摘要
人工神经网络是属于机器学习领域中的监督学习,利用神经网络的非线性逼近能力,对病害的发生时间、感染程度以及病害迁移方向进行预测。相比于传统的病害预测方式,人工神经网络具有实时性高、精度高、容错性强等特点,为农业现代化提供了另一种技术支撑。本文通过研究资料发现,农业病害的发生与气候有巨大的关系,可结合这点并利用神经网络的强大学习能力对历史病害数据与发生程度进行分类处理,最后再利用已训练好的网络模型预测未来病害的发生趋势。
Artificial neural network belongs to supervised learning in the field of machine learning.It uses the nonlinear approximation ability of neural network to predict the occurrence time,infection degree and disease migration direction.Compared with the traditional way of disease prediction,artificial neural network has the characteristics of high real-time,high precision and strong fault tolerance,which provides another technical support for agricultural modernization.In this paper,we found that the occurrence of agricultural diseases had a great relationship with the climate.Combined with this point,we used the strong learning ability of neural network to classify the historical disease data and the degree of occurrence.Finally,we used the trained network model to predict the occurrence trend of future diseases.
作者
陈民
胡雪琼
鲁韦坤
周文文
陈亚平
李晓君
曹志勇
CHEN Min;HU Xueqiong;LU Weikun;ZHOU Wenwen;CHEN Yaping;LI Xiaojun;CAOZhiyong(College of Big Data,Yunnan Agricultural University,Kunming Yunnan 650201;Yunnan Climate Center,Kunming Yunnan 650034;Yunnan Plant Protection and Quarantine Station,Kunming Yunnan 650034;College of Plant Protection,Yunnan Agricultural University,Kunming Yunnan 650201;Dalian Puwan Engineering Project Management Co.,Ltd.,Dalian Liaoning 116100)
出处
《现代农业科技》
2020年第21期136-140,共5页
Modern Agricultural Science and Technology
基金
云南省气候中心项目“高原特色农作物主要病虫害气象预测模型研发”(YYHZ190701)。
关键词
农业病害
预测
人工神经网络
agricultural disease
prediction
artificial neural network
