A Model-Adaptive Clustering-Based Time Aggregation Method for Low-Carbon Energy System Optimization

Yuheng Zhang,Vivian Cheng,Dharik S. Mallapragada,Jie Song,Guannan He

DOI: https://doi.org/10.1109/tste.2022.3199571

IF: 8.31

2023-01-01

IEEE Transactions on Sustainable Energy

Abstract:Intermittent renewable energy resources like wind and solar introduce uncertainty across multiple time scales, from minutes to years, on the design and operation of power systems. Energy system optimization models have been developed to find the least-cost solution that manages the multi-timescale variability using an optimal portfolio of flexible resources. However, input data that capture such multi-timescale uncertainty are characterized with a long time horizon and high resolution, which brings great difficulty to solving the optimization model. Here we propose a model-adaptive time aggregation method based on clustering to alleviate the computational complexity, in which the energy system is solved over selected representative time periods instead of the full time horizon. The proposed clustering method is adaptive to various energy system optimization models or settings, because it extracts features from the optimization models to inform the clustering process. Results show that the proposed adaptive method can significantly lower the error in approximating the solution of the optimization model with the full time horizon, compared to traditional time aggregation methods.