基于优化RBF神经网络的缝纫平整度客观评价  

Objective Evaluation of Sewing Properties Based on Optimized RBF Neural Network

作　　者：吴懋刚[1] 潘永惠[1] 范蕤[1] 

机构地区：[1]江阴职业技术学院计算机科学系,江苏无锡214405

出　　处：《通化师范学院学报》2010年第10期37-38,66,共3页Journal of Tonghua Normal University

摘　　要：该文提出了一种基于优化RBF神经网络的丝织物缝纫性能客观评价系统.该系统根据丝织物面料的FAST力学指标与缝纫性能之间的关系,运用神经网络的方法进行面料缝纫性能评价和预测,并提出了一种基于模糊核聚类的优化径向基神经网络学习算法.经过大量实验,该系统对一般丝型织物的缝纫平整性能评价预测具有客观、快速和高效等优点.According to the relations between FAST mechanical index of silk fabric and sewing properties,the paper proposed an objective evaluation system of silk fabric＇s sewing properties based on optimized Fuzzy kernel clustering and RBF neural network.The experimental results demonstrate that the proposed system could efficiently be used as an objective prediction system of sewing properties with high accuracy.

关 键 词：FAST系统 缝纫平整度 模糊核聚类 RBF神经网络 客观评价 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]