Influence of different damage fragment identification methods on the effect of accelerated editing of stress loads on combine harvester chassis
GU Yuanyang
YANG Zihan
LIU Xuke
ZHANG Zhihao
GUO Zhiqiang
SONG Zhenghe
Abstract：[Objective]The operating conditions of combine harvesters are complex and variable.No load spectrum accelerated editing method adapted to their field load characteristics is available,which restricts the improvement of structural reliability design and durability test efficiency.Three different damage fragment identification methods were applied to the accelerated editing of chassis stress loads of combine harvesters.To solve the aforementioned problems,the time-domain damage retention identification method,the wavelet transform-based envelope damage identification method and the time-frequency analysis-based cumulative power spectral density damage identification method were employed.The strain signals measured at the hinge positions of the combine harvester rear axle were separately subjected to accelerated spectrum compilation processing by the three aforementioned methods,and the influences of different damage identification approaches on the accelerated compilation performance of stress load spectra for the combine harvester chassis were explored.[Methods]Firstly,vertical stress loads at the hinge point of the rear axle of the chassis of a domestic large double axial flow combine harvester were collected.Preprocessing and working condition division were completed to clarify the damage contribution ratio of each working condition.Secondly,the time-domain damage identification method was adopted.The optimal 2 s calculation window interval was determined to realize the screening of load fragments based on damage distribution.Then,high-frequency components were extracted based on wavelet transform.Envelope analysis was performed to identify high-damage load fragments.Finally,the accumulative power spectral density damage identification method based on time-frequency analysis was adopted.High-energy damage regions were located through continuous wavelet transform and power spectrum superposition.[Results]The results show that under the condition of pseudo-damage retention ratio of about 98％,the signal compression ratio of the accumulative power spectral density damage identification method based on time-frequency analysis reaches 40.92％.The acceleration effect is improved by 12.45％and 16.56％respectively compared with the time-domain damage identification method and the envelope damage identification method based on wavelet transform.The grain unloading condition with a damage ratio of only 0.7％can be completely eliminated,but deficiencies exist in the consistency of power spectral density in the high frequency band.Reference can be provided for the accelerated editing of load spectrum in the field of agricultural machinery.
Keywords：Combine harvesterLoad spectrumAccelerated editingWavelet transformDamage fragment identification
Publication Date：2026-07-31
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：9( 40-48 )