Realizing the similarity identification between future and historical transmission sections can provide guidance for the formulation of future transmission section control measures

which is of great significance for ensuring the safe and stable operation of the power system. However

the existing methods fail to fully consider the selection of key sectional features and sample similarity measurement methods. Therefore

this paper propose an improved fuzzy C-means (FCM) clustering algorithm embedded with multi-scale key feature information for sections similarity identification. First

in order to obtain more sectional feature samples

this paper establishes a hierarchical decision model to filter and improve the basic feature quantities of the power grid. Then

an optimized total transfer capability (TTC) calculation model considering thermal stability and transient power angle stability constraints is proposed to obtain the TTC. Subsequently

an adaptive FCM clustering algorithm considers both sample morphology and numerical features is proposed. With the aim of improving the effectiveness and objectivity of the algorithm

the optimal number of clusters is adaptively selected by the multi-indicator evaluation method

and the cosine distance is introduced to consider the sample morphological features. Finally

the effectiveness of the proposed method is verified by taking the IEEE 39-bus system and a local actual power grid as examples.