基于灰色线性回归组合模型的集成电路封装设备故障趋势预测  

Fault Trend Prediction of Integrated Circuit Packaging Equipment Based on Gray Linear Regression Model

作　　者：张燕龙 陈亮希 陈兴玉[1] 郭磊 张红旗[1] 黄魁 ZHANG Yanlong;CHEN Liangxi;CHEN Xingyu;GUO Lei;ZHANG Hongqi;HUANG Kui(No.38 Research Institute of CETC,Hefei 230088,China;School of Mechanical Engineering,Southeast University,Nanjing 211189,China)

机构地区：[1]中国电子科技集团公司第三十八研究所,安徽合肥230088 [2]东南大学机械工程学院,江苏南京211189

出　　处：《机械与电子》2021年第3期20-23,共4页Machinery & Electronics

摘　　要：针对集成电路封装打孔机冲针监测数据的特点,分别构建灰色GM(1,1)模型和线性回归模型,开展故障趋势预测;在此基础上,采用组合预测的思想,运用灰色关联度方法融合灰色GM(1,1)模型和线性回归模型,建立灰色线性回归组合模型对设备进行故障趋势预测。结果表明,灰色线性回归组合模型的预测精度优于单一预测模型,可以用于集成电路打孔机设备的故障趋势预测。According to the characteristics of the monitoring data of punching machine for IC packaging,grey GM(1,1)model and linear regression model were constructed respectively to carry out fault trend prediction.On this basis,the grey linear regression combination model is established by combining grey GM(1,1)model and linear regression model with the idea of combination prediction.The results show that the prediction accuracy of gray linear regression combination model is better than that of single prediction model,and it can be used to predict the failure trend of IC puncher.

关 键 词：集成电路封装设备 故障趋势预测 灰色模型 线性回归模型 

分 类 号：TP206.3[自动化与计算机技术—检测技术与自动化装置]