The model externality in the Brazilian power system: a call for a new data and model governance

Keynote session 2 (TAI 2024, PUC-Rio, October 31, 2024)

Speaker: Alexandre Street

Abstract: The Brazilian power system heavily relies on complex models to determine generation and reserve schedules, price signals, and security standards, which are the foundation for market incentives, flexibility utilization, and supply adequacy. The need to modernize these models and data governance becomes paramount as the energy sector undergoes a transformative shift towards low-cost and low-emission electricity. This presentation addresses the externalities of outdated modeling practices and advocates for a robust data and model governance framework. Specifically, it highlights the critical role of storage opportunity cost assessment and the impact of optimistically biased uncertainty forecasts and system description in the policy forward-looking. We advocate for a new governance structure that brings incentives for increasing the efficiency of data-driven dispatch and pricing rules in reducing costs, increasing reliability, and signalizing the system’s opportunity costs for market participants. Within this new governance, we propose 1) the regular benchmarking of official data and models against independent standards, 2) new processes to internalize ad hoc out-of-the-market rules, and 3) accountability based on monitoring indexes for adherence and accuracy of the inputs and outputs of the models against observed data.

Download the PDF file

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

_]