Sparse Representation of Diesel Engine Multi-source Impact Signals and Its Application to Fault Diagnosis
Diesel engines are widely used in different fields such as ships,nuclear power plants and vehicles,and their monitoring and fault diagnosis is of great significance.With the development of equipment health monitoring,the pressure of data storage increases significantly,and sparse signal representation has become an effective solution.In this paper,a sparse representation method for diesel engine multi-source impact signals is proposed based on decomposition signal(DS)dictionary.And the sparse coefficient is applied as a feature to diagnose abnormal faults in diesel engine valve clearances.Firstly,the variational time-domain decomposition(VTDD)is used to process the signals for acquiring the decomposition signals.Then,the decomposed signals are integrated to form the DS dictionary.And the orthogonal matching pursuit(OMP)algorithm is used to realize the sparse representation of the original signal and the decomposed impact signals.Finally,the sparse coefficients are used as features for diagnosing abnormal faults in diesel engine valve clearances.Test results show that the proposed method has a good application effect with a fault diagnosis accuracy above 90%,indicating its effectiveness in solving the challenges of diesel engine monitoring and fault diagnosis.
fault diagnosisdiesel enginevibration and impactsignal processingsparse representation