【摘要】 为实现离心泵设备的状态监测与智能维护，论文基于机理建模与数据驱动建模相结合的思想建立了离心泵数字孪生属性模型。首先基于CFD建立离心泵数值计算模型，实现对离心泵水力属性与其内部流场物理属性的初步孪生。其次，针对数值计算模型的结果误差，基于数据驱动的方法对其进行误差补偿，进而实现对离心泵水力属性的精确孪生。最后，以某石化企业航空煤油输送离心泵为实验对象，利用其实际运行数据对论文所建数字孪生模型进行验证。验证结果表明论文所建离心数字孪生属性模型准确的实现了对于离心泵水力属性及物理属性的孪生。更多还原

【Abstract】 In order to realize the condition monitoring and intelligent maintenance of centrifugal pump equipment,this paper establishes the digital twin attribute model of centrifugal pump based on the idea of combining mechanism modeling with data driven modeling. Firstly,the numerical calculation model of centrifugal pump is established based on CFD to realize the preliminary twinning of the hydraulic properties of centrifugal pump and the physical properties of its internal flow field. Secondly,in view of the result error of the numerical calculation model,the error compensation is carried out based on the data driven method,so as to realize the accurate twin of the hydraulic properties of the centrifugal pump. Finally,the digital twin model is validated by the actual operation data of an aviation kerosene centrifugal pump in a petrochemical enterprise. The verification results show that the centrifugal digital twin attribute model established in this paper accurately realizes the twin of hydraulic and physical attributes of centrifugal pumps.更多还原