Estimating the inertia in a (micro)grid based on small power steps

The inertia in a power grid is an important indicator for its robustness, in particular, its ability to recover from faults that would cause the frequency to deviate significantly from the nominal frequency, possibly leading to loss of synchronism. Hence, the increased penetration of renewable sources with little inertia has brought with it serious concerns about grid stability and robustness. Transmission system operators in developed countries, as well as in private companies, are developing and implementing tools to estimate the inertia in the system, and to mitigate the effects of increased penetration of renewable sources on the inertia. We propose a method for the real-time estimation of the total inertia, as well as equivalent local inertia constants (to be defined). The method involves injecting small steps of power (both positive and negative) from energy storage units located in different areas of the grid and measuring the step response of the frequency in the same points. The steps are small relative to the capacity of the power grid, but by averaging over several measurements, the effect of the noise can be reduced. Processing the step responses measured in different points allows obtaining inertia estimates. We give a theoretical motivation for our method and demonstrate it experimentally in a microgrid consisting of four areas, each containing a virtual synchronous machine.