基于数据挖掘的舰务器材消耗规律预测  

Forecasting of Consumption Law for Equipment and Material for Naval Ship Service Based on Data Mining

作　　者：赵平[1] 刘宝平[1] 黄栋[1] 

机构地区：[1]海军工程大学装备经济管理系,湖北武汉430033

出　　处：《装备制造技术》2014年第7期66-69,共4页Equipment Manufacturing Technology

基　　金：2011年国家社科基金军事学项目资助(编号:11GJ003-072)

摘　　要：舰务器材的影响因素繁多,消耗规律复杂,在实际预测工作中,很难进行较为精确地预测。以涂料为例,提出了一套基于数据挖掘的消耗规律预测方法。运用粗糙集属性化简技术降低数据维数,将化简结果作为输入建立支持向量机回归模型,解决了小样本非线性拟合效果不理想的缺陷,较真实地反映出涂料的消耗规律。The equipment and material for naval ship service are often affected by various factors. Thus the consumption law is so complicated that it is hard to forecast correctly in reality. Taking paint for example, a way to forecast consumption law based on data mining is proposed. Rough sets theory was applicator to simplify the influence fator, then the simplified resuhs were put into the SVM to carry out training and forecasting,which solved the problem that small sample is inappropriate for nonlinear fitting and recovered the consumption law for paint factually.

关 键 词：数据挖掘 消耗规律 舰务器材 预测 涂料 

分 类 号：U668.33[交通运输工程—船舶及航道工程]