Optimizing the reactive power balance between a distribution and transmission grid through iteratively updated grid equivalents

This paper focuses on a scenario with a high amount of renewable generators (DGs) in the distribution grid; a local grid operator (DSO) utilizes reactive power provision by the DGs, to improve the reactive power balance at the connection points to the transmission grid. At the same time, the transmission grid operator (TSO) aims to optimize his voltage, by computing reactive power setpoints for the DSO. This is a decentralized optimization problem, where two optimizers ("DSO" and "TSO") balance the reactive power flow between their grid areas. No optimizer has detailed information about the neighbouring grid area and uses a very simple equivalent model for it. In case these equivalents are updated iteratively, we find that both optimizers mostly converge within only a few iterations for a realistic Danish grid topology. However, it is also found that the accuracy of the result highly depends on the built-in component models that each optimizer uses, within its own grid area.