No abstract

No abstract

To encourage and support innovation, synthetic electric grids are fictional, designed systems that mimic the complexity of actual electric grids but contain no confidential information. Synthetic grid design is driven by the requirement to match wide variety of metrics derived from statistics of actual grids. In order to scale these systems to 10,000 buses or more, robust reactive power planning is needed, accounting for power flow convergence issues. This paper addresses reactive power planning and power flow convergence in the context of large synthetic power grids. The iterative algorithm presented by this paper supplements a synthetic transmission network that has been validated by a dc power flow with a realistic set of voltage control devices to meet a specified voltage profile, even with the constraints of difficult power flow convergence for large systems. The algorithm is illustrated with an example new synthetic 10,000 bus system, geographically situated in the western United States, which is publicly available and useful for a variety of research studies. An analysis is shown validating the synthetic system with actual grid characteristics.

To enable greater innovation in power systems, our research seeks to create entirely fictitious synthetic power system networks that capture the functionality, topology, and defining characteristics of the actual U.S. transmission system, and thus provide realistic test cases for research, without revealing any sensitive information. Creation of these models relies only on publicly available data and statistics derived from the actual grid. This paper outlines two fundamental steps for the creation of synthetic power system models: geographic load and generator substation placement and assignment of transmission line electrical parameters.

Geomagnetically induced currents (GICs) are a result of the changing magnetic fields during a geomagnetic disturbance interacting with the deep conductivity structures of the Earth. When assessing GIC hazard, it is a common practice to use layer‐cake or one‐dimensional conductivity models to approximate deep Earth conductivity. In this paper, we calculate the electric field and estimate GICs induced in the long lines of a realistic system model of the Pacific Northwest, using the traditional 1‐D models, as well as 3‐D models represented by Earthscope's Electromagnetic transfer functions. The results show that the peak electric field during a given event has considerable variation across the analysis region in the Pacific Northwest, but the 1‐D physiographic approximations may accurately represent the average response of an area, although corrections are needed. Rotations caused by real deep Earth conductivity structures greatly affect the direction of the induced electric field. This effect may be just as, or more, important than peak intensity when estimating GICs induced in long bulk power system lines.