水科学进展
水科學進展
수과학진전
2010年
1期
63-68
,共6页
张展羽%王声锋%段爱旺%王斌
張展羽%王聲鋒%段愛旺%王斌
장전우%왕성봉%단애왕%왕빈
天气预报%参考作物腾发量%最小二乘支持向量机%预测模型
天氣預報%參攷作物騰髮量%最小二乘支持嚮量機%預測模型
천기예보%삼고작물등발량%최소이승지지향량궤%예측모형
weather forecasts%reference evapotranspiration%least squares support vector machines%prediction mode
利用最小二乘支持向量机(LS-SVM)方法, 建立了基于天气预报的参考作物腾发量(ET_0)的预测模型.对广利灌区1997～2006年逐日气象信息中的天气类型和风速等级进行量化后,以不同天气预报信息作为输入量,建立10种验证方案,对2007年的逐日ET_0进行预测.经验证,方案1～方案7精度均令人满意,其中方案1精度最高.方案1的输入量为气温、天气类型、风速等级3项的预测值,该方案的模型预测值与计算值的统计参数分别为:均方根偏差E_(RMS)为0.5182mm,相对偏差E_R为0.1878,决定系数R~2为0.8648,认同系数I_A为0.9669,回归系数R_C为0.9867;方案7精度亦较好,且以上指标统计参数依次为0.6576mm、0.2332、0.9866、0.7747及0.9866,该方案输入量只有气温项,实用性很强.
利用最小二乘支持嚮量機(LS-SVM)方法, 建立瞭基于天氣預報的參攷作物騰髮量(ET_0)的預測模型.對廣利灌區1997～2006年逐日氣象信息中的天氣類型和風速等級進行量化後,以不同天氣預報信息作為輸入量,建立10種驗證方案,對2007年的逐日ET_0進行預測.經驗證,方案1～方案7精度均令人滿意,其中方案1精度最高.方案1的輸入量為氣溫、天氣類型、風速等級3項的預測值,該方案的模型預測值與計算值的統計參數分彆為:均方根偏差E_(RMS)為0.5182mm,相對偏差E_R為0.1878,決定繫數R~2為0.8648,認同繫數I_A為0.9669,迴歸繫數R_C為0.9867;方案7精度亦較好,且以上指標統計參數依次為0.6576mm、0.2332、0.9866、0.7747及0.9866,該方案輸入量隻有氣溫項,實用性很彊.
이용최소이승지지향량궤(LS-SVM)방법, 건립료기우천기예보적삼고작물등발량(ET_0)적예측모형.대엄리관구1997～2006년축일기상신식중적천기류형화풍속등급진행양화후,이불동천기예보신식작위수입량,건립10충험증방안,대2007년적축일ET_0진행예측.경험증,방안1～방안7정도균령인만의,기중방안1정도최고.방안1적수입량위기온、천기류형、풍속등급3항적예측치,해방안적모형예측치여계산치적통계삼수분별위:균방근편차E_(RMS)위0.5182mm,상대편차E_R위0.1878,결정계수R~2위0.8648,인동계수I_A위0.9669,회귀계수R_C위0.9867;방안7정도역교호,차이상지표통계삼수의차위0.6576mm、0.2332、0.9866、0.7747급0.9866,해방안수입량지유기온항,실용성흔강.
A reference evapotranspiration (ET0) prediction model is developed based on the least squares support vector machines. Weather forecasts are used for ET0 predictions. The model can be trained with daily weather parameters including quantified weather types and wind grades, etc. Different combinations of daily weather parameters can be tested in the model training processes. In this study, the daily weather parameters are obtained from the Guangli irrigation district during the period 1997-2007. The 1997-2006 data are used for the model training, and a total of 10 daily ET0 forecasting schemes are established. Predictions of daily ET0 using each of the 10 schemes are validated with the observations from 2007. The results show that the scheme using the air temperature and the quantified weather types and wind grades as model predictors is able to give the best model performance; and the corresponding statistics are the root mean square error (E_(RMS)) of 0.5182, the relative error (E_R) of 0.1878, the coefficient of determination (R~2) of 0.8648, the I_A of 0.9669, and the regression coefficient (R_C) of 0.9868. Acceptable daily ET0 predictions are also obtained with other 6 schemes, among which the simplest scheme using only the air temperature as the model predictor can also produce fair results as revealed by E_(RMS) = 0.6576,E_R = 0.2332,R~2 = 0.9866,I_A = 0.7747 and R_C = 0.9680. The latter scheme shows a strong potential in practical applications.
