Spreading of Disturbances in Realistic Models of Transmission Grids: Dependence on Topology, Inertia and Heterogeneity

Abstract

The energy transition towards more renewable energy resources (RER) together with the reduction of conventional energy sources profoundly affects the frequency dynamics and electromechanical stability of electrical power networks. Here, we investigate the effect of different levels of grid inertia and its heterogeneous distribution on propagation patterns of disturbances in realistic models of power grids, including the Nigerian transmission network, the Ghana transmission network and the IEEE 118 bus test-grid as well as a Cayley-tree network, as a model grid without cycles, and the Square-grid power network as a strongly meshed model grid. These studies are conducted with the DigSILENT PowerFactory software and compared with results obtained from swing equations. Results show that the times of arrival of frequency deviations as function of geodesic distance from the contingency are more clustered and decay faster in more meshed grids than in tree- like or weakly meshed grids. In order to take into account the inhomogeneous distribution of inertia, we define an effective distance and find it to be more strongly correlated with the arrival time than the geodesic distance. A ballistic equation for the arrival time, with a velocity derived from the swing equation, provides then a strict lower bound for all effective distances. The fitted velocity decays with a power law with grid inertia. Decreasing the magnitude of grid inertia by increasing the magnitude of RER is thereby found to increase the travel velocity of disturbances and reduce the time a disturbance needs to spread throughout the grid. Published at https://www.nature.com/articles/s41598-021-02758-2