摘要
研究农业生产中病虫害灾变临界的准确监控方法。生物灾变复杂的生态过程,是生态系统中物质流、能量流和信息流的内变量众多,且变量之间并非简单的线性关系,灾变发生是种群数量不连续的突然变化,灾变的突发是小概率事件,又具有多因素、非线性和不确定性的特点,其大尺度的时空变化过程未必呈周期性,有的事例甚至绝无仅有,传统的线性物联网监控的条件被推翻,造成物联网监控困难。为解决上述问题,提出一种以物联网结构为基础的物联网监控模型。建立异常物联网监控信号搜索模型,计算信号特征模糊聚类概率。根据病虫害灾变多衡量标准,对获取的异常物联网监控信号进行分析,从而实现大区农业中的病虫害灾变临界监控。实验结果表明,利用本文算法进行大区农业中病虫害灾变临界的物联网监控,可以极大地提高监控的准确性,从而获取病虫害灾变临界的准确数据。
The method to monitor accurately catastrophe crisis of plant diseases and insect pests in agricultural production was researched. An IoT monitor model based on the structure of IoT was put forward. The model of searching abnormal loT monitoring signal was set up to calculate fuzzy clustering probability of signal features. Ac cording to the pest catastrophe measure, abnormal IoT monitoring signals were analyzed. The lot monitoring of plant diseases and insect pests catastrophe critical in regional agriculture experimental results show that the algorithm pres ented in this paper for IoT monitoring of plant diseases and insect pests catastrophe critical in regional agriculture, can greatly improve the accuracy of the monitoring, so as to obtain the pest catastrophe critical data accurately.
出处
《计算机仿真》
CSCD
北大核心
2014年第4期299-302,共4页
Computer Simulation
关键词
病虫害
灾变临界
物联网监控
特征模糊聚类概率
Plant diseases and insect pests
Cataclysm crisis
Internet monitoring
Characteristics of the fuzzyclustering probability
