Authors

Abstract

The feed rate is an important index for evaluating the performance of a combine harvester. Determining how to accurately reflect the feed rate during harvesting and establishing a reliable detection model is a major focus of current research and an important basis for the next step of feed rate stable control. This paper provides an overview of feed rate detection methods and stable control techniques for combine harvesters. It reviews methods that estimate the feed rate based on the inclined conveyor extrusion pressure, header power, and threshing unit energy consumption. Additionally, it introduces machine learning-based methods that incorporate multiple influencing factors to predict the feed rate. A comparison of different noise reduction techniques used in feed rate detection is also presented, analyzing their effectiveness. Furthermore, this study examines feed rate control methods in combine harvesters, discussing various control approaches with an emphasis on methods that stabilize the feed rate by adjusting header height and harvester forward speed. In response to the current issues of inadequate detection accuracy in feed rate monitoring, limited adaptability, and instability in control systems, it is pointed out that future research needs to innovate in developing advanced sensor technology, optimizing automatic control algorithms as well as data fusion and analytical methodologies.

Abstract in Chinese

喂入量是衡量联合收割机性能的重要指标，通过何种方式将收获时机的喂入量反应出来并建立可靠的检测模型是当前研究的重点，也是下一步稳定控制喂入量的重要依据。该文概述了联合收割机喂入量检测方法以及喂入量稳定控制方法的研究现状，综述了基于过桥挤压力、割台功率、脱粒元件功耗等反应收割机喂入量的方法，介绍了利用机器学习结合多种影响喂入量的因素预测喂入量的方法，并对比了喂入量检测中不同降噪方法的降噪效果的优劣，接着对联合收割机喂入量控制方法进行了梳理，探讨了不同控制策略，着重分析了通过调节喂入量高度、收割机前进速度使喂入量保持稳定的方法。结果表明现阶段喂入量检测方法均能在一定程度上反映出喂入量变化情况，但普遍存在稳定性差、检测精度不高等问题。通过控制喂入量有效减少了收获损失率，但控制误差易受田间环境影响、波动大。针对上述问题，指出今后研究需要在发展先进传感器技术、优化自动控制算法、数据融合与分析等方面开拓创新。