摘要
提出了一种适用于温室环境参数融合的混合算法,使用D-S理论推测某一时刻环境状态,用小波网络实现宏观决策,此算法解决了传统D-S证据理论中概率赋值困难的问题,同时也为小波神经网络提供了比较准确实用的训练数据。最后以温室樱桃西红柿为例,通过仿真说明此混合算法在农作物温室培养过程中对于环境参数的融合效果,为温室的控制提供了准确可信的依据。
A greenhouse environment parameters applicable to the integration of hybrid algorithm is presented, which can solve the issues of probability assignment by the traditional theory of D-S evidence, as well as to provide an accurate and practical training data for WNN. Take the high economic value of cherry,tomatoes for example,it shows algorithm in greenhouse cultivation of crops in the process of application.
出处
《传感器与微系统》
CSCD
北大核心
2008年第11期32-34,37,共4页
Transducer and Microsystem Technologies
基金
国家"863"计划资助项目(2006AA10A301)
关键词
D—S证据理论
小波网络
数据融合
温室控制
D-S evidence theory
wavelet-network
information fusion
greenhouse control
