Quantitative Intelligent Evaluation of Transient Power Angle Stability based on Cluster Grouping

Aiming at the problem that it is difficult to achieve the quantitative evaluation of transient power angle stability based on artificial intelligence, a quantitative intelligent evaluation method of transient power angle stability based on cluster grouping is proposed. Firstly, the key features of transient power angle stability are selected by the method of comprehensive physical mechanism and data mining. According to the difference requirement parameters of the key features of the train samples, the similar samples are minimized and clustered into groups. The center sample of each group is determined based on the comprehensive minimum of the stable mode difference index. Based on each group of center samples, a transient power angle stability margin estimation model is established and parameters are optimized by linear programming. Determine the effective group to which the new mode belongs by the effective mode distance of each group of samples. Use the model corresponding to each effective group to estimate the margin of the new mode, and take the minimum value of the estimated margin of all effective groups as the estimated value of the transient stability margin of the new mode to realize quantitative intelligent evaluation of transient power angle stability. The effectiveness of the proposed method is verified by an example of power grid.