Improved selection of critical network elements for flow-based market coupling based on congestion patterns

Abstract

Flow-based market coupling provides zonal day-ahead markets with appropriate signals of possible real-time congestions by incorporating information on local load and generation patterns. It relies on predictive parameters, notably the base case and generation shift keys. Also it only monitors part of the grid, through selecting critical network elements. In consequence, it naturally falls short of a nodal pricing-based cost-optimal solution. Based on a test network with three flow-based market coupling zones, we show that the results of flowbased market coupling converge to nodal pricing solutions with an increasing amount of re-configured market zones. We identify if re-configured market zones can help to improve the selection of critical network elements and lead to cost reductions even in the original market zone setting. We find that around 90% of the cost reductions from a market zone re-configuration can be maintained when the critical network elements, obtained from the re-configured market zones, are used for the original market zones. This is a strong indication that, both in reality as well as model-based research of flow-based market coupling, the selection of critical network elements should be based on expected congestion patterns. To find these congestions, we conduct a nodal pricebased market zone re-configuration that helps to identify lines with different congestion signals. This approach can constitute a helpful addition to static and assumption-based selection criteria for critical network elements, such as the often-used zone-to-zone power transfer distribution factors that strongly rely on assumptions like generation shift keys.

A clustering approach to the definition of robust, operational and market efficient delineations for European bidding zones

“…They applied K‐Means and Hierarchical clustering on single dispatch scenarios in order to obtain series of grid partitions, then used spectral clustering to build coherent zone delineations regarding scenario weighting and grid topology. This differs from previously mentioned initiatives like [11, 20, 25] that use multiple scenarios—thus multidimensional LMP vectors—as a direct input for K‐Means or Hierarchical clustering. This approach therefore returns grid partitions that are robust to the full set of input dispatch scenarios.…”

“…[20], Burstedde [21], or Breuer and Moser [22]. From a more recent perspective, contributions have been restraining their system operation assessment to the measurement of intrazonal and cross‐zonal LMP differentials, as in [11, 15, 19, 23–25]. This characterizes a strong focus on the sophistication of grid partitioning towards the alignment of zone borders with grid congestions.…”

“…This approach was notably adopted by Felling and Weber [11] and Schônheit et al. [25]. However, top‐down algorithms like K‐Means are less suitable for adjacency constraints, which leads authors to integrate spatial coordinates in their calculation of the distance between buses.…”

Decarbonizing the industry sector and its effect on electricity transmission grid operation—Implications from a model based analysis for Germany

“…The following model description strips down the equation system to a minimum to foster an increased understanding of the model exercise carried out in this research. A full description of the base version of ELMOD herein applied is provided in Schönheit et al (2020). Further reading on modelling congestion management can be found in Hobbie et al (2022).…”

Uncertainty-Aware Capacity Allocation in Flow-Based Market Coupling

“…Over the past years, several academic publications provided various modeling approaches to the different components of FBMC. Byers and Hug (2020) provided insights into hte fundamental modeling process and numerically highlights the importance of its parametrization, Schönheit et al (2020b); Finck et al (2018) evaluated the impact of different generation shift keys, a core parameter that describes how generators will participate in cross-border exchange, Schönheit et al (2021a) described the relation of bidding zone configuration on the selection of network elements that are part of the capacity calculation.…”