原棉质量对纱线质量的鱼骨图-旋转森林预测模型  

Fishbone Diagram and Rotation Forest Predictive Model of Raw Cotton Quality for Yarn Quality

作　　者：雷英娜 

机构地区：[1]兴城市产品质量监督检验所,辽宁葫芦岛125100

出　　处：《工业技术创新》2015年第5期495-501,共7页Industrial Technology Innovation

摘　　要：原棉质量的影响因素与纱线质量的关系具有不确定性,针对该问题,综合应用相关分析法、鱼骨图理论及旋转森林原理构建原棉质量对纱线质量的预测模型。采用相关分析法及原因型鱼骨图模型分析原棉影响因素与纱线质量的关联度,计算各因素权重。用实测18组纺纱数据对影响因素加权的旋转森林模型进行训练,测试结果良好。用13组数据作为测试数据,进行量预测,并且与其他预测模型进行对比。研究结果表明:鱼骨图模型可获得因素与纱线质量的关系,量化输入变量的重要性;结合旋转森林预测模型能更好地考虑各因素对纱线质量的综合影响,与其他方法进行比较,验证了鱼骨图-旋转森林模型在纱线质量预测中具有较高的精确度。The relationship between factors of raw cotton quality and yarn quality is uncertain,so correlation analysis,fishbone diagram,and rotation forest algorithm are used to make the model of yarn quality prediction model from the perspective of quantitative analysis.Among them,correlation analysis and fishbone diagram are used to analysis the correlation degree of influence factors of raw cotton and yarn quality.Then the correlation degree is as the weight to predict by rotation forest algorithm,Eighteen groups of data are taken as training data.Additional thirteen groups as test data to verify it and compare with the other prediction method.The results show that the fishbone diagram can obtain the relationships among indexes and yarn quality,and the indexes＇ weights.The proposed method considering the mainfold factors can improve the prediction precision of raw cotton quality for yarn quality.Through compared with the other test data,the result shows that the model has a high accuracy.

关 键 词：纺纱质量 预测模型 鱼骨图 旋转森林 

分 类 号：TS11[轻工技术与工程—纺织材料与纺织品设计]