Abstract: In this paper,based on the Grey System Theory,the principle and method of Grey Clustering Analysis is introduced,and is also used to classify the flood disaster degree of Xinjiang＂96.7＂floods.In the application of Grey Clustering Analysis,combining the disaster losses of every regions in Xinjiang＂96.7＂floods,disaster areas are taken as the clustering objects（ i=I,II,III ⋯ X）,some essential factors of disaster losses as the clustering quotas（j=1^*,2^*,⋯4^*）.Divide the floods disaster into five grades（k=1,2,⋯5）according to the level of economy development and population density of Xinjiang.The clustering coefficient and disaster grade of the disaster regions in Xinjiang＂ 96.7 ＂floods were made out by using this method.The results show that the extreme heavy disaster includs Changji,Tulufan,Bazhou and Bintuan;the heavy disaster includs Urumqi,Tacheng and the middle disaster includs Bozhou,Hami,Kerzhou and Kashger;which are identical to the flood disaster losses of Xinjiang＂96.7＂floods.All mentioned above indicates that the Grey Clustering Analysis is an effective method to deal with the classification of disaster grade.

Key words: Flood Disaster, Grey Clustering, Classification of Grade