日光温室湿度日预测的季节时序模型应用研究

日光温室内湿度环境的调控是设施农业生产监测与预测的重要内容，现以季节性时间序列模型的理论和方法，对石家庄高邑地区的代表温室内，2008年10月至2009年6月间的月平均日湿度值为分析数据，建立了当地温室的日湿度AR（p）模型，用实测数据与模型的预测数据相比较，均方根误差RMSE和相对误差RE分别为9.90和 -8.65%，说明时间序列模型能较好地拟合与预测日湿度的变化趋势,可为日光温室的湿度环境调控书提供理论依据和决策支持。

Application of Seasonal Time Series Model in Forecasting Greenhouse Daily Humidity

Control of humidity environment in greenhouse is an important item of humidity monitoring and forecasting.In this paper,the method of seasonal time-series analysis was proposed to predict the air humidity of sunlight greenhouse,the data collected from Oct.2008 to June 2009 in some Gaoyi greenhouse which is representative in Shijiazhuang area was studied.Using the theories and methods of seasonal time series model,an AR （p） model to forecast the local greenhouse inner daily humidity was built up.By comparing the measured value with the estimated value,the RMSE and the average RE were 9.90,-8.65%,it is showed that the AR model can preferably fit the greenhouse inner humidity,and also can be used theory instruction and management supplying in sunlight humidity environment control.