Abstract: In recent years, mine water damage accidents have become more frequent. In order to reduce the harm caused by mine water inrush accident to coal mine production and human life and property safety, how to judge mine water Inrush source quickly and accurately is more and more important. Aiming at the above problems, a rapid identification method of mine water inrush source coupled with PCA-ELM model is proposed, and validates the method with the water source data of Zhaogezhuang mine as an example. Firstly, the principal component analysis method is used to reduce the dimension of the multi-dimensional variable data, and then the ELM algorithm is used to simulate the selected principal component. The results show that the method has stable classification performance and short classification learning time, which satisfies the condition of rapid identification of water inrush source.