基于BP神经网络的农业机械数量预测  

Prediction of the Number of Agricultural Machinery Equipment Based on BP Neural Network

作　　者：王笑岩[1] 王石[2] 周琪[3] 

机构地区：[1]朝阳师范高等专科学校数学计算机系,辽宁朝阳122000 [2]沈阳工程学院机械学院,沈阳110866 [3]渤海大学数理学院,辽宁锦州121000

出　　处：《农机化研究》2015年第3期11-14,共4页Journal of Agricultural Mechanization Research

基　　金：国家自然科学基金青年基金项目(61304003);辽宁省高教学会"十二五"高等教育科研项目(GHYB110216)

摘　　要：为了提高农业机械化水平、农业生产效率和优化农业产业结构,保证在农业机械生产与实际需求的一致性,在制订农业机械化水平发展规划过程中需要对农业机械数量进行预测。为此,采用基于遗传算法的BP神经网络预测算法,对我国从1997-2013年期间以农机总动力、中大型拖拉机数量和小型拖拉机数量为内容的主要农业装备数量进行预测。预测结果表明,利用遗传算法与BP神经网络相结合的方法预测全国农业机械装备数量,农机总动力预测值与绝对值平均误差为1.080%、农用大中型拖拉机数量预测值与绝对值平均误差为1.352%、小型拖拉机数量预测值与绝对值平均误差为1.765%,预测精度稳定,该预测方法适用于本时间序列预测问题。In order to improve agricultural mechanization and productivity , optimize agricultural structure and guarantee the consistency between agricultural mechanical production and actual needs , the number of agricultural machines should be estimated in the process of planning the development of agricultural mechanization .The paper selected BP neural net-work estimation method based on genetic algorithm to estimate the number of major agricultural devices , including total power of agricultural machinery , the number of middle-and large-sized tractors and the number of small tractors .The estimation result showed that through making use of the combination of genetic algorithm and BP neural network to esti -mate the number of agricultural machines in the whole country , the average error between estimated value and the abso-lute value of total power of agricultural machinery , the number of middle-and large-sized agricultural tractors and the number of small agricultural tractors was 1 .080%, 1 .352%and 1 .765%respectively .The estimation precision was sta-ble.Besides, the method was applicable in the estimation of time sequence .

关 键 词：BP神经网络 预测 农业机械 遗传算法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] S232.9[自动化与计算机技术—控制科学与工程]