基于灰色线性回归组合模型的故障率预测  

Failure Rate Prediction Based on the Grey Linear Regression Model

作　　者：邵延君[1] 潘宏侠[1] 马春茂[2] 刘永姜[1] 

机构地区：[1]中北大学机械工程与自动化学院,太原030051 [2]西北机电工程研究所,咸阳712099

出　　处：《振动．测试与诊断》2014年第4期664-667,776,共4页Journal of Vibration,Measurement & Diagnosis

基　　金：国家自然科学基金资助项目(51175480);山西省自然科学基金资助项目(2012011046-12)

摘　　要：针对武器装备在使用过程中故障率变化的复杂性,提出了灰色线性回归组合模型的故障率预测方法。该模型用线性回归方程和指数方程的和来拟合故障率曲线,它可以改善线性回归模型中没有指数增长趋势和GM(1,1)模型中没有线性因素的不足。通过对装备备件故障率的预测分析表明,灰色线性回归组合模型在故障率预测精度上优于单一的灰色模型和线性回归模型,且不要求对所提供的历史数据具有典型的分布规律。该模型的预测结果可以为装备的维修工作提供决策依据。Aiming at the complexity of the failure rate variety in weapon usage,a failure rate prediction method using agrey linear regression model(GM)is proposed.The model uses the sum of the linear regression equation and exponential equation to fit the failure rate curve.The deficiency caused by the lack of an exponential growth trend in the linear regression equation and the lack of linear factors in GM(1,1)can be improved.According to the analysis and prediction of the spare equipment,the results show that the grey linear regression model is superior to both the individual GM model and the linear regression model in the prediction accuracy of the failure rate.Moreover,the historical data does not require a typical distribution.The prediction results of the model can thus be regarded as the decision-making basis for equipment maintenance.

关 键 词：GM(1 1)模型 线性回归模型 灰色线性回归组合模型 故障率 

分 类 号：TH318[机械工程—机械制造及自动化]