摘要
甘蔗转运车负责甘蔗田间的转运工作,针对广西60%以上甘蔗种植区都是丘陵地的情况,转运车工况不同于平原地带,除了正常工作状态以外,还会出现无载重车体倾斜、过载的存在安全隐患的工作状态。在课题组自主开发的一种适于丘陵地区作业的双剪叉式可升降的转运车的基础上,为了准确预测转运车节点应力,保证转运工作的平稳高效运行,使用长短期记忆网络构建转运车关节点应力预测模型,实现对节点应力数据地预测。使用转运车工作过程中的节点应力数据进行模型训练与测试分析。试验结果表明,在短期节点应力预测中,基于长短期记忆网络(Long-Short Term Memory,LSTM)的转运车检测模型可以有效地预测节点应力变化。
Sugarcane transit car is responsible for transport the sugarcane field for more than 60%of guangxi sugar cane area is hilly land,transport vehicle conditions different plain area,in addition to the normal working state,there will be no load bodywork tilt overload safety working condition based on the research of independent development of a suitable for hilly region homework can be elevating double shear transfer car,on the basis of node stress in order to predict the transfer car,ensure the smooth and efficient operation of transport.The LSTM neural network model was constructed with the deep learning framework keras,and the node stress time series data were regressed and predicted.The node stress data of the whole working process of the transport vehicle under certain experimental conditions were used for simulation.The results showed that the artificial neural network(Long-)based on Long-and short-term memory was used.
作者
袁泓磊
李尚平
李向辉
李凯华
张伟
黄宗晓
YUAN Hong-lei;LI Shang-ping;LI Xiang-hui;LI Kai-hua;ZHANG Wei;HUANG Zong-xiao(College of Information Science and Engineering,Guangxi University for Nationalities,Nanning 530000,China;College of Mechanical&Engineering,Guangxi University,Nanning Guangxi 530004,China)
出处
《装备制造技术》
2020年第4期1-4,30,共5页
Equipment Manufacturing Technology
基金
广西创新驱动发展专项资金科技重大专项“走式田间甘蔗收集搬运车的研究及开发”(桂科AA17202015-5)。
