灰色神经网络下的多变量土壤含水量预测模型

Prediction Model of Multivariate Soil Water Content Constructed by the Grey Model Combined with Neural Network and Three-parameter Interval Grey Number

土壤含水量是评价农业墒情的重要指标,其变化规律对节水农业具有重要意义。基于土壤含水量预测问题中测量数据的特征和土壤含水量变化的内在联系,结合灰色GM(0,N)模型和BP神经网络模型的优势,构建BPSGM(0,N)模型。以河南省新郑市土壤含水量预测问题为例,基于实际测量数据,对该模型的预测值进行残差分析和关联度检验。结果表明,BPSGM(0,N)模型在土壤含水量预测方面具有良好的效果。

The soil water content is an important index of evaluating soil moisture,the change law of soil water content is of great significance in water-saving agriculture. In the paper,combing the advantages of grey model GM（ 0,N） and BP neural network model,the model BPSGM（ 0,N） was constructed based on the characteristics of predicted data and the internal relation in the change of soil water content in the prediction problem of soil water content. Regarding the prediction of soil water content of Xinzheng City in Henan Province as an example,the residual value analysis and correlation test of the predicted values of the model were carried out based on the actual measurement data. The results show that the model BPSGM（ 0,N） can obtain good forecasting results for soil water content.