Comparing the Topological and Electrical Structure of the North American Electric Power Infrastructure

The topological (graph) structure of complex networks often provides valuable information about the performance and vulnerability of the network. However, there are multiple ways to represent a given network as a graph. Electric power transmission and distribution networks have a topological structure that is straightforward to represent and analyze as a graph. However, simple graph models neglect the comprehensive connections between components that result from Ohm’s and Kirchhoff’s laws. This paper describes the structure of the three North American electric power interconnections, from the perspective of both topological and electrical connectivity. We compare the simple topology of these networks with that of random (Erdos and Renyi, 1959), preferential-attachment (Barabasi and Albert, 1999) and small-world (Watts and Strogatz, 1998) networks of equivalent sizes and find that power grids differ substantially from these abstract models in degree distribution, clustering, diameter and assortativity, and thus conclude that these topological forms may be misleading as models of power systems. To study the electrical connectivity of power systems, we propose a new method for representing electrical structure using electrical distances rather than geographic connections. Comparisons of these two representations of the North American power networks reveal notable differences between the electrical and topological structure of electric power networks.

💡 Research Summary

The paper investigates both the topological (graph‑theoretic) and electrical structures of the three major North American power interconnections – Eastern (EI), Western (WI) and Texas (TI). While power transmission and distribution networks can be represented as simple undirected, unweighted graphs, such representations ignore the physical laws (Ohm’s and Kirchhoff’s) that govern power flows. To address this gap, the authors first construct large‑scale topological models of the grids: the IEEE‑300 test case (300 nodes, 409 edges) and detailed models of EI (41,228 nodes, 52,075 edges), WI (11,432 nodes, 13,734 edges) and TI (4,513 nodes, 5,532 edges). All buses—generators, loads, and pass‑through nodes—are treated equally, and parallel lines are collapsed into single edges, yielding average degrees close to 2.5 for all three interconnections.

For comparison, synthetic networks of equivalent size and edge count are generated using three classic random‑graph families: Erdős‑Rényi (random), Watts‑Strogatz (small‑world) and Barabási‑Albert (preferential‑attachment/scale‑free). The authors then evaluate a suite of standard graph metrics: degree distribution, clustering coefficient (C), characteristic path length (L), network diameter (d_max), and degree assortativity (r).

Statistical testing (Kolmogorov‑Smirnov) shows that the degree distributions of the real power grids are best described by exponential (Weibull) tails rather than power‑law (scale‑free) behavior. The estimated power‑law exponent α is ≈3.5 for most cases, but the hypothesis that the grids follow a true power‑law is rejected at the 0.001 level for all but the IEEE‑300 vs. random graph comparison. Maximum node degree in the real grids never exceeds 30, indicating the absence of the ultra‑high‑degree hubs typical of scale‑free networks.

Clustering coefficients are extremely low (C≈0.000–0.008) for the interconnections, far below the values observed in small‑world networks (C≈0.1–0.3). Average path lengths and diameters are intermediate: larger than random graphs but far smaller than regular lattices, confirming that power grids are neither purely random nor regular. Assortativity values hover around zero, with slight negative tendencies in the synthetic graphs and slight positive tendencies in the real grids, suggesting only weak degree‑degree correlation.

Recognizing that topological metrics alone cannot capture the physics of power flow, the authors introduce an “electrical distance” based on the linearized DC power‑flow model. This distance matrix reflects the effective impedance between any pair of buses, incorporating both Ohm’s law and Kirchhoff’s current law. By constructing a graph where edge weights are derived from these electrical distances, the authors reveal a structure that can differ dramatically from the pure topological graph. For instance, two buses that are topologically adjacent may be electrically distant (high impedance), while geographically remote buses can be electrically close (low impedance) due to parallel paths and network redundancy.

The comparative analysis demonstrates that the three North American interconnections deviate substantially from the canonical random, small‑world, and scale‑free models in all examined metrics. Moreover, the electrical‑distance representation uncovers hidden connectivity patterns that are invisible to a purely topological view, implying that vulnerability assessments based solely on graph attacks may be misleading.

In conclusion, the study argues that power‑grid modeling must go beyond simple unweighted graphs. Incorporating electrical distances yields a more faithful representation of the network’s functional connectivity, which is essential for accurate reliability analysis, cascade‑failure modeling, and control‑strategy design. The authors suggest that future work should develop hybrid models that blend topological and electrical information, and apply them to tasks such as optimal islanding, resilience planning, and real‑time operation of large‑scale power systems.