涡轮机叶片同步振动参数辨识方法研究(建模仿真)

摘要

A new method used to identify a turbine's blade synchronous vibration parameters based on the blade tip-timing (BTT) non-contact type blade vibration measurement technique and previous studies in this field was proposed here.According to the measured relationship between blade vibration displacement and rotating speed,the parameters,such as,the center frequency of blade's harmonic resonance points,the maximum amplitude of blade vibration,and its initial phase were computed with the nonlinear least square fitting algorithm.If the blade assembly was operated at a constant rotating speed being close to the frequency of its resonance point,different blade vibration displacements were measured with sensors at different positions.The double-frequency vibration displacements of various blades at their corresponding harmonic resonance points were computed with the global autoregressive with instrumental variables (GARIV) method.A blade's Campbell diagram was made with its vibration parameters obtained using GARIV method and the sweep frequency fitting method.Based on this method,the locating scheme of BTT sensors was proposed,interference factors existing in the actual measurements were analyzed in detail,the feasibility of the proposed method was verified.The test verification of this method will be published later.%基于叶尖定时非接触式叶片振动测量技术,提出一种需要较少传感器又能提高测量精度的叶片同步振动参数分析方法.通过低转速到高转速运行过程,测量出叶片振动位移随转频变化之间的关系,根据非线性最小二乘拟合算法获得叶片谐共振点处的中心频率、振动最大幅值、相位、振动恒偏量,但无法获得叶片振动倍频值.对谐共振动点附近处的频率值进行恒速运转,不同位置处的传感器获得不同的振动位移,通过GARIV算法获得各叶片在对应谐共振频率点处的倍频值,结合扫频拟合获得的叶片振动参数可做出叶片坎贝尔图.基于以上的识别方法,提出了叶尖计时传感器的布置方法,并对实际测量中可能存在的干扰因素进行了详细的仿真分析,验证了方法的可行性.该辨识方法的实验验证将在《涡轮机叶片同步振动参数辨识方法研究(实验研究)》一文中给出.