Abstract:Aiming at the problems that mining truck engine have poor fault diagnosis accuracy in the case of small samples, a mining truck engine intelligent failure diagnosis method based on the improved sparrow search algorithm optimizing convex radius-margin-based SVM model (F-SVM) was proposed. Firstly, a chain search strategy was introduced to solve the problem of insufficient global search ability in the sparrow search algorithm. Secondly, following the principle of optimal location, the location update of the joiners was optimized to improve its convergence performance. Finally, the improved sparrow algorithm was adopted to optimize the kernel parameter g and penalty factor C of F-SVM, and then a fault diagnosis model of mining truck engine was constructed. The experimental results prove that the prediction accuracy of the proposed method is higher, which is 21.5% and 4.1% higher than the traditional SVM and F-SVM models, respectively. The model can better realize the diagnosis of common faults of mining truck engines. It is suitable for fault prediction of small sample data, and provides a reference for intelligent fault diagnosis of mining machinery and equipment.