摘要
汽封是汽轮机里控制漏汽量的重要部件,汽封中的汽流力也是引起汽流激振的主要因素之一。传统的汽封动力特性系数计算主要为有限元法和有限差分法等。然而这些方法始终存在计算精度和网格划分、计算复杂程度的矛盾。提出基于径向基函数神经网络(RBFNN)的汽封动力特性系数计算模型,并以某厂660MW超超临界机组高压轴封为研究对象,建立该轴封动力特性系数的RBFNN计算模型。计算结果表明,该模型可大大提高计算效率,计算精度较高,满足工程计算的要求,为电厂快速、实时分析汽封中的汽流力对轴系的稳定性影响提供了方便。
Steam seal is an important part of steam turbine to control the leakage of steam.The flow force in steam seal is one of the main factors that cause steam flow excitation.The caculation of dynamic characteristics of the steam seal mainly relies on finite element method and finite difference method which always have the contradiction between the computational accuracy and the complexity.In this paper,the model based on the radial basis function neural network(RBFNN)of the dynamic characteristics of the steam seal is presented.The high pressure seal of a660MW unit of a factory is taken as the research object and the calculation results show that the model has high accuracy and can greatly improve the computational efficiency.It provides convenience for real-time analysis and calculation of characteristics of steam seal and rotor stability in power plants.
作者
邓敏强
傅行军
DENG Min-qiang;FU Xing-jun(National Engineering Research Center of Turbo-generator Vibration, Southeast University, Nanjing 210096,China)
出处
《汽轮机技术》
北大核心
2018年第1期59-62,65,共5页
Turbine Technology
关键词
汽封
径向基函数
动力特性系数
电厂实时分析
汽流激振
sealing
radial basis function
dynamic characteristics
real time analysis of power plant
excited vibration
