Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (17): 218-220.DOI: 10.3778/j.issn.1002-8331.2009.17.066

• 工程与应用 • Previous Articles     Next Articles

Application of neural network based on improved PSO in data fusion of greenhouse temperature

ZHANG You-jun1,XIONG Wei-li1,ZHANG Lin2,XU Bao-guo1   

  1. 1.School of Communication and Control Engineering,Jiangnan University,Wuxi,Jiangsu 214122,China
    2.Electronic Engineering Department,Tsinghua University,Beijing 100084,China
  • Received:2008-04-09 Revised:2008-06-26 Online:2009-06-11 Published:2009-06-11
  • Contact: ZHANG You-jun


张酉军1,熊伟丽1,张 林2,徐保国1   

  1. 1.江南大学 通信与控制工程学院,江苏 无锡 214122
    2.清华大学 电子系,北京 100084
  • 通讯作者: 张酉军

Abstract: The temperature distribution in the greenhouse influenced by many kinds of environmental factors is uneven.In order to get precise data,the neural network based on improved PSO is proposed for greenhouse data fusion,and the distributing diagram approach is used to eliminate the careless mistake data.Data fusion technology gets efficient data,providing precise information for greenhouse’s management.The results show that the precision of the collected data is improved and the careless error caused by disabled sensors is eliminated effectively.

摘要: 由于温室环境受到各种因素影响,导致分布在各点的温度值不均匀,为了获得温度的准确值,提出了基于改进PSO的神经网络对其进行数据融合,并且采用分布图法剔除多传感器离异数据,最终得到准确有效的数据,为温室管理提供了精确的信息。仿真结果表明,采用这种方法可以提高温度采集的准确性,并且有效地消除了由于传感器失效引起的误差。