ANA BEATRIZ CARVALHO WERLANG

Título: Multistage Hydroelectric Bidding Problem with Regularized Linear Decision Rules

Data: 03/09/2024, 14h

Orientadores: Alexandre Street de Aguiar | PUC-Rio & Davi Michel Valladão | PUC-Rio

Resumo: This study introduces an innovative methodology aimed at optimizing revenue for hydropower producers amidst the complexities of competitive energy markets. Conventional static models often fail to adequately capture the dynamic interplay between water inflows and market conditions, leading to suboptimal revenue outcomes. In contrast, the proposed approach leverages Linear Decision Rules (LDRs) to dynamically adjust bidding curves in response to evolving scenarios. By optimizing revenue while accounting for operational constraints and uncertainties, the proposed model offers a robust framework for bidding strategy enhancement. A comparative analysis between the dynamic approach and static methodologies showcases substantial enhancements in revenue generation for hydropower assets. Notably, the dynamic approach not only outperforms static methods but also achieves comparable results to system optimization benchmarks with significantly reduced data requirements and computational effort, thus offering a promising avenue for enhancing the profitability of hydropower operations in competitive energy markets.