Improved Synthetic Power Grid Modeling with Correlated Bus Type Assignments

This paper presents our study results on the correlated assignment of generation, load, or connection buses in a given grid topology and the development of an optimized search algorithm to improve the proposed synthetic grid model, called RT-nestedSmallWorld. A numerical measure, called "Bus Type Entropy", was proposed in an initial study on this subject to characterize the correlation of bus type assignments in realistic grids. In this paper its definition has been redefined and improved with the help of some newly obtained data of realistic grids so that the scaling property of a real-world grid's entropy value versus the network size can be effectively captured with a curve-fitting approach. With the help of the derived scaling function, we will be able to determine an appropriate target entropy value that a correlated bus type assignment should assume in a specific Nbus grid network. Therefore our previously proposed synthetic power grid modeling has been enhanced with a direct search procedure for the best bus type assignments in a specific Nbus grid topology, saving the mandatory, but most the time unattainable, requirement of a set of realistic grid data with a comparable network size for identifying the search target. Finally the performance of the proposed approach is examined based on some available grid data including the IEEE test cases, the NYISO-2935, the ERCOT-5633 and the WECC-16994 systems.