An Accelerated Editing Method for Stress Signal on Combine Harvester Chassis Using Wavelet Transform

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 13; p. 4100
• Main Authors: Huang, Shengcao, Yang, Zihan, Song, Zhenghe, Yu, Zhiwei, Guo, Xiaobo, Chen, Du
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 30.06.2025; MDPI
• Subjects: acceleration editing; Agricultural production; Case studies; combine harvester; Editing; Efficiency; Farm machinery; Field study; Fourier transforms; Harvest; load spectrum; Machinery; Magneto-electric machines; Methods; Optimization; Product development; Strain gauges; wavelet transform; Wavelet transforms; Working conditions; combine harvester; acceleration editing; load spectrum; wavelet transform
• DOI: 10.3390/s25134100
• ISSN: 1424-8220

This paper presents a load spectrum acceleration editing method based on wavelet transform. The principle of the method is to decompose the target signal using wavelet transform to obtain high-frequency wavelet components, which are classified and combined based on their frequency components for accelerated editing. During the damage segment identification stage, a threshold selection method based on the pseudo-damage gradient of the segment identification results is proposed. An envelope-based damage identification method is used to extract high-damage segments from the original signal, which are then concatenated to form an accelerated signal. Using the stress signal on the chassis of a combine harvester as a case study, the effectiveness of various accelerated editing methods is compared, with a discussion on the selection of wavelet function parameters. The results indicate that, compared to the time-domain damage retention method and the traditional wavelet transform accelerated editing method, the proposed improvement enhances the acceleration effect of the time-domain signal by 7.76% and 15.92%, respectively. The accelerated signal is consistent with the original signal in terms of statistical parameters and power spectral density. Additionally, we also found that an appropriate selection of the wavelet function’s vanishing moment can further reduce the time-domain signal length of the accelerated result by 4.8%. This study can provide beneficial experiential references for load spectrum development in the accelerated durability testing of agricultural machinery.