摘要
精确预测已成为制约装备器材保障决策的重难点问题。针对小样本消耗数据装备器材历史数据少、复杂性高、非线性等特点,利用灰色模型构建方便、计算简单、善于挖掘影响因素内部联系的优点,以及LS-SVM在非线性映射分析和稳定性高的优点,将两种方法进行组合,设计了一种灰色LS-SVM预测模型,用于解决小样本数据的装备器材需求预测问题。最后,以某建制单位小样本数据器材的需求预测为例,验证了该方法的可行性和优越性。
Accurate prediction has become an important and difficult problem in equipment support decision-making.In view of the characteristics of small sample consumption data,such as less historical data,high complexity and nonlinearity,this paper used the advantages of Grey model,such as convenient construction,simple calculation,being good at mining the internal relations of influencing factors,and the advantages of LS-SVM in nonlinear mapping analysis and high stability,combing the two methods to predict,and designed a Grey LS-SVM prediction model to solve the problem of equipment demand prediction based on small sample data.Finally,the feasibility and superiority of the method was verified by the demand forecast of small sample data equipment in an organic unit.
作者
贾琦
杨帆
王铁宁
JIA Qi;YANG Fan;WANG Tiening(Army Armored Academy,Beijing 100072,China;The Northern War Zone Army,Jinan 250000,China;The No.96901 st Troop of PLA,Beijing 100095,China)
出处
《兵器装备工程学报》
CAS
CSCD
北大核心
2021年第4期170-174,共5页
Journal of Ordnance Equipment Engineering
基金
军队总装科研项目。