Investigation On Method Of Turbine Blade Vibration Measurement Based On Data Fusion Method

Turbine machinery plays an extremely important role in defense and industrial production as a power plant.Blade is the core component of turbine machinery,which is commonly used in energy conversion.Monitoring the status of the blade is to ensure that the rotating equipment can be safe and stable operation.Blade tip timing(BTT)is an important method of non-contact blade vibration measurements.The main contents of this paper are as follows:1.This paper studied the effect of speed fluctuation on the accuracy of blade tip timing and corresponding solution.Based on the method of BTT,key phase interpolation method is introduced to solve the larger error caused by speed fluctuation with once per revolution probe,while it can also be adaptive under the constant speed operation.First,the simulation model was established to compare the results of the key phase interpolation method with the result of the single phase under different frequencies and speed fluctuations.Finally,tests were carried out by the twist blade test bench.The results showed that the accuracy of results were significantly improved by using key phase interpolation method when the fluctuation period was large or the amplitude of rotation speed fluctuation was big.2.The effects of axial displacement and radial vibration on the blade vibration measurement under the BTT method are discussed in detail.Aiming at the twist blade which is common in engineering application,the method correcting the time of the axial(ToA)displacement to the sensor is proposed to reduce the influence of axial displacement.By monitoring the rotor's axis trajectory,the rotor vortex is decomposed to obtain the positive and negative precession of the rotor,and the influence of the rotor vortex on the vane velocity is analyzed,and the concrete derivation formula is given.The accuracy of the axial displacement correction method was verified by twist blade test bench.The effect of the rotor vortex on the blade linear velocity is negligible,but it is still necessary to discuss the time at which the blade reaches the sensor.3.A new method used to identify the blade synchronous vibration parameters based on BTT method and previous research in the actual environment.According to the relationship between the rotation speed and the blade vibration displacement,the partial vibration parameters not include Engineer Order(EO)corresponding to the harmonic resonance of blade are obtained by fitting algorithm.The blade assembly is operated at constant which is near the frequency of resonance center.Different blade vibration displace can be obtained by the sensors at different position,and the blade synchronous vibration Engine Order(EO)can be obtained by Global Autoregressive with Instrumental Variables(GARIV)method.Based on the new identification method,the layout scheme of the blade tip sensor is determined,and the possible interference factors in the actual measurement are simulated and analyzed,and the feasibility of the method is verified.4.A high-speed straight blade test bench was established.In the experimental study,the parameter identification of synchronous vibration of blade is completed and the Campbell diagram of blade vibration is accurately plotted under the excitation of six magnets.Meanwhile,the experimental study and analysis on the harmonic vibration of blade with different numbers of excitation.The accuracy of the new identification method is verified by strain gauge measurement.This method can provide a technical approach for blade fault early warming and non-contact measurement of dynamic stress.