基于支持向量自回归的大型离心式水泵振动状态预测研究  

Study on Large-scale Centrifugal Pump Vibration Forecast Based on Support Vector Autoregressive

作　　者：吴兴伟[1] 

机构地区：[1]沈阳工程学院能源与动力工程系,沈阳110136

出　　处：《节水灌溉》2010年第4期52-55,共4页Water Saving Irrigation

基　　金：水利部"948"科技创新项目(CT200516);辽宁省教育厅科技攻关项目(05L385)

摘　　要：为了预测大型离心式水泵在运行中的振动状态,提高其运行的安全性和经济性,采用了统计学习理论中的核心算法支持向量机建立了大型离心式水泵振动预测模型(SVAR)。并通过实例,与基于灰度方法建立的预测模型(GM)和基于自回归方法建立的预测模型(AR)进行了比较。结果表明:基于支持向量自回归的大型离心式水泵振动预测模型具有精度高、速度快、易于建模的特点。应用该方法建立的预测模型能够很好地预测大型离心式水泵运行中的振动情况,有效地避免其在运行中由振动引起的故障,提高运行的安全性和经济性。In order to forecast the large-scale centrifugal pump operating vibration and increase its operating safety and economy,the kernel algorithm of Statistical Learning Theory（SLT）,namely Support Vector Machine（SVM）,is applied to set up a forecast model for large-scale centrifugal water vibration.The model based on SVAR is compared with the model based-on Gray Model and the model based-on Autoregressive by a case.The result indicates that the forecast model for Large-scale Centrifugal pump vibration based-on SVAR has many advantages,such as high-precision,high calculation velocity,modeling easily.The model based on SVAR can forecast the vibration condition of operating large-scale centrifugal pump,and avoid the fault due to vibration.

关 键 词：支持向量机 统计学习理论 大型离心式水泵 振动预测 

分 类 号：TK262[动力工程及工程热物理—动力机械及工程]