Research On Turbine Blade Temperature Field Reconstruction And Temperature Distribution Characteristics

As the most efficient and widely used thermal power devices,gas turbines have an important role in many fields such as aerospace,aviation,ship and energy.Turbine blade is one of the highest working load components in the process of gas turbine operation.As it is working in high temperature,high pressure and high speed rotating state for a long time,its surface coating is easy to falling off,defecting and fracturing,which will not only affect the service life of gas turbine,but also causing potential safety hazards and even happening accidents.Therefore,temperature measurement and analysis of turbine blade are necessary for the development of gas turbines.The way to get the temperature characteristics is not only difficult but also important in the research of turbine blades.The processing technology,cooling performance and service life of blades can be analyzed through the temperature distribution characteristics of blade surface.Computational Fluid Dynamics(CFD)is used to simulate the temperature field of turbine blade in my research.The blade model,gas volume,temperature and pressure are set up to simulate the blade operating state,so we can get the temperature field data of blade surface.According to the obtained blade model,an improved method of radiation temperature measurement is proposed,which can realize the full scanning of blade surface.The radiation pyrometer collects the temperature data on a straight line from the leading edge to the trailing edge of the blade in the axial direction,and can not accurately reflect the temperature distribution of the blade surface.So an improved scheme for the probe part of the radiation thermometer is proposed.Controlling the rotation of the rotating mirror,the rotation and expansion of the probe by stepping motor,the probe can move both in the axial and radial direction,and the scanning area can cover the entire blade and reconstruct the temperature field.According to the obtained temperature field data,drawing the normalized temperature curves of suction surface and pressure surface,analyzing the temperature distribution characteristics with the environmental factors such as blade shape,curvature and gas flow direction.In order to research the uniformity of temperature distribution of turbine blades in the same cycle,adopting the method of feature extraction and comparison.The dimensionless index is extracted as the time-domain feature,and the frequency-band energy feature after wavelet packet decomposition is taken as the frequency-domain feature.Both of them constitute the eigenvector representing the temperature distribution information of the blade.Finally,the principal component eigenvector is obtained by principal component analysis(PCA),which is linearly reconstructed to remove the correlation.The mean value of temperature distribution characteristic vectors of all blades in the same rotation period is taken as the reference characteristic vectors of temperature distribution in that period.The temperature distribution characteristic vectors of each blade are compared with the reference characteristic vectors and the relative deviation between them is calculated.The consistency of the temperature distribution of the blades can be judged according to the deviation value.The research results can be used to observe the blade working state from the angle of temperature characteristics,and provide a research basis for blade fault diagnosis and life prediction.