纺织工业中的虚拟加工技术与模式

简要介绍了纺织品加工过程、人工神经网络（ANN）及其相关算法的特征。通过ANN技术建立的原料、纺纱、织造和后整理预测／反演模型，能够优化生产工艺，预测与控制产品质量，是纺织设计与虚拟加工的基础。采用主因子、聚类、案例模板和ANN等算法完成对输入参数的归纳、筛选与增补，是提高预测／反演模型精度和效率的有效步骤。以此构建的模块组合式虚拟加工系统，对纺织工业的快速、准确和理性加工，纺织品的低成本和高质量实现，具有重要意义。

Virtual Manufacturing Technology and Models for Textile Industry

The textile processing,artificial neural network (ANN)and other relevant algorithms are briefly introduced. Based on ANN,prediction/inversion models are set up respectively for material qualifying,spinning,weaving and finishing processes,which can be effectively used for optimiz- ing manufacturing technology and parameters,predicting and controlling product quality,laying foundation for textile designing and virtual manufacturing.Through the selection and supplement of the corresponding parameters for the prediction/ inversion ANN models by means of main-factor,or clustering, or case-based analysis,or ANN,the input parameters are screened out and optimized,so as to improve the efficiency and accuracy of calculation.The modular-assembled virtual processing system thus constructed is vital important for the textile industry to realize quick-response,accurate and rational processing and to reduce production costs,improve product uualitv and shorten manufacturing cycle.