摘要
模糊规则的提取和隶属度函数的学习是模糊推理系统设计中重要而困难的问题。自适应神经模糊推理系统 (ANFIS)方法基于Sugeno模糊模型 ,其结构类似于神经网络 ,采用反向传播算法和最小二乘法调整模糊推理系统的参数 ,并能自动产生模糊规则。本文应用该方法给出了对一个典型系统建模的仿真实例 ,取得了良好的效果。
Extraction of fuzzy rules and learning of parameters of membership functions play an essential role in the design of a fuzzy inference system but they are difficult. Adaptive Neural-Fuzzy Inference System (ANFIS) method is based on Sugeno fuzzy model and has a structure similar to neural network that tunes the parameters of the fuzzy inference system with backpropagation algorithm and least-square method and can produce fuzzy rules automatically. This paper gives the simulation example of modeling a typical system with ANFIS method and good result is obtained.
出处
《计算机仿真》
CSCD
2002年第4期47-49,共3页
Computer Simulation
关键词
自适应神经模糊推理系统
建模
仿真
Fuzzy Inference System
ANFIS
modeling and simulation
fuzzy clustering