摘要
将投影寻踪分类(PPC)模型与基于实数编码的加速遗传算法(RAGA)相结合,同时优化多个指标参数,将高维数据指标转化为低维空间上的一维投影值,建立RAGA-PPC模型,用于造纸纤维原料分类,并对造纸纤维原料进行综合评价。结果表明,基于RAGA-PPC模型的评价结果与造纸纤维原料实际分类结果一致,此方法客观可靠,精度高,具有一定的应用前景。
In this paper,the projection pursuit classification(PPC)and accelerated genetic algorithm(RAGA)based on real coding were combined and optimized multiple index parameters to convert high-dimensional data indexes into one-dimensional projection.Based on the RAGA-PPC model,the various papermaking raw materials were classified,and evaluated effectively.The conclusion demonstrated that evaluation result of RAGA-PPC model was consistent with the actual category of the papermaking raw materials.Furtherly this method has more advantages in objectivity,reliability,accuracy and has practical application prospects.
作者
赵静远
熊智新
梁龙
房桂干
ZHAO Jingyuan;XIONG Zhixin;LIANG Long;FANG Guigan(Jiangsu Provincial Key Lab of Pulp and Paper Science and Technology,Nanjing Forestry University,Nanjing,Jiangsu Province,210037;Institution of Chemical Industry of Forestry Products,CAF,Nanjing,Jiangsu Province,210042)
出处
《中国造纸学报》
CAS
CSCD
北大核心
2020年第3期53-58,共6页
Transactions of China Pulp and Paper
基金
中国林科院林业新技术所基本科研业务费专项资助(CAF,基金号:CAFYBB2019SY039)。
关键词
RAGA算法
PPC模型
分类
综合评价
RAGA algorithm
PPC model
classification
comprehensive evaluation