湖北农业科学
湖北農業科學
호북농업과학
2014年
8期
1969-1971
,共3页
BP神经网络%粮食产量%预测%河南省
BP神經網絡%糧食產量%預測%河南省
BP신경망락%양식산량%예측%하남성
BP neural network%grain yield%prediction%Henan province
利用河南省1996-2011年粮食生产相关影响因素指标数据，建立了基于BP神经网络的河南省粮食产量的预测模型。该模型以播种面积、农田有效灌溉面积、农用机械总动力、化肥施用量、农药施用量5个指标为网络的输入值，粮食产量为输出值进行网络训练。预测结果表明，最大预测误差小于2%，说明BP网络模型具有较高的预测精度和稳定性，可为粮食产量预测提供一种新途径。
利用河南省1996-2011年糧食生產相關影響因素指標數據，建立瞭基于BP神經網絡的河南省糧食產量的預測模型。該模型以播種麵積、農田有效灌溉麵積、農用機械總動力、化肥施用量、農藥施用量5箇指標為網絡的輸入值，糧食產量為輸齣值進行網絡訓練。預測結果錶明，最大預測誤差小于2%，說明BP網絡模型具有較高的預測精度和穩定性，可為糧食產量預測提供一種新途徑。
이용하남성1996-2011년양식생산상관영향인소지표수거，건립료기우BP신경망락적하남성양식산량적예측모형。해모형이파충면적、농전유효관개면적、농용궤계총동력、화비시용량、농약시용량5개지표위망락적수입치，양식산량위수출치진행망락훈련。예측결과표명，최대예측오차소우2%，설명BP망락모형구유교고적예측정도화은정성，가위양식산량예측제공일충신도경。
The data of factors affecting food production of Henan province from 1996 to 2011 were used to forecast the grain yield by back propagation neural network. The prediction model of grain yield of Henan province was set up based on BP neural network. The model used 5 indicators including sown area, the effective irrigation area, the total power of agricultural machinery, the consumption of chemical fertilizer and the amount of pesticide application as the network input value, and grain yield as the output to do the network training. The prediction results showed that the maximum prediction error was less than 2%, indicating that the model had high accuracy and stability of prediction and would provide a new way for predicting grain yield.
