Effects of Network Characteristics on Optimal Power Flow Convergence in Large Networks

This paper presents a novel method for analysing how key network characteristics affect the optimal power flow (OPF) convergence rates in large networks. As power systems become increasingly complex, there is a growing need for effective modelling of large-scale networks across voltage levels, including integrated transmission and distribution modelling. However, with challenges caused by differences in scale, components, and computational limits, the ability to create and analyse large-scale models requires attention. Existing modelling methods focus on overcoming limitations through new solvers or co-simulation processes. Thus, it is valuable to understand what aspects of a network make an OPF calculation simpler or harder to solve. The ability of the OPF solver to find a solution is utilised as a metric of network scalability. This is then assessed as various network properties are manipulated in order to evaluate which properties impact the OPF convergence rate as network scale increases. The described method is implemented on the IEEE300 test case and demonstrates the impact of maximum bus voltage constraints, generation capacity, and generator cost functions on OPF convergence.