纺织学报 ›› 2018, Vol. 39 ›› Issue (06): 142-148.doi: 10.13475/j.fzxb.20170803507

• 管理与信息化 • 上一篇    下一篇

应用粗糙集和支持向量机的熔喷非织造布过滤性能预测

  

  • 收稿日期:2017-08-22 修回日期:2018-03-07 出版日期:2018-06-15 发布日期:2018-06-15

Prediction on filtration performance of melt blown nonwoven fabric based on rough set theory and support vector machine

  • Received:2017-08-22 Revised:2018-03-07 Online:2018-06-15 Published:2018-06-15

摘要:

为预测熔喷非织造布的过滤性能,提出基于属性约简和支持向量机的预测方法。运用粗糙集理论在ROSETTA 环境下对含有9 个参数的熔喷非织造纤网结构参数全集进行约简,得到6 个各含3 个参数的约简集。分别将参数全集及各个约简集作为输入建立基于支持向量机(SVM)和BP 神经网络(BP-ANN)的28 个过滤性能预测模型,运用交叉验证法进行模型结构参数优化。结果表明:以含厚度、纤维直径和孔径的约简集为输入,基于SVM模型预测准确度最高;其对过滤效率和过滤阻力的预测精度均超过98%,且CV 值均小于2%,表明这3 个参数是影响熔喷非织造布过滤性能的核心要素;基于SVM 模型的预测准确度总体优于基于BP-ANN模型的。

关键词: 熔喷非织造布, 纤网结构, 属性约简, 支持向量机, 交叉验证法, 过滤性能

Abstract:

In order to predict the filtration performance of melt blown nonwoven, a prediction method based on attribute reduction and support vector machine was introduced. Six reducts, each including three parameters, were extracted from the complete parameter set of fiber web structure of melt blown nonwoven fabric in ROSETTA environment using rough set theory. Twenty eight models, each based on either a support vector machine (SVM) or a back-propagation artificial neural network (BP-ANN), were established to predict the filtration performance by taking the parameters of each reduct and the complete parameter as inputs. A k-fold cross validation technique was applied to access the optimized structural parameters of the models. The results show that the prediction accuracy of the SVM-based model taking thickness, fiber diameter and pore as input parameters is higher than that of any other model. The values of its prediction precision for both filtration efficiency and pressure drop are higher than 98% and their variation coefficients are both lower than 2%. This indicates that these three parameters can be considered as key factors injluencing the filtration performances of melt blown nonwoven fabric. Generally, the prediction performance of SVM-based models are better than that of BP-ANN-based models.

Key words: melt blown nonwoven fabric, fiber web structure, attribute reduction, support vector machine, cross validation, filtration performance

[1] 李佳平 沈国康 欧耀明 孟想 辛斌杰. 应用连续投影算法及最小二乘支持向量机的单组分纺织品识别[J]. 纺织学报, 2018, 39(08): 46-51.
[2] 朱金铭 钱建华 曹晨 曹原. 聚醚砜非织造布复合膜的空气过滤性能[J]. 纺织学报, 2018, 39(07): 55-62.
[3] 于宾 赵晓明 漆东岳. 腈纶预氧化丝/芳纶针刺滤材的性能[J]. 纺织学报, 2018, 39(03): 61-66.
[4] 张陆佳 林兰天 陈春敏 申炎仃 高琮. 基于主成分分析的纤维拉伸断裂声发射信号识别[J]. 纺织学报, 2018, 39(01): 19-24.
[5] 陈锋 姬忠礼 齐强强. 静电纺聚丙烯腈纳米纤维复合滤材的制备及其气液过滤性能[J]. 纺织学报, 2017, 38(09): 8-13.
[6] 张建新 张银露 胡旭东. 光谱优化处理结合多层次支持向量机的混合染液浓度检测方法[J]. 纺织学报, 2017, 38(07): 90-94.
[7] 申莹 邓炳耀 刘庆生 夏赛男 姚鹏飞. 不同填充密度的梯度结构复合滤材的制备及其性能[J]. 纺织学报, 2017, 38(07): 23-27.
[8] 李东 万贤福 汪军. 采用傅里叶描述子和支持向量机的服装款式识别方法[J]. 纺织学报, 2017, 38(05): 122-127.
[9] 赵伟伟 汪滨 王娇娜 裴广玲 李从举. 静电纺聚酰胺6纳米纤维膜的制备及其性能[J]. 纺织学报, 2017, 38(03): 6-12.
[10] 余改丽 张弘楠 张娇娇 左晓飞 覃小红. 高效低阻聚丙烯腈/石墨烯纳米纤维膜的制备及其抗菌性能[J]. 纺织学报, 2017, 38(02): 26-33.
[11] 刘雷艮 潘志娟 牛建涛. 壳聚糖改性静电纺聚砜纤维膜对染料的过滤性能[J]. 纺织学报, 2016, 37(10): 1-7.
[12] 王敏 韩建 于斌 朱斐超 Solitario Nesti 宋卫民. 双组分橘瓣型纺粘水刺材料的过滤和力学性能[J]. 纺织学报, 2016, 37(09): 16-20.
[13] 陈孝之 谢莉青. 织物颜色配准到标准色卡的计算机识别与仿真[J]. 纺织学报, 2016, 37(05): 150-0.
[14] 任煜 李猛 尤祥银. 驻极处理对聚乳酸熔喷材料性能的影响[J]. 纺织学报, 2015, 36(09): 13-17.
[15] 刘雷艮 沈忠安 洪剑寒. 静电纺高效防尘复合滤料的制备及其性能[J]. 纺织学报, 2015, 36(07): 12-0.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!