Topology Reconstruction of a Class of Electrical Networks with Limited Boundary Measurements

Abstract

We consider the problem of recovering the topology and the edge conductance value, as well as characterizing a set of electrical networks that satisfy the limitedly available Thevenin impedance measurements. The measurements are obtained from an unknown electrical network, which is assumed to belong to a class of circular planar passive electrical network. This class of electrical networks consists of R, RL, and RC networks whose edge impedance values are equal, and the absolute value of the real and the imaginary part of the edge impedances are also equal. To solve the topology reconstruction and the set characterization problem, we establish a simple relation between Thevenin impedance and the Laplacian matrix and leverage this relation to get a system of multivariate polynomial equations, whose solution is a set of all electrical networks satisfying the limited available Thevenin's impedance measurements. To confine the search space and generate valid electrical networks, we impose the triangle and Kalmanson's inequality as constraints. The solution to a constrained system of multivariate polynomial equations is a set of reconstructed valid electrical networks. For simple algorithmic solutions, we use Gröbner basis polynomials. This paper shows that the triangle and the Kalmanson's inequality holds for general circular planar passive R, RL, and RC electrical networks if certain boundary conditions lie within a convex cone. Numerical examples illustrate the developed topology reconstruction method.