EURO 2025 Leeds
Abstract Submission

1096. NetZero-Opt: Advancing Time Series Aggregation for Computational Efficiency in Net-Zero Power Systems Optimization

Invited abstract in session MA-44: Advancing Energy System Models, stream Energy Economics & Management.

Monday, 8:30-10:00
Room: Newlyn 1.01

Authors (first author is the speaker)

1. Luca Santosuosso
Institute of Electricity Economics and Energy Innovation, TU Graz
2. Sonja Wogrin
Graz University of Technology

Abstract

Our power grid stands as the most complex human-engineered infrastructure ever created. Decision-making within this intricate system relies heavily on optimization, with the key challenge being to balance modeling accuracy with computational efficiency. Time series aggregation (TSA) methods are employed to streamline the input data for aggregated optimization models, aiming to reduce complexity of full-scale optimization problems while retaining representativeness. However, current TSA methods either rely on a priori input data analysis, mistakenly assuming that minimizing clustering error reduces output error, or use heuristics without theoretical guarantees. The ERC-funded NetZero-Opt project addresses these challenges by advancing TSA. The project stems from the groundbreaking discovery of a new TSA method, leveraging simplex algorithm-based input data clustering. For the illustrative example of an economic dispatch problem, this approach reduces the input space from 8760 hours down to just 3 representative hours, while preserving solution accuracy exactly. These findings demonstrate tremendous potential for reducing complexity in optimization and raise key research questions: (1) Can this method be extended to nonconvex problems? (2) Can it establish a performance guarantee for exact TSA in optimization? This talk will explore the findings of NetZero-Opt and discuss ongoing strategies to address these research questions, as well as other future directions for the project.

Keywords

Status: accepted

Back to the list of papers