粘衬组合面料缝纫外观平整性能的预测模型

Prediction of the fused composites performance in tailored garments

分析了粘衬后组合面料的缝纫性能的变化趋势,并利用人工神经网络的方法建立了基于粘衬组合面料FAST力学性能指标的缝纫外观平整性客观评价模型,采用斯皮尔曼秩相关(Spearman′s rho)分析法确立模型评价指标.经检验,该模型整体预报精度较好,稳定性高,可用于粘衬织物缝纫外观平整度等级预测.

The changes of seam-pucker grades in fused composite was researched by experiment and the prediction model was set up to evaluate the fabric seam-pucker grade after fusing the interlining by means of artificial neural network system. Spearman＇s rho analysis method was used to get the evaluation indexes. After training, the best model was gained in which predictive precisions of all kinds of woven fabrics in the experiment is above 90 %.