Maximizing the algebraic connectivity of meshed electrical pathways used as current return network

Etienne Goddet (1, 2) , Nicolas Retiere (1, 3) , Vojislav Stojanovic (2) , Anca Dieudonne (2) , Jerome Genoulaz , Jean-Michel Guichon (1, 4)

1 G2ELab - Laboratoire de Génie Electrique de Grenoble
2 SAFRAN Electrical & Power Div EWISe
3 G2Elab-SYREL - G2Elab-SYstèmes et Réseaux ELectriques
4 G2Elab-MAGE - G2Elab-Modèles, Méthodes et Méthodologies Appliqués au Génie Electrique

This paper proposes an innovative method for optimizing the connectivity of meshed electrical pathways used to carry return current and provide grounding in more composite aircrafts. In normal operation, these networks should ensure a low voltage drop between power sources and electrical loads. The networks are modelled as graphs and spectral graph analysis is used to study their complexity. Thanks to the insight into graph's spectrum, a multi-scale reading of meshed networks topology is proposed and an optimization problem is defined to decrease the DC resistance of the network while keeping constant its total weight. It is based on the maximization of the second eigenvalue of the graph's Laplacian matrix. The optimization problem is applied to a mock-up of a real aircraft current return network. The paper is concluded by checking the DC voltage drops in steady state conditions. c

Mots clés: Graph theory; Spectral analysis; Optimization; Algebraic connectivity; Current return networks

Domaines: Optimisation et contrôle [math.OC]; Energie électrique

Etienne Goddet, Nicolas Retiere, Vojislav Stojanovic, Anca Dieudonne, Jerome Genoulaz, et al.. Maximizing the algebraic connectivity of meshed electrical pathways used as current return network. Mathematics and Computers in Simulation, 2019, 158, pp.18--31. ⟨10.1016/j.matcom.2018.05.002⟩. ⟨hal-02350910⟩