摘要
土壤水分含量是影响作物生长和农田水土环境的一个重要指标,土壤水分预测对农业生产中水分的合理利用与管理具有重要意义。应用支持向量机理论,采用土壤水分含量影响较大的降雨量、蒸发量、相对湿度和地下水埋深作为模型输入因子,土壤水分含量为输出因子,建立了土壤水分预测模型,并与灰色理论水分预测模型进行了对比。结果表明:支持向量机模型预测土壤含水量的最大误差为5.83%,平均误差为2.44%,预测结果优于灰色预测理论,模型的预测精度为0.957。
Soil moisture is of an important index that affects crops growth and farmland environment.The prediction of soil moisture has significance to the rational utilization and management of moisture in agricultural production.It builds soil moisture predictive model by using the theory of supporting vector,using precipitation,evaporation,relative humidity and groundwater level that have greater influence to the soil moisture as input factor of the model and using soil moisture as output factor and compares with gray theory soil moisture predictive model.The outcomes show that the maximum error soil moisture of the model is 5.83% and average error 2.44%.The predictive results are better than that of the gray predictive theory and the predictive accuracy of the model is 0.957.
出处
《人民黄河》
CAS
北大核心
2011年第3期68-69,72,共3页
Yellow River
