论文部分内容阅读
为提高纺织面料成分的检测水平和分析速度,以134个毛涤混纺织物为研究对象,自行设计采样装置,采集其近红外光谱;然后,经一阶导数等光谱预处理方法,结合偏最小二乘法,建立了毛涤混纺织物中毛含量近红外的定量校正模型。为验证模型的实用性,对129个毛涤样品进行了预测,将预测结果进行方差分析,得到该方法与标准SN/T 0464-2003方法的结果不存在显著性差异的结论,并据此制定了纺织品原料组分近红外快速检测初筛方法的标准操作规程,经实际应用证明效果显著。
In order to improve the testing level and analysis speed of the textile fabrics, 134 wool-polyester blended fabrics were selected as the research object, and the sampling device was designed by ourselves to collect the near-infrared spectra. Then, by the first-order derivative and other pretreatment methods, Multiplication method was used to establish a quantitative correction model of wool content in wool and polyester blended fabrics. In order to verify the practicability of the model, 129 samples of wool and polyester were predicted, and the results were analyzed by ANOVA and the conclusion that there was no significant difference between the method and the standard SN / T 0464-2003 method was obtained The standard operating procedure of the near-infrared rapid testing method of raw materials for textile raw materials has been proved to be effective by practical application.