基于时序残差辨识的终霜日灰预测模型研究  

Grey Prediction Model Research of the Last Frost Date Based on Time Series Residual Error Identification

作　　者：晁林海 汪顺勤 刘升 

机构地区：[1]安徽宿州农委,安徽宿州234000 [2]安徽宿州气象局,安徽宿州234000 [3]安徽萧县气象局,安徽萧县235200

出　　处：《安徽农业科学》2009年第16期7562-7563,共2页Journal of Anhui Agricultural Sciences

摘　　要：终霜冻出现的早晚是影响宿州市春播作物和越冬作物生长发育的重要因素。根据季节事件灰预测机理和时序变量灰化处理思想建立时序残差GM(1,1)修正模型,模型精度有所提高,对终霜期的长期预测具有一定的参考价值。The last frost appearance sooner or later is an important factor that affects spring sowing and growth of winter crop in Suzhou. According to the principle of seasonal gray prediction mechanism and the thought of time series variables gray processing, time series residual error GM （ 1,1 ） correction model is established, and model accuracy is improved, it has certain reference value for long -- term forecasting the last frost time.

关 键 词：终霜 时序残差 灰预测 

分 类 号：S166[农业科学—农业气象学]