一种加权模糊推理网络模型及其应用

提出了一种加权模糊推理网络的结构模型和学习算法，该网络的基本信息处理单元为模糊推理神经元，融合了模糊逻辑能够较完整地表达领域规则和先验知识，以及神经网络自适应环境的优点。根据模糊推理规则的量化表示形式和微分方程数值解的动力学思想推导出了该网络模型的学习算法。该算法具有稳定、收敛速度快，且能较好地避免网络学习陷入局部极值点。以油田生产复杂水淹层识别问题为例，验证了模型和算法的有效性。

Weighted fuzzy reasoning network and its application

A weighted fuzzy reasoning network structure model and its algorithm are proposed. The basic unit of information processing in this network is fuzzy reasoning neuron. The algorithm combines the excellence of fuzzy logic that can express domain rules and me-tempirical knowledge wholly with the virtue of neural network that can adapt to the environment automatically. A novel study algorithm of the network is shown that is based on fuzzy reasoning rule and numerical method for differential dynamic systems. This algorithm has better stability and fast speed quality of convergence. This algorithm can overcome some shortages of falling into local minima. Finally an experiment example is proposed to illustrate the availability of the network and algorithm by water flooded layer identification in oil field producing.