Load distribution and operating point estimation in inverter-dominated grids by electrical signature analysis

High-frequency oscillatory excitations are common undesired by-product of inverter-coupled devices in alternating cur-rent (AC) and direct current (DC) grids. This paper introduces a novel concept that utilizes the characteristic spectra, defined as the device’s signature obtained from the oscillatory excitations, to estimate the operating point of devices within an electrical grid. The uniqueness of this concept lies in its ability to approximate operating points for multiple devices with a single measurement point in the grid. This is achieved by calculating the spectral decomposition of the grid measurement and solving an optimization problem for a superposition of known device signatures to minimize the deviation from the spectral grid components. The findings of this study demonstrate that by leveraging the undesired effects of modern power electronics, information about devices in the grid can be obtained that might otherwise be inac-cessible. This is particularly important for grids where additional transducers cannot be used or the grid cannot be altered for measurements, yet detailed knowledge of device operating points are required. The fact that operating points can be estimated through voltage readings as well, opens numerous possibilities for analyzing electrical grids.