基于扩展输入空间的迟滞神经网络模型  

Neural network based hysteresis model on expanded input space

作　　者：谭永红[1] 张新良[2] 董瑞丽[1] 

机构地区：[1]上海师范大学信息与机电工程学院,上海200234 [2]上海交通大学电子信息与电气工程学院自动化系,上海200240

出　　处：《上海师范大学学报（自然科学版）》2009年第3期229-236,共8页Journal of Shanghai Normal University(Natural Sciences)

基　　金：上海师范大学重学科项目(DZL811);上海市教委科技创新重点项目(09ZZ141);上海师范大学前瞻性项目(DYL200809);国家自然科学基金(60572055)

摘　　要：针对迟滞的非光滑、多值映射特性进行分析,提出了反映迟滞主要特征的迟滞算子的构造方法,并将迟滞算子引入扩展输入空间,实现了将迟滞的多值映射转换为一一映射,建立了关于迟滞的神经网络模型.避免了对非光滑的迟滞求梯度,获得了简单的模型结构和较高的建模精度.An approach to construct the hysteretic operator is proposed based on the analysis on the characteristic of hysteresis with non-smooth and multi-valued mapping. By introducing the proposed hysteretie operator into the expanded input space, the multi-valued mapping of hysteresis can be transferred into a one-to-one mapping. Thus, the neural network based model can be derived based on the expanded input space. This method avoids the calculation of gradients of the non-smooth hysteresis with respect to its input. Then, an accurate model with simpler architecture has been obtained.

关 键 词：迟滞 非光滑系统 扩展输入空间 模型 

分 类 号：TP2[自动化与计算机技术—检测技术与自动化装置]