A method of transient stability assessment of power system

Abstract

The invention relates to a power system transient stability evaluation method, which belongs to the technical field of power system stability analysis. Firstly, the transient stability simulation is carried out for different operating conditions and preset faults of the power system, and the data of the power system before the fault is collected. Then, the similarity evaluation index of different preset faults is constructed based on the Jaccard distance and the Hausdorff distance, and the data sets of different preset faults are clustered based on the similarity evaluation index; the different preset faults in each cluster are clustered. The multi-task Siamese neural network with shared weights is sequentially trained to obtain multiple multi-task Siamese neural networks for transient stability evaluation. This method considers the similarity of different preset faults in the transient stability assessment, and uses different data sets of similar preset faults to train the twin network, which is beneficial to improve the generalization ability of the transient stability assessment model, thereby improving the transient stability assessment results. accuracy.

Description

technical field [0001] The invention belongs to the technical field of power system stability analysis, and relates to a power system transient stability evaluation method. Background technique [0002] Transient stability damage is an important cause of large-scale power outages in power systems. How to quickly and accurately judge the transient stability of power systems is one of the important issues to be considered in power system security prevention and control. In recent years, data-driven methods such as support vector machines and extreme learning machines have been used to analyze the transient stability of power systems under preset faults. In general, since the fault has not yet occurred, steady-state data is often used as the input feature. Since the transient stability of the power system under different preset faults in the same operating mode is different, the data sets under different preset faults are generally used to construct multiple machine learning m...

AI Technical Summary

Problems solved by technology

If different transient stability evaluation models are trained separately for similar data sets, similar data sets cannot be fully utilized, which is not conducive to the improvement of model performance under limited data sets.

Images