基于反向传播的非单点模糊系统辨识研究  

Research on Non Singleton Fuzzy System Identification Using Back Propagation

作　　者：董宸 

机构地区：[1]中国能源建设集团江苏省电力设计院有限公司,南京211102

出　　处：《仪器仪表用户》2017年第5期32-35,共4页Instrumentation

摘　　要：针对实际输入输出数据测量噪声对模糊辨识精度的影响,提出了基于反向传播的非单点模糊逻辑系统。模糊推理系统基于非单点模糊化,提高了系统的抗干扰能力。利用反向传播确定输入变量的方差,提高了系统的辨识精度。用Mackey-Glass混沌数据在不同信噪比噪声情况下,验证了文章中算法的有效性和实用性。In according with the identification effected by noise of real input and output data, a non-singleton fuzzy logic system based on back propagation is proposed in this paper. The fuzzy inference system uses non-singleton fuzzy, which strengthen the ability of immunity. The input variables variance is decided by back propagation, and improves the identification accuracy of fuzzy system. The effectiveness and practicality of which are illustrated by Mackey-Glass chaos datas under different signal to noise ratio noise.

关 键 词：模糊基函数 非单点模糊化 模糊辨识 反向传播 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]