基于IOWGA算子的转子系统磨粒浓度组合预测方法
|
摘要
为了准确地得到转子系统滑动轴承磨损的预测结果,基于IOWGA算子的组合预测方法的理论,提出一种最小二乘支持向量机回归预测、灰色预测及指数平滑法相结合的组合预测方法,建立一种新的预测转子系统滑动轴承磨损磨粒浓度的组合预测模型及其评价指标体系。在单跨双圆盘转子试验台上进行试验,提取铁谱片上滑动轴承磨损的浓度数据,利用各项预测模型对磨损磨粒浓度变化趋势进行预测,并比较各项预测模型的预测结果。实例结果表明:基于IOWGA算子的组合预测模型的预测精度较高,是预测润滑油中磨粒浓度变化趋势的一种有效方法。
In order to get accurately wear prediction results of the sliding bearing in rotor system,based on theory of the Induced Ordered Weighted Geometric Averaging(IOWGA) operator combination forecasting method,a combination forecasting method combining the least squares support vector machine regression,the grey forecast and the exponential smoothing method was proposed,and a combined forecasting model of wear particle concentration of a rotor system and the evaluation index system were established.The experiment was carried out in the single span double disc rotor test rig,the concentration data of the wear debris of the sliding bearing on the iron spectrum were extracted,the wear particle concentration change trend was predicted by combination forecasting model,and the forecast results of the forecast model were compared.The results show that the combined forecasting model based on IOWGA operator posses higher prediction accuracy than the three single forecasting methods,which is an effective method to predict the particle concentration change trend in lubricating oil.
In order to get accurately wear prediction results of the sliding bearing in rotor system,based on theory of the Induced Ordered Weighted Geometric Averaging(IOWGA) operator combination forecasting method,a combination forecasting method combining the least squares support vector machine regression,the grey forecast and the exponential smoothing method was proposed,and a combined forecasting model of wear particle concentration of a rotor system and the evaluation index system were established.The experiment was carried out in the single span double disc rotor test rig,the concentration data of the wear debris of the sliding bearing on the iron spectrum were extracted,the wear particle concentration change trend was predicted by combination forecasting model,and the forecast results of the forecast model were compared.The results show that the combined forecasting model based on IOWGA operator posses higher prediction accuracy than the three single forecasting methods,which is an effective method to predict the particle concentration change trend in lubricating oil.
引文
[1]孙虎儿,杨兆建,梁群龙,等.突变扭矩激励下转子系统横振响应[J].振动、测试与诊断,2011,31(5):622-625.SU H E,YANG Z J,LIANG Q L,et al.Transverse vibration response of rotor system under mutational torque excitation[J].Journal of Vibration,Measurement&Diagnosis,2011,31(5):622-625.
[2]张鄂.铁谱技术及其工业应用[M].西安:西安交通大学出版社,2001.
[3]李霜,杨晓京,郭志伟.基于LS-SVM的柴油机润滑油中磨粒含量预测[J].润滑与密封,2009,34(2):46-48.LI S,YANG X J,GUO Z W.Forecasting of wear particle concentration in diesel engine lubricating oil by least squares support vector machine[J].Lubrication Engineering,2009,34(2):46-48.
[4]周伟,景博,邓森.基于IGA和LS-SVM航空发动机磨粒识别[J].润滑与密封,2013,38(1):14-18.ZHOU W,JING B,DENG S.Wear particle pattern identification of aeroengine based on LS-SVM and improved genetic algorithm[J].Lubrication Engineering,2013,38(1):14-18.
[5]张彦.制动器摩擦衬片磨损量的等维灰色预测[J].润滑与密封,2009,34(2):30-33.ZHANG Y.The application of grey equidimensional prediction model in the wear of car braking friction disk[J].Lubrication Engineering,2009,34(2):46-48.
[6]吴越,王玫,陈勇.线性回归模型诊断和在线预测刀具磨损量的方法研究[J].机械设计与制造,2009(6):236-238.WU Y,WANG M,CHEN Y.A methodology of tool wear diagnostic using Linear model and prognostic using double exponential smoothing[J].Machinery Design&Manufacture,2009(6):236-238.
[7]陈华友.组合预测方法有效性理论及其应用[M].北京:科学出版社,2008.
[8]陈华友,盛昭瀚.一类基于IOWGA算子的组合预测新方法[J].管理工程学报,2005,19(4):36-39.CHEN H Y,SHENG S H.A kind of new combination forecasting method based on induced ordered weighted geometric averaging(IOWGA)operator[J].Journal of Industrial Engineering and Engineering Management,2005,19(4):36-39.
[2]张鄂.铁谱技术及其工业应用[M].西安:西安交通大学出版社,2001.
[3]李霜,杨晓京,郭志伟.基于LS-SVM的柴油机润滑油中磨粒含量预测[J].润滑与密封,2009,34(2):46-48.LI S,YANG X J,GUO Z W.Forecasting of wear particle concentration in diesel engine lubricating oil by least squares support vector machine[J].Lubrication Engineering,2009,34(2):46-48.
[4]周伟,景博,邓森.基于IGA和LS-SVM航空发动机磨粒识别[J].润滑与密封,2013,38(1):14-18.ZHOU W,JING B,DENG S.Wear particle pattern identification of aeroengine based on LS-SVM and improved genetic algorithm[J].Lubrication Engineering,2013,38(1):14-18.
[5]张彦.制动器摩擦衬片磨损量的等维灰色预测[J].润滑与密封,2009,34(2):30-33.ZHANG Y.The application of grey equidimensional prediction model in the wear of car braking friction disk[J].Lubrication Engineering,2009,34(2):46-48.
[6]吴越,王玫,陈勇.线性回归模型诊断和在线预测刀具磨损量的方法研究[J].机械设计与制造,2009(6):236-238.WU Y,WANG M,CHEN Y.A methodology of tool wear diagnostic using Linear model and prognostic using double exponential smoothing[J].Machinery Design&Manufacture,2009(6):236-238.
[7]陈华友.组合预测方法有效性理论及其应用[M].北京:科学出版社,2008.
[8]陈华友,盛昭瀚.一类基于IOWGA算子的组合预测新方法[J].管理工程学报,2005,19(4):36-39.CHEN H Y,SHENG S H.A kind of new combination forecasting method based on induced ordered weighted geometric averaging(IOWGA)operator[J].Journal of Industrial Engineering and Engineering Management,2005,19(4):36-39.