Comparison of two coherency identification methods for frequency response assessment on a large-scale power system with fluctuating wind power generation

The injection of wind energy would have an impact on the local frequency stability. Therefore, the average system frequency cannot comprehensively reflect this impact due to the large inertia in the system. Coherency identification technology is a solution to this problem and two types of methods have been developed: model-based method and measurement-based method. The representative method of each type is slow coherency identification method and dynamic coherency determination method respectively. This paper compares these two methods for the coherency identification in a large power system with wind farm connected, considering the wind fam location and the number of wind farms. To demonstrate an effective comparation of two methods quantitatively, the paper proposes a frequency evaluation index: System Frequency Connection Strength (SFCS) to evaluate two methods based on measurement results. The comparison results are illustrated in the 16M5A system.