摘要
自然通风状态下温室内空气温度的合理估测是全开型玻璃温室夏季温度控制的重要依据。为此,建立了基于RS-485总线的分布式多传感器温度测控网络,并采用分布图法和基于均值的数据融合方法对测量结果修正融合,实现了全开型玻璃温室温、湿度的精确测量。在此基础上,以室外空气温度、太阳辐射强度、室内通风速率以及室内空气相对湿度作为室内空气温度的影响因子,针对各因子的非平稳时间序列建立了室内空气温度时间序列模型,同时引入了协整方法避免伪回归现象,引入了误差修正方法提高模型预测精度。实际验证表明,时间序列模型预测数据与实测数据吻合良好,可较好地预测温室内温度,该研究成果为全开型玻璃温室的温度控制提供了理论依据。
A reasonable estimate of greenhouse internal air temperature is an important basis to control the internal air temperature in a naturally ventilated open air greenhouse in summer.The distributed multi-sensor temperature measurement network based on RS-485 bus was developed in this paper.The distribution method and the data fusion method based on meaning value were introduced to improve the greenhouse air temperature and greenhouse air relative humidity measurement accuracy.Furthermore,the main external environment variables that influence the greenhouse internal air temperature were external air temperature,ventilation rate,solar radiation,and greenhouse air relative humidity-both of them were non-stationary variables.A temperature time series model was established using the co-integration and error correction methods.In this paper,the co-integration method was used to avoid spurious regression,the error correction method was used to improve the prediction accuracy of the model.A comparison between the forecasts and the measured data shows that the model was able to forecast internal air temperature reasonably well.These researches provide a theoretical basis for the temperature control in the open air greenhouse.
出处
《农机化研究》
北大核心
2010年第11期178-184,共7页
Journal of Agricultural Mechanization Research
基金
国家"863"计划项目(2006AA10A311)
上海市重点学科建设项目资助(B209)
上海市科学技术委员会科技攻关计划项目(073919104)
关键词
全开型玻璃温室
数据融合
协整
误差修正
时间序列模型
the open air greenhouse
data fusion
co-integration
error correction
time series model
