Topology identification for multiple-bus DC MicroGrids via primary control perturbations

We propose a Least Squares Estimation procedure for estimating the admittance matrix of multi-bus DC MicroGrids (MGs). In the proposed solution, the generators simultaneously inject training signals in the form of small deterministic perturbations of the primary droop control parameters and measure the related steady state deviations of the bus voltage. When the training signals meet sufficient excitation conditions seen in the number of injected perturbations, the admittance matrix can be uniquely identified. The measurements are collected and processed by a topology identification and monitoring system that produces unbiased estimate of the admittance matrix. The numerical evaluations show that the estimator can recover the admittance matrix of the MG with very high precision, proving its practical viability.