基于PSO-BP的甘蔗联合收割机物流堵塞预测预警研究

Prediction and early warning research on logistics blockage of sugarcane combine based on PSO-BP

在甘蔗的倒伏、弯曲,以及蔗节、蔗叶、泥块等杂物的共同作用之下,甘蔗联合收割机物流通道中的刀盘和切段辊处容易形成堵塞,为提高甘蔗联合收割机工作效率,降低甘蔗联合收割机物流通道发生堵塞的概率。以甘蔗联合收割机刀盘转速、切段辊转速、刀盘工作压力、切段辊工作压力等信号为输入变量,以甘蔗联合收割机运行状态为结果建立BP预测模型,并采用PSO-BP算法对预测模型进行优化。结果表明:PSO-BP神经网络预测模型对甘蔗联合收割机堵塞预测的准确率为98.75%,运行时间为0.18 s,预测准确率比BP神经网络提升了2.5%,用时减少了0.94 s。基于PSO-BP预测,应用LabVIWE软件开发了甘蔗联合收割机物流堵塞预测及预警系统,实现对甘蔗联合收割机刀盘和切段辊液压参数的实时监测和物流状态的预测和预警。研究结果可为优化甘蔗联合收割机的防堵控制设计提供依据。

Under the common action of cane falling,bending,cane knot,cane leaves,mud and other debris,the cutter head and cutting roll in the logistics channel of the sugarcane combine harvester are easily blocked.In order to improve the working efficiency of the sugarcane combine harvester and reduce the probability of the sugarcane combine harvester logistics channel clogging.The BP prediction model was established with input variables such as cutter head speed,cutter roll speed,cutter head working pressure and cutter roll working pressure of sugarcane combine harvester,and PSO-BP algorithm was used to optimize the prediction model.The results show that the accuracy of PSO-BP neural network prediction model is 98.75%,the running time is 0.18 s,the prediction accuracy is 2.5%higher than BP neural network,and the time is reduced by 0.94 s.Based on PSO-BP prediction,LabVIWE software was used to develop the logistics congestion prediction and early warning system of sugarcane combine harvester,which realized the real-time monitoring of hydraulic parameters of cutter head and section roll of sugarcane combine harvester and the prediction and early warning of logistics status.The results can provide a basis for optimizing the anti-blocking control design of sugarcane combine harvester.