Abstract:Purpose:The article proposes a fault diagnosis method for mining excavator engines, addressing the issues of limited datasets and low diagnostic accuracy. The method is based on a one-dimensional convolutional kernel, pooling kernel residual network, and transfer learning strategy. Method:A Random Forest (RF) classifier is used to perform dimension selection on the initial dataset, removing features with low importance to improve the model's learning efficiency and classification accuracy. The filtered 10-dimensional dataset is then used to pre-train a one-dimensional residual network (ResNet18_1D), and the training results are retained. Random noise is added to expand the dataset, and the parameters from the end of the one-dimensional residual network training are used as the initial parameters for transfer learning (TL). Five-fold cross-validation training is conducted using the expanded dataset, and the trained model parameters are saved and outputted. The best-performing model is called for testing, and the classification results are output. Result:The proposed RFTL-1DNet model was tested on a fault dataset from a mining excavator engine in Henan, and the experimental results show that this method significantly outperforms other methods in terms of fault diagnosis performance with over 99% diagnostic accuracy for mining excavator engine states. Conclusion: The model has a high classification accuracy for common engine faults and can quickly diagnose faults beyond the maintenance plan. Significance: It proposes a new method for intelligent mining equipment management.