Stochastic analysis of economic viability of photovoltaic panels installation for big consumers in Brazil

•Correlation among peak demand, energy generation and consumption.•Simulation with a statistical model.•Optimization model to compute peak demand contract and PV panels quantity.•The method considers the Expected Value and the Conditional-Value-at-Risk.•Analysis performed for a real commercial consumer in Brazil. In the last few years, the growth of solar power worldwide has been remarkable and, different from other sources, it allows small and big consumers to play an important role in the electric system. In Brazil, for big consumers, apart from the benefit associated with energy cost reduction, the photovoltaic system can also reduce the peak demand as well, making the investment even more attractive. Therefore, one of the main challenges for consumers is to accurately estimate the impact of photovoltaic systems on their costs. To do that, the estimation of their future energy consumption, peak demand and energy generation from photovoltaic systems is important to properly compute the economic advantages of such investments. This paper proposes to solve this problem by simulating future energy scenarios of energy consumption, generation and peak demand and correlating them to compute the optimum quantity of photovoltaic panels to be installed and the peak demand contract with the utility by solving a mixed integer linear stochastic optimization model. In the first part of this work, a Box & Jenkins modelling is used to estimate the parameters of the energy consumption, generation and peak demand in a correlated way. In the second part, a stochastic optimization model is applied using a convex combination of the Expected Value and Conditional Value-at-Risk, which were used as risk metrics to compute the optimum number of panels and the best peak demand contract. To illustrate the proposed approach, a case study of a real big consumer is presented, considering a specific contract applied in Brazil. The results allow us to analyze the investment in photovoltaic systems considering the risk level of the consumer and the correlation of all variables involved. In addition to that, the proposed paper can be used as a reference model to be applied in different modalities in Brazil and other countries.