In order to make full use of historical monitoring data of mine hoist in the operation process, the information fusion technology is introduced into the fault diagnosis of the hoist to determine the fault reason and find the site of the fault position accurately, and a fault diagnosis method based on three layers information fusion is proposed. This method establishes the principal component model according to the PCA, carries on the feature extraction and realizes the data level fusion. It chooses the Elman neural network as the fusion algorithm, constantly adjusts the weight to train the characteristics extracted from various information sources, and completes the fusion process in feature layer. It uses DS evidence theory to fuse the information provided by the feature layer to obtain the diagnosis results on the decision layer. Finally, according to the principle of PCA fault diagnosis, the position of fault occurred is located and the diagnosis process is completed. This fusion method can be used to realize the integration of space and time through the reasonable selection, synthesis and utilization of the measured information. It has been applied to the mine hoist braking system, and the experimental results show that the method can make full use of a large amount of historical data to diagnose system and improve the reliability of the system.