神经模糊系统 ,即把神经网络与模糊逻辑结合起来 ,用神经网络来构造模糊系统 ,使建立的储层参数反演模型既能处理输入信息 ,又能嵌入专家的模糊性知识 ,提高了模型的抗干扰能力和预测精度 ,同时 ,也克服了人工神经网络技术的储层参数反演与预测在实际应用中暴露出的一些难点。通过基于神经模糊系统的网络结构、建模步骤及应用的研究 ,与常规多元线性回归分析、模糊系统建模相比较 ,获得了较高的储层参数反演预测精度。 The neural fuzzy system combines neural network and fuzzy logic to construct fuzzy system by neural network. The established reservoir parameter inversion model can not only process the input information, but also embed the expert’s fuzzy knowledge to improve the model’s Anti-interference ability and prediction accuracy. At the same time, it overcomes some difficulties exposed by the inversion and prediction of reservoir parameters by artificial neural network in practical application. Compared with the conventional multiple linear regression analysis and fuzzy system modeling, the network structure, modeling procedure and application research based on the neuro-fuzzy system get higher inversion accuracy of reservoir parameters.