缝纫机缝纫性能测量评价系统的设计  

Design of testing-and-evaluating system for sewing performance of sewing machines

作　　者：林君焕[1,2] 张国庆[2,3] 张树有[3] 

机构地区：[1]宁波大学信息科学与工程学院,浙江宁波315211 [2]台州职业技术学院机电学院,浙江台州318000 [3]浙江大学机械工程学系,浙江杭州310027

出　　处：《纺织学报》2013年第1期116-121,共6页Journal of Textile Research

基　　金：浙江省自然科学基金资助项目(Y1080974)

摘　　要：针对缝纫机缝纫性能评价难的问题,设计了一种基于BP神经网络的缝纫机缝纫性能客观测量评价系统。在缝纫机机械性能与缝纫性能之间构建了1个3层神经网络模型,该神经网络模型由4个子网络模型构成。分别负责4个缝纫性能指标的评价,同时研究了面料可缝性、缝纫性能、面料缝纫质量三者之间的量化关系。提出样本采集方案,设计了一套基于VB与MatLab的应用软件和基于USB数据采集卡的测量系统,在此基础上进行了实验研究。结果表明,BP神经网络在给定的样本下学习收敛效果较好,对被测缝纫机缝纫性能的评价与实际结果一致。A testing-and-evaluating system based on BP-neural network is designed to deal with the difficulty in evaluating sewing performance of sewing machines.A neural network with three levels composed of four sub-networks which are used to evaluate the four sewing performance indexes respectively is modeled between mechanical properties and sewing performance,and the quantized relationship between sewing performance,sewability and sewing quality of fabrics is researched.The sample collecting method for BP-learning is presented and a set of application software base on VB and MatLab and testing system based on USB-DAQ card are designed,and then experiments are performed using the designed system.Test results show that BP-neural network achieves satisfactory learning effect under a given sample condition,and its evaluated results are in good agreement with actual values.

关 键 词：缝纫性能 测量评价系统 BP神经网络 缝纫机 

分 类 号：TS941[轻工技术与工程—服装设计与工程] TP206.1[自动化与计算机技术—检测技术与自动化装置]