Análise do curtailment eólico utilizando a simulação de Monte Carlo pseudo-sequencial

Abstract

The increasing penetration of variable renewable energy sources in Brazil,in conjunction with events such as the 2023 blackout, has heightened both the frequency and significance of curtailment events in the brasilian National Interconnected System (SIN). Traditional reliability methods focus on the analysis of energy supply deficits and are widely used in the literature, making them promising tools for curtailment evaluation, given the strong resemblance between curtailment scenarios and load shedding problems. This work proposes a methodology based on Pseudo-Sequential Monte Carlo Simulation (PSMCS) to estimate renewable generation curtailment by leveraging well-established reliability concepts adapted to a surplus energy context. The proposed methodology incorporates the N-1 security criterion, component failures, and power losses, and classifies curtailment events according to the guidelines of the Brazilian regulatory agency ANEEL—namely: energy-related (ENE), reliability-related (CNF), and external unavailability (REL). The approach was applied to the RTS96REN test system, using combinations of historical series of load, wind, and hydrology. Results show that including the N-1 criterion and component failures significantly increases curtailment indices, while accounting for electrical losses reduces the estimated values due to an increase in effective system demand. The methodology proved to be both effective and computationally efficient, presenting a promising alternative for planning and operational studies in power systems with high penetration of variable renewable energy sources.