Abstract
The use of diesel cylinder head vibration signal for fault diagnosis must eliminate signal interference non-periodic and random components. The periodic components related to the working cycle of the diesel engine are retained, so as to achieve the purpose of fault feature extraction. A time-frequency correlation-based diesel engine fault feature extraction method is proposed in this paper. First, the Wavelet transform is applied to the vibration signals of diesel engines collected in three continuous working cycles to realize the time-frequency distribution of the signals. Then three time-frequency distributions are estimated by cross-correlation, so as to eliminate noise interference and extract periodic transient characteristics. The experimental results of simulation signals and real signals show that this method can effectively extract the periodic transient impact characteristics of diesel engines.
Highlights
The quality of diesel combustion state is directly related to power, economy and other performance indicators, and it is of great significance to detect its state in time and effectively
Wavelet transform in non-stationary and time-varying signal has obvious advantages, and can automatically adjust the time window and frequency window size according to the characteristics of the signal to achieve frequency analysis signal, which has been widely applied in the field of fault diagnosis [3,4,5]
The feature extraction method of diesel engine combustion state based on time-frequency correlation transforms three continuous working cycle signals into two-dimensional time-frequency distribution through continuous wavelet transform
Summary
The quality of diesel combustion state is directly related to power, economy and other performance indicators, and it is of great significance to detect its state in time and effectively. It is necessary to adopt appropriate signal processing methods in order to effectively extract combustion characteristics of diesel engines and ensure the accuracy of detection and diagnosis [1]. Wavelet transform in non-stationary and time-varying signal has obvious advantages, and can automatically adjust the time window and frequency window size according to the characteristics of the signal to achieve frequency analysis signal, which has been widely applied in the field of fault diagnosis [3,4,5]. The Wavelet transform is applied to the vibration signals of diesel engines collected in three continuous working cycles to realize the time-frequency distribution of the signals. The time-frequency correlation estimation of two working cycle signals based on wavelet transform can analyze the distribution of correlation in time-frequency domain.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
