Considering the abnormal fuel injection failure of small aviation piston engines

based on the engine’s in-cylinder pressure and cylinder head vibration signal

a fault diagnosis method combining variational modal decomposition and cuckoo search optimization support vector machine was used to diagnose abnormal fuel injection failure of the engine. The method used variational mode decomposition to process the in-cylinder pressure signal and cylinder head vibration signal of the engine to obtain an intrinsic mode function

perform singular value decomposition and energy feature extraction on the intrinsic mode function

and the data sets of in-cylinder pressure and cylinder head vibration were input into the support vector machine optimized by the cuckoo search algorithm for training and testing. Results showed that this method can better identify the faults of abnormal fuel injection of the engine

and the accuracy of the fault recognition classification of the pressure in the cylinder and the vibration signal of the cylinder head was 95.32% and 92.47%

respectively

verifying the effectiveness of the method.