability of irradiance over this period is determined, and monthly averages and standard deviations of irradiation are calculated. A photovoltaic solar energy project is simulated as participating in an auction organized by the Brazilian National Agency of Electric Energy (ANEEL) with typical characteristics of auctions held since 2014 to the present date for this generation source, with energy destined to the Regulated Contracting Environment. Monte Carlo method is applied on the irradiation of the model, based on the determined statistical distribution, and the risk of the financial return falling below a specified value is calculated, using a Real Options approach to estimate the effect of energy output shortfall.