| As the heart component of large machinery,diesel engine involves vehicles,boats,locomotives,electric power and other fields.Whether the diesel engine is running smoo thly has a direct impact on the reliability and safety of the entire power system.As the core component of the diesel engine,the piston connecting rod assembly directly affects the operation of the diesel engine.In the daily work of a diesel engine,the piston and connecting rod assembly is an important structure for converting heat energy into mechanical energy.It is prone to wear and other failures in harsh working environments.The friction between the mating pairs can cause vibration.This vibration information is actually a manifestation of the operating status of the diesel engine.Under the condition that the diesel engine is not disintegrated,the operation status of the diesel engine can be monitored by analyzing the vibration information during the operation of the diesel engine and extracting the fault characteristics.Through the analysis and comparison of various adaptive decomposition methods,this paper determines that the Variational Modal Decomposition(VMD)is the main analysis method;the method of combining VMD and independent component analysis(ICA)is used to reduce the noise of the vibration signal,and blind source separation is carried out for the concurrent faults of the diesel engine.Through research,the following conclusions are drawn:(1)Through comparative study of multiple adaptive decomposition methods such as Empirical Mode Decomposition,Ensemble Empirical Mode Decomposition,Complete Ensemble Empirical Mode Decomposition,Local Mean Decomposition,Intrinsic Time scale Decomposition,and Variational Modal Mecomposition,etc.,Research shows that VMD has better resolution capabilities than other methods,and can better suppress modal aliasing and end effects.It is concluded that VMD is the most suitable adaptive decomposition method for analyzing diesel engine vibration signals.(2)VMD and IC A are used to reduce the noise of diesel engine vibration signal,and the fault component and noise component are screened out accurately.The noise in the vibration signal is effectively removed.(3)Based on VMD and blind source separation,the blind source separation of the single-channel observation signal is realized,and the source signal component is successful y separated from the mixed signal.(4)Weichai WD615 diesel engine is taken as the research object.The faults of piston,piston pin and connecting rod bearing and the concurrent faults of piston and piston pin were set respectively.The vibration signals of diesel engine body were measured,and the best measuring point and the best test speed were determined.The vibration signal of piston is decomposed by VMD.The energy values of the piston under different wear cond itions were extracted as characteristic parameters to realize the recognition of piston wear state.VMD-ICA is used to accurately screen out the fault components and noise components,and the noise in the vibration signal of connecting rod bearing is removed.By using the method of VMD and Blind Source Separation,the fault of piston and piston pin of diesel engine was diagnosed successfully. |