欢迎访问《中国农学通报》,

中国农学通报 ›› 2010, Vol. 26 ›› Issue (8): 341-344.

• 工程 机械 水利 装备 • 上一篇    下一篇

温室环境进化RBF网络控制模型的研究

赵斌1,2,王克奇1,匡丽红2   

  • 收稿日期:2009-11-16 修回日期:2009-12-09 出版日期:2010-04-20 发布日期:2010-04-20
  • 基金资助:

    大豆精准生产技术系统构建与应用

Research on Greenhouse Environmental RBFNN Control Model Based on ES

Zhao Bin1,2, Wang Keqi1, Kuang Lihong2   

  • Received:2009-11-16 Revised:2009-12-09 Online:2010-04-20 Published:2010-04-20

摘要:

温室环境模型是制定控制策略、优化控制方法的重要基础。采用径向基网络构建北方地区的温室环境控制模型,以温室内的温度和湿度为输出,采用进化策略(ES)优选网络中心,简化网络结构,并与OLS算法进行性能对比。试验结果表明:进化RBF网络具有更小的网络结构,并在逼近能力和泛化能力上优于OLS算法。该模型为设计北方温室环境控制方案和结构奠定了良好基础。

关键词: 黄粉虫, 黄粉虫, 诱导, 抗菌肽

Abstract:

The mode of greenhouse environment is a important basis of instituting control tactic and optimizing control method. The RBFNN is constructed environmental control model inside greenhouse in the north of China. NN outputs were temperature and humidity inside greenhouse. NN hidden unit centers were optimized, and NN configuration was simplified by means of ES. Both algorithms were compared. Experiment result is showed: approximate and generalization ability of RBFNN based on ES is better to OLS algorithm. The model is beneficial to design control scheme and configuration of the north greenhouse.