Identification of inertia constants using time-domain vector fitting

The increasing penetration of converter-interfaced renewable energy sources in the quest to decarbonize power grids has introduced significant challenges considering the safe and reliable operation of modern power systems. Of critical threat are those posed by the reduced system inertia. As a result of these concerns, system operators must monitor and determine close to real-time their system inertia in order to provide proper preventive remedial actions. This is mostly feasible by using measured signals. In this paper, a new measurement-oriented method, based on the time-domain vector fitting technique, is developed for the estimation of inertia constants of synchronous generation units from normal operating condition data and transient responses. The impact of various parameters on the efficiency of the proposed method is investigated by means of Monte Carlo analysis in a simplified system frequency response model. Simulations, conducted in two benchmark power system models, namely the IEEE 9-Bus and the IEEE 39-Bus Test Systems, are used to evaluate the performance of the proposed method. Comparisons with respect to the widely adopted ARMAX-based method are also performed. The applicability of the proposed method is also demonstrated using laboratory measurements.