Combines fuzzy logic inference system with Neural Network, a fault diagnosis method for diesel engine based on Adaptive Network-based Fuzzy Inference System (ANFIS) is studied. The theory and arithmetic and structure of this system are introduced, and a fault model for the fuel injection system of a diesel engine is built with the help of Matlab Fuzzy Toolbox. Simulation results demonstrate that the proposed approach can efficiently diagnose the faults and is better than the fault diagnosis method based on BP neural network.