摘要
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。
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.
出处
《纺织导报》
CAS
北大核心
2005年第7期10-10,12-16,22,共7页
China Textile Leader
基金
国家技术创新项目(02CJ-14-05-01)