Evaluating the Neural Network Ensemble Method in Predicting Soil Moisture in Agricultural Fields

Soil is an important element in the agricultural domain because it serves as media that bridges water consumption and supply processes. In this study, a neural network ensemble (NNE) method was employed to predict soil moisture eliminate effects of random initial parameters (NN) on model accuracy. The constructed NNE predicts daily root zone continuously for whole crop growing season processes were separately modeled. profile divided into multiple layers modeled separately. Weather data (including air temperature, humidity, wind speed, net radiation, precipitation), rooting depth, hesternal each layer used input. A calibrated quality maize (Zea mays L.) generate training evaluation data. result showed with 100 randomly initialized NN models, achieved average R2 0.96 nRMSE 5.93%, suggesting learned dynamics well sufficiently improved robustness prediction high