应用粗糙集和支持向量机的熔喷非织造布过滤性能预测  

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

作　　者：金关秀[1] 张毅[1] 楼永平[1] 祝成炎[2] JIN Guanxiu1, ZHANG Yi1. LOU Yongping1, ZHU Chengyan2(1. Zhejiang Industry Polytechnic College, Shaoxing, Zhejiang 312000, China;2. Zhejiang Sci-Tech University, Hangzhou, Zhejiang 310018, China)

机构地区：[1]浙江工业职业技术学院,浙江绍兴312000 [2]浙江理工大学,浙江杭州310018

出　　处：《纺织学报》2018年第6期142-148,共7页Journal of Textile Research

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。In order to predict the filtration performance of melt blown nonwoven fabric, 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 lower than 2%. This indicates that three parameters can be considered as key factors influencing 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.

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 

分 类 号：TS171[轻工技术与工程—纺织材料与纺织品设计]