Font Size: a A A

Fault Feature Extraction Of Planetary Gearbox Based On Vibration Signal Separation

Posted on:2020-05-05Degree:MasterType:Thesis
Country:ChinaCandidate:Y LongFull Text:PDF
GTID:2392330599455705Subject:Mechanical design and theory
Abstract/Summary:PDF Full Text Request
Planetary gearboxes are widely used in large mechanical transmission systems due to its unique structure,small volume,large transmission ratio and strong carrying capacity.However,planetary gearboxes usually operate in harsh environment,which causes critical components such as gears or bearings to be more susceptible to failure.The transmission path of vibration signals in planetary gearbox is time-varying because of the planetary motion of the planet gear,so the rotation of the planet carrier has amplitude modulation effect on vibration signals,and the multiple gear pair meshing vibration and the bearing vibration interact with each other.Therefore,faulty information of planetary gearboxes is weak,which makes the traditional fault feature extraction method of fixed-shaft gearboxes can not apply to planetary gearboxes directly.Thus,how to extract fault features of planetary gearboxes effectively is of great significance.The windowed vibration separation technology can restrain the influence of time-varying transmission path of vibration signals of planetary gearboxes.Based on the existing research,an improved windowed vibration separation technology is proposed in this paper.Firstly,the peak value is utilized to select the initial position of windowed interception,and the multi-tooth-width Tukey is used for windowed interception.Then,the multi-tooth-width vibration signals intercepted by windowed interception are divided into single-tooth-width vibration signals and re-spliced into multi-group vibration separation signals according to the meshing tooth sequence.Finally,a group of vibration separation signals with the richest fault information is determined according to the kurtosis to gain a higher signal-to-noise ratio.Aiming at the fault feature extraction of planet gears or sun gears,a fault feature extraction method based on improved windowed vibration separation and variational mode decomposition is proposed in this paper.Firstly,the synchronous periodic components of interest in planetary gearbox are extracted by synchronous averaging technology to improve the signal-to-noise ratio.Then,the vibration separation signal of a single planet gear or a sun gear is constructed by improving the windowed vibration separation technology to suppress the influence of the time-varying transmission path of the vibration signal of the planetary gearbox.Then,the vibration signal similar to a fixed-axis gearbox is obtained.Finally,the decomposition method adaptively chooses the mode which contain abundant fault information for envelope demodulation analysis,extracts fault features,and overcomes the shortcomings of filtering and demodulation analysis by artificial selection of parameters.Fault information of planet bearings is often submerged in gear meshing vibration leading to be weak.In order to solve this problem that the fault feature of planet bearing is difficult to extract,a fault feature extraction method of the planet bearing based on vibration signal separation is proposed in this paper.The improved windowed vibration separation technology is used to suppress the modulation effect of planetary carriage rotation,then,discrete random separation technology is utilized to separate the vibration signals of planet bearings from the vibration signals of gear meshing to extract the fault characteristics of planet bearings.In this study,2K-H single-stage planetary gearbox in NGW planetary gear box comprehensive experimental platform is selected as the research object.The effectiveness of the proposed method is verified by the simulation fault analysis of a planet gear,a sun gear and a planet bearing.
Keywords/Search Tags:Planetary gearboxes, Fault feature extraction, Improved windowed vibration separation, Variational mode decomposition, Discrete random separation
PDF Full Text Request
Related items