基于神经网络的多功能收割机发动机性能仿真及优化

发布时间：2018-01-12 23:02

本文关键词：基于神经网络的多功能收割机发动机性能仿真及优化　出处：《农机化研究》2017年12期 　论文类型：期刊论文

【摘要】：随着全球经济和工业的快速发展,能源危机和环境保护问题越来越突出,传统内燃式发动机受到了巨大冲击,因此研究高效、节能的发动机显得尤为重要。为此,研究一种压缩比为10.6多功能收割机,并设计了基于MatLab仿真平台的GT-Power神经网络模型,利用神经网络训练和测试的数据自动进行仿真和储存,采用拉丁超立方采样算法设计试验,简化运算过程提高寻优效率。实验结果表明:神经网络模型转矩、比油耗和温度等参数模型预测误差很小、精度很高,可用于多功能收割机发动机的性能预测,使其各项指标参数最优化。
[Abstract]:With the rapid development of global economy and industry, the energy crisis and environmental protection are more and more prominent. The traditional internal combustion engine has been greatly impacted, so the research is highly efficient. Energy saving engine is particularly important. Therefore, a 10.6 multifunctional harvester with compression ratio is studied, and a GT-Power neural network model based on MatLab simulation platform is designed. The neural network training and testing data are automatically simulated and stored, and the Latin hypercube sampling algorithm is used to design the experiment, which simplifies the operation process and improves the optimization efficiency. The experimental results show that the torque of the neural network model is improved. The prediction error of specific fuel consumption and temperature model is very small and the precision is very high. It can be used to predict the performance of multi-function harvester engine and optimize its parameters.
【作者单位】：江苏经贸职业技术学院;
【基金】：江苏省农业科技自主创新资金项目(CX(16)60732) 江苏经贸职业技术学院创新项目(JSJMY015)
【分类号】：S225;TP183
【正文快照】： 0引言由于农作物收获期短,秋收时期时间非常紧张,收割机进行作业时,一片农田区域往往需要在1周之内完成收割任务,收割压力大,因此对收割机的要求非常高,甚至会出现收割机不歇驾驶员轮流休息的抢收局面。收割机连续工作时间长,发动机油门一拉到底,对发动机的可靠性要求比较苛刻

【参考文献】