摘要
甘蔗联合收割机收获质量对制糖工艺有极大影响,但测量难度大,难以直接获得。针对上述问题,以甘蔗联合收割机切割机构、行走机构、切段机构、风机机构的负载压力信号和转速信号为输入变量,以含杂率和损失率为输出变量,建立了一种GA-BP神经网络预测模型。GA-BP神经网络预测模型对甘蔗收获质量的预测结果平均MSE为0.0937,平均R 2为0.8915,和建立的BP神经网络进行对比模型的均方误差降低了18.17%,决定系数提高了2.59%,进一步说明了利用遗传算法优化BP神经网络的权值和阈值建立的GA-BP模型比传统的BP神经网络具有更好的预测精度。研究结果为测量甘蔗收获质量提供了一种新思路,并为甘蔗收割机不同工况下提高甘蔗收获质量的各子系统的协调联动控制策略提供了理论依据。
The harvest quality of sugarcane combine harvester has great influence on sugar processing technology,but it is difficult to measure and obtain directly.Aiming at the above problems,a GA-BP neural network prediction model is established by taking the load pressure signal and rotate speed signal of the cutting mechanism,walking mechanism,chopper mechanism and fan mechanism of sugarcane combine harvester as the input variables,and the impurity rate and loss rate as the output variables.GA-BP neural network prediction model on the quality of the sugarcane harvest predicted results average MSE is 0.0937,the average R 2 is 0.8915.Compared with the BP neural network model,MSE reduced by 18.17%,the R 2 reduced by 2.59%.These resulits further illustrates prediction accuracy of the GA-BP model is better than the traditional BP neural network.This article provide a new way to measure the sugarcane harvest quality,and a theoretical basis for t-he coordinated linkage control strategy of each subsystem to improve the sugarcane harvest quality under different working conditions of the sugarcane harvester.
作者
陈远玲
王肖
孙英杰
张阳
Chen Yuanling;Wang Xiao;Sun Yingjie;Zhang Yang(College of Mechanical Engineering,Guangxi University,Nanning 530004,China)
出处
《农机化研究》
北大核心
2022年第2期187-191,共5页
Journal of Agricultural Mechanization Research
基金
国家自然科学基金项目(51665004)。
关键词
甘蔗收割机
收获质量
含杂率
损失率
BP神经网络
遗传算法
sugarcane harvester
harvest quality
impurity rate
loss rate
BP neural network
genetic algorithm(GA)
