Data-Driven Regionalization of Decarbonized Energy Systems for Reflecting Their Changing Topologies in Planning and Optimization

Abstract

The decarbonization of energy systems has led to a fundamental change in their topology since generation is shifted to locations with favorable renewable conditions. In planning, this change is reflected by applying optimization models to regions within a country to optimize the distribution of generation units and to evaluate the resulting impact on the grid topology. This paper proposes a globally applicable framework to find a suitable regionalization for energy system models with a data-driven approach. Based on a global, spatially resolved database of demand, generation, and renewable profiles, hierarchical clustering with fine-tuning is performed. This regionalization approach is applied by modeling the resulting regions in an optimization model including a synthesized grid. In an exemplary case study, South Africa’s energy system is examined. The results show that the data-driven regionalization is beneficial compared to the common approach of using political regions. Furthermore, the results of a modeled 80% decarbonization until 2045 demonstrate that the integration of renewable energy sources fundamentally changes the role of regions within South Africa’s energy system. Thereby, the electricity exchange between regions is also impacted, leading to a different grid topology. Using clustered regions improves the understanding and analysis of regional transformations in the decarbonization process.

The strong effect of network resolution on electricity system models with high shares of wind and solar

“…Other clustering methods applied in the literature are not suitable for the co-optimization of supply and grid technologies: these include clustering based on electrical distance [70,12,23,2,10] (which we do not use because we want to optimize new grid reinforcements that alter electrical distances), spectral partitioning of the graph Laplacian matrix [34] (avoided for same reason), an adaptation of -means called -means++ combined with a max-regions algorithm applied to aggregate contiguous sites with similar wind, solar and electricity demand [65] (avoided since we want a coherent clustering of all network nodes and assets), hierarchical clustering based on a database of electricity demand, conventional generation and renewable profiles including a synthesized grid [47] (avoided for the same reason and because we do not want to alter the topology of the existing transmission grid), -means clustering based on renewable resources as well as economic, sociodemographic and geographical features [20] (avoided because we need a clustering focused on network reduction), as well as clustering based on zonal Power Transfer Distribution Factors (PTDFs) [19,53,64] (avoided because they encode electrical parameters that change with reinforcement), Available Tranfer Capacities (ATCs) [63] (avoided because they depend on predefined dispatch patterns) and locational marginal prices (LMP) [66] (again avoided because they depend on pre-defined dispatch patterns).…”

“…In the planning literature that considers a high share of renewables in the future energy system, the effects of clustering applied separately to wind, solar and demand were investigated in [65], but neglected potential transmission line congestion within large regions. In [47] the previous study was extended by including a synthesized grid and renewable profiles, but it ignored the existing topology of the transmission grid. Effects of varying the resolution were not considered in either of the studies.…”