Evaporation from sub-surface reservoirs is a phenomenon that has drawn a considerable amount of attention, over recent years. An accurate prediction of the sub-surface evaporation rate is a vital step towards drawing better managing of the reservoir’ water system. In fact, the evaporation rate and more specifically from sub-surface is considered as highly stochastic and non-linear process that affected by several natural variables. In this research, a focuses on the development of an Artificial Intelligence (AI) model, to predict the evaporation rate has been proposed. The model’s input variables for this model include temperature, wind speed, humidity and water depth. In addition, two AI models have been employed to predict the sub-surface evaporation rate namely: Generalized Regression Neural Network (GRNN) and Radial Basis Function Neural Network (RBFNN) as a first attempt to utilize AI models in this topic. In order to substantiate the effectiveness of the AI model, the models have been applied utilizing actual hydrological and climatological in an arid region, for two soil types: fine gravel (F.G) and coarse gravel (C.G). The prediction accuracy of these models has been assessed through examining several statistical indicators. The results showed that the Artificial Neural Networks (ANN) model has the capacity for a highly accurate evaporation rate prediction, for the subsurface reservoir. The correlation coefficient for the fine gravel soil, and coarse gravel soil, was recorded as 0.936 and 0.959 respectively.

从地下储层中蒸发是一种现象，近年来引起了相当多的关注。准确预测地下蒸发速率是迈向更好地管理储层水系统的重要一步。实际上，蒸发速率，更具体地说是从地下的蒸发速率，被认为是高度随机且非线性的过程，受几个自然变量的影响。在这项研究中，重点是开发人工智能（AI）模型以预测蒸发速率。该模型的模型输入变量包括温度，风速，湿度和水深。此外，已经采用了两种AI模型来预测地下蒸发速率：广义回归神经网络（GRNN）和径向基函数神经网络（RBFNN）作为在本主题中利用AI模型的首次尝试。为了证实效力的人工智能模型，这些模型已经应用于实际利用水文和气候在干旱地区，对于两个土壤类型：细砾（FG）和粗砂砾（CG）。这些模型的预测准确性已通过检查几个统计指标进行了评估。结果表明，人工神经网络（ANN）模型具有用于地下储层的高精度蒸发速率预测的能力。细砂砾土和粗砂砾土的相关系数分别记录为0.936和0.959。