论文部分内容阅读
影响石油可采储量的因素是多种多样的，很难用一简单的表达式来描述两者之间的关系。神经网络专家系统为解决这一问题提供了新途径。具体过程为（１）选取储层参数（累积厚度、温度、有效孔隙度、有效渗透率、压力）和原油参数（含油饱和度、地下原油粘度和密度）等８个参数作为特征参数；（２）将参数进行标准化和归一化处理；（３）以樊家油田８个已知采油区作为学习样本对网络进行训练；（４）运用已训练好的神经网络专家系统对未知油区进行预测。预测结果与实际情况吻合很好，误差都在允许范围之内。
The factors that affect oil recoverable reserves are varied and it is difficult to describe the relationship between the two in a simple expression. Neural network expert system provides a new way to solve this problem. The specific process is as follows: (1) Eight parameters such as reservoir parameters (cumulative thickness, temperature, effective porosity, effective permeability and pressure) and crude oil parameters (oil saturation, viscosity and density of underground crude oil) ) To standardize and normalize the parameters; (3) Eight known zones in Fanjia Oilfield are used as training samples to train the network; (4) The trained expert system of NN is used to predict the unknown areas . The predicted results are in good agreement with the actual situation and the errors are within the allowable range.