Construction method and case study of digital twin system for combine harvester

文献类型： 外文期刊

第一作者： Yin, Yanxin

作者： Yin, Yanxin;Ma, Bowen;Meng, Zhijun;Chen, Liping;Yin, Yanxin;Meng, Zhijun;Chen, Liping;Liu, Mengnan;Ma, Bowen;Zhang, Yawei;Zhang, Bin;Wen, Changkai

作者机构：

关键词： Digital twin; Combine harvester; Kinematic relationship modeling; Fuel consumption prediction; LightGBM

期刊名称：COMPUTERS AND ELECTRONICS IN AGRICULTURE （ 影响因子：8.9； 五年影响因子：9.3 ）

ISSN： 0168-1699

年卷期： 2024 年 226 卷

页码：

收录情况： SCI

摘要： How to further improve the working performance of combine harvester is increasingly focused. Because the harvest time of crops in agricultural production is generally short, the number of field trials of harvesters is limited, which severely hindered the study of combine harvesters. The digital twin system can conduct a large number of simulation tests on the harvester in a virtual environment, and it is not limited by operation time and operation scenarios, which has obvious advantages. Addressing the issue that current digital twin systems for agricultural machinery rely heavily on large-scale physical engines and the combine harvester digital twin system lacks a method for modeling multi-component complex transmissions; this study introduces a lightweight network-based approach to construct a digital twin system for combine harvester, encompassing multiple subsystems such as physical, virtual, model calculation, data interaction, and human-computer interaction. Among them, it studies the complex transmission and motion pattern classification of critical components of combine harvester, proposing a method for modeling the kinematic relationships of complex motions in critical components, accurately modeling different types of physical activities provides crucial technical support for the precise mapping of digital twin systems. Ultimately, using the Lovol GM100 combine harvester as a case study and leveraging the CMOnlineLib and HTML lightweight network, a digital twin system was developed for the combine harvester, including creating a LightGBM model capable of predicting fuel consumption. Field tests demonstrate that the digital twin system for the combine harvester operates stably and reliably, with the fuel consumption prediction model under full-load conditions achieving an average error of 0.24 L/h, a maximum error of 0.84 L/h, and an average relative error of only 1.09 %. This research offers a novel approach to enhancing the digital twin technology and increasing the intelligence level of combine harvester.

分类号：

  • 相关文献
作者其他论文 更多>>

微信小程序