基于灰色理论和时间序列模型的润滑油中磨粒含量预测分析  

Wear Particle Content Prediction Analysis in Lubricant Based on Gray Theory and Time Series Model

作　　者：陈志伟[1] 米东[1] 徐章遂[1] 刘斌[2] 

机构地区：[1]军械工程学院电气工程系 [2]中国人民解放军71320部队80分队

出　　处：《润滑与密封》2007年第5期147-149,共3页Lubrication Engineering

摘　　要：目前机械设备润滑油中磨粒含量的预测一般用简单线性拟合的方法,或者利用单一的预测模型,预测精度较低。针对润滑油中磨粒含量预测时存在的问题,提出了一种综合运用灰色预测理论和时间序列模型的方法对磨粒含量进行预测。该方法充分运用灰色预测能够反映磨粒含量变化的总体趋势和时间序列分析模型可以很好地进行细节分析的优点。将该方法应用到某型自行火炮润滑油分析数据中,证明了该方法的有效性。At present, the simple linear fitting method or the single model is usually applied in the prediction of the wear particle content in lubricant, so the prediction precision is lower. Aiming at the problems of the prediction precision, gray theory and time series model were applied together in the prediction. The rough trend of the wear particle content change can be reflected through gray theory, and the detail of the change can be reflected through time series model, so the method can make a good prediction result in theory. The prediction method was applied in the lubricant analysis data of some self- propelled gun,which proves that the method is effective.

关 键 词：灰色理论 时间序列模型 润滑油 

分 类 号：TH117[机械工程—机械设计及理论]