Abstract:With the large-scale grid integration of new energy sources, the proportion of synchronous generators in the power system is decreasing, resulting in lower system inertia and higher time variability. Therefore, there is an urgent need for a fast and accurate system inertia assessment method. Usually, the power system inertia can be roughly estimated by the synchronous generator equivalent model method, but this method does not consider the influence of the load model enough and has a large error. Besides, most of the existing inertia estimation methods rely on the system measurement data after disturbance, which is difficult to reflect the real situation of the current system inertia in the context of the increasingly variable operation mode. Therefore, a fast estimation method of system inertia based on extreme learning machine (ELM) is proposed. On the basis of constructing the equivalent load model of the system, the method uses the ELM algorithm to evaluate the influence of system load on system inertia, so as to improve the estimation accuracy of system inertia only considering the influence of synchronous generator, and realize the fast estimation of system inertia. The fast estimation of system inertia is realized. Based on this, the proposed method is tested with several standard systems, and the results show that the method can quickly and accurately estimate the actual inertia of the system.