Research On Infrared Radiation Temperature Measurement Method Of Turbine Blades Based On Effective Emissivity

Turbine blades are important components of aviation engines and gas turbines.Accurate measurement of turbine blade temperature is of great significance for ensuring safe operation and health monitoring of engines.Infrared radiation temperature measurement technology is considered the best method for temperature measurement of new generation turbine blades due to its advantages of non-contact,high heat resistance,fast response speed,and wide temperature measurement range,However,the presence of a large number of high-temperature environments around the tested blades,represented by hot end components,can generate strong reflected radiation interference,leading to significant errors in temperature measurement results.This article focuses on the key issues of reflected radiation analysis and calculation in multispectral radiation temperature measurement of turbine blades in high-temperature environments,as well as the correction of reflected radiation temperature measurement errors.The main tasks completed are as follows:Aiming at the analysis and calculation of reflected radiation generated on the surface of the blade to be measured under high temperature environment,an analysis model of reflected radiation transfer of turbine blade was constructed.First,a three-dimensional discrete model of turbine blade was established by using discrete triangular panel division,which transformed the calculation of reflected radiation between non isothermal and irregular shaped blades into the calculation of radiation transfer between isothermal and regular shaped discrete blade panels;Secondly,by utilizing visual screening between blade surface elements,radiation transfer path occlusion judgment,and radiation angle coefficient calculation,the radiation transfer relationship between the high-temperature environment surface elements and the tested surface elements is solved,achieving the radiation angle coefficient calculation and reflection radiation analysis calculation of the tested blade in the high-temperature environment.Aiming at the problem of recognition and classification of multispectral emissivity models under the interference of reflected radiation,firstly,a calculation method of emissivity spectral shape classification features based on alpha spectrum LM(Levenberg Marquardt)algorithm was proposed,and the multispectral radiation equation was solved by using alpha spectral analysis and LM algorithm.The spectral shape features that are consistent with the real emissivity normalization are solved as the actual input of the classifier to solve the classification feature calculation problem of the multi spectral emissivity model under the influence of reflected radiation;Secondly,a emissivity model classifier based on particle swarm optimization least squares support vector machine(PSO-LSSVM)is designed,and the regularization parameters and Gaussian radial basis function kernel width parameters of the classifier are optimally selected by using particle swarm optimization algorithm to realize the recognition and classification of multispectral emissivity models under high temperature and strong reflection environment.Simulation experiments show that the accuracy of emissivity model recognition and classification can reach 93.0%.Aiming at the problem of error correction of turbine blade reflected radiation temperature measurement under high temperature environment,firstly,combined with reflected radiation analysis and calculation and emissivity model recognition and classification,a turbine blade multispectral infrared radiation temperature measurement algorithm based on effective emissivity is proposed,the constraint relationship between effective emissivity and blade temperature is determined,and the target equation of multispectral radiation temperature measurement is established;Secondly,a multi-spectral radiation data processing algorithm based on the dual population social group optimization algorithm(DPSGO)is proposed to solve the objective equation and achieve true temperature inversion of turbine blades.Finally,simulation comparative analysis and experimental verification were conducted on the proposed method and key steps.The experimental results showed that the maximum temperature measurement error of the radiation temperature measurement method proposed in this paper was 7.8K,and the average temperature measurement error was 2.7K.This can effectively correct the reflection radiation temperature measurement error and meet the radiation temperature measurement requirements of turbine blades under high temperature and strong reflection background.