基于BP神经网络的甘蔗收获机切割器振动性能研究  

Research on Vibration Performance of Sugarcane Harvester Cutter Based on BP Neural Network

作　　者：钟家勤[1] 李尚平 何永玲[1] 何维[1] 王跃飞[1] Zhong Jiaqin;Li Shangping;He Yongling;He Wei;Wang Yuefei(Guangxi Colleges and Universities Key Laboratory Breeding Base of Coastal Mechanical Equipment Design,Manufacturing and Control, Qinzhou Unversity,Qinzhou 535000,China;College of Chemistry and Chemical Engineering, Guangxi University For Nationalities,Nanning 530004,China)

机构地区：[1]钦州学院广西高校临海机械装备设计制造及控制重点实验室培育基地,广西钦州535000 [2]广西民族大学化学化工学院,南宁535004

出　　处：《农机化研究》2019年第3期193-198,213,共7页Journal of Agricultural Mechanization Research

基　　金：国家自然科学基金项目(E050303);广西高校临海机械装备设计制造及控制重点实验室项目(GXLH2016ZD_08);广西教育厅基础能力提升项目(2017KY0807)

摘　　要：针对小型甘蔗收获机切割器不平衡对切割器轴向振动的影响,为实现切割器振动的有效预测以及自动控制信号的获取,通过正交试验并利用BP神经网络技术与回归分析构建出了切割器螺旋以及刀盘振动的BP神经网络模型和回归模型。分析结果表明:基于BP神经网络建立模型的切割器螺旋与刀盘的振动正确拟合率达到了88.89%,且相对误差基本上在5%以内,而回归模型的切割压力正确拟合率只有38.89%。因此,基于BP神经网络建立的模型具有较高的精度,通过此BP神经网络模型,有效地解决了复杂信息特征的提取问题,减少了试验研究的次数与成本,为进一步的切割器刀盘以及螺旋振动的自动控制系统的研发奠定了基础。In this paper the minitype sugarcane harvester cutter imbalance of cutter axial vibration, in order to effectively predict the realization of cutter vibration and obtain automatic control signal, the orthogonal experiment was constructed by cutter and cutter spiral BP neural network model and regression model of vibration analysis by using BP neural network and regression. The analysis results show that the correct fitting of cutter and cutter vibration spiral BP neural network model of the disk at a rate of 88.89% and the relative error is less than 5% basically based on the regression model of the cutting pressure correct fitting rate is only 38.89%, so BP neural network based on the established model has a high accuracy, the BP neural network the model effectively solves the problem of extracting complex features, reduce the number and cost of testing research, lay the foundation for the further development of the cutter wheel and spiral vibration automatic control system.

关 键 词：甘蔗收获机 切割器 螺旋 振动 BP神经网络 

分 类 号：S225.53[农业科学—农业机械化工程]