油层的非均质性归因于岩石特性的非均质和非线性空间分布，而孔隙度和渗透率是非均质性的集中表现。近年来，人工神经网络（ＡＮＮ）技术在油藏描述中的应用十分广泛，使用ＡＮＮ技术为试井选择合适的油藏模型则更具实用性。本文提出了表征油藏非均质性的一种人工神经网络途径，介绍了用ＡＮＮ技术建立较为可靠的孔隙度、渗透率解释模型，并对丘陵油田三间房组油层非均质性进行了研究 The heterogeneity of the reservoir is attributed to the heterogeneous and nonlinear spatial distribution of rock properties, while porosity and permeability are the concentrated manifestations of heterogeneity. In recent years, Artificial Neural Network (ANN) has been widely used in reservoir description. It is more practical to use ANN technology to select the appropriate reservoir model for well testing. In this paper, an artificial neural network approach to characterize reservoir heterogeneity is proposed. A more reliable interpretation model of porosity and permeability using ANN technique is introduced. The heterogeneity of reservoir in Sanfangfang Formation in Hilly Oilfield the study