EMD (empirical mode decomposition) based computation method for recognizing sensitivity of signal source

Abstract

The invention discloses an EMD (empirical mode decomposition) based computation method for recognizing sensitivity of a signal source. The computation method comprises main steps as follows: vibration signals under the normal oil supply parameter and under the adjusted oil supply parameter are set, and a correlation coefficient between the vibration signal under the adjusted oil supply parameter and an IMF (intrinsic mode function) component of the vibration signal is computed; a correlation coefficient between the IMF component and the vibration signal under the normal oil supply parameter is computed; a correlation coefficient difference value generated through oil supply parameter change is computed; the correlation coefficient difference value is subjected to standard processing according to the function, and a sensitivity factor is defined; a sensitive IMF component is selected according to the factor; the sensitive IMF component is subjected to wavelet time-frequency analysis, so that a diesel engine vibration source corresponding to the component is determined. With the adoption of the method, the diesel engine vibration source closely related with the oil supply parameter change is revealed, the recognition precision of the vibration source closely related with the diesel engine oil supply parameter change can be improved, and reasonable selection of the diesel engine oil supply parameter and vibration noise control are realized.

Description

technical field [0001] The invention belongs to the calculation technology of relevant parameters of an internal combustion engine, and in particular relates to a calculation method for reasonably selecting fuel supply parameters of a diesel engine and controlling vibration and noise. Background technique [0002] Although most of the vibration noise is generated by the characteristics of the internal combustion engine's own structure, due to the difference in ignition methods, the combustion noise of a diesel engine is louder than that of a gasoline engine. With the wide application of diesel engines, research and manufacturers have tried to adopt various methods and technologies to reduce combustion noise while taking into account diesel engine emissions and fuel economy, and achieved good results. For example, optimize the design of the fuel injection system, or adjust the relevant parameters in the fuel supply process to improve the combustion process and reduce noise an...

AI Technical Summary

Problems solved by technology

However, the time domain signal is limited to the overall evaluation of vibration and noise, and cannot measure the main frequency components in the signal and the changes of each excitation source

Frequency domain analysis is to decompose the vibration signal of the diesel engine into its frequency components, and compare the signal characteristics under different system parameters according to the shape characteristics of the power spectrum to determine the vibration status of the diesel engine. Vibration signals produce similar power spectra, it is difficult to judge them

Images