| As the global aviation industry continues to develop rapidly,aircraft engines are playing an increasingly important role in modern aviation technology.The research and development level of aviation engines reflects a country’s technological innovation level and national economic level,and is also related to a country’s national defense security and industrial competitiveness.Therefore,it has become a key research and development target for various industrial powers in the world.With the development of aviation technology,the performance improvement of aeroengines has also brought many new challenges.under extreme conditions such as high speed,high temperature,and high load,how to ensure the stability and reliability of aeroengines is a critical issue.Turbine shaft is one of the important components of an aeroengine,which connects turbine,compressor,and other components,and transmits power to generate thrust for the engine.Under working conditions,turbine shafts are exposed to extreme environments for a long time,which can easily lead to fatigue failure and serious safety accidents.Conducting fatigue life research on turbine shafts can ensure the reliability and safety of aircraft engines,which is of great significance.Regarding the turbine shaft,there are various uncertainties that affect the prediction results during the study of its fatigue life,mainly including the uncertainty of the material properties,geometric dimensions,prediction models,and loads applied to the turbine shaft.Characterizing the distribution of fatigue life prediction process data and conducting hypothesis testing on the evaluated fatigue life distribution is of great significance for improving the accuracy and credibility of prediction results.This thesis conducts probability fatigue life prediction for a certain type of aviation engine turbine shaft,and the main research content is as follows:(1)Finite element simulation was conducted on the turbine shaft.A finite element three-dimensional model of the turbine shaft was established,simplified and imported into finite element analysis software.Boundary conditions were applied based on the material and load conditions of the turbine shaft,as well as the assembly relationship with surrounding components.The dangerous parts of the turbine shaft and their stress-strain response were divided into grids and solved.The obtained finite element analysis results of the turbine shaft were consistent with the experimental results.(2)The random variables of the turbine shaft were characterized by distribution.Analyzed the uncertainty of the turbine shaft,identified multiple sources of uncertainty for the turbine shaft,identified the main random variables,and conducted distribution characterization research on their classification based on the characteristics of the sample data of each random variable.For random variables with small sample characteristics,MCMC simulation method was used to infer and estimate them based on Bayesian theory to obtain a more accurate posterior distribution.(3)Probabilistic fatigue life prediction was conducted on the turbine shaft.A modified average stress fatigue life prediction model was proposed,and its prediction accuracy was verified through material data and comparison with existing models’ prediction results.Latin hypercube sampling was used to obtain the random variable sample data of the turbine shaft,and the finite element simulation and fatigue life prediction model calculation were carried out in turn.The stress strain distribution and fatigue life data of the dangerous parts of the turbine shaft were obtained.(4)Hypothesis testing was conducted on the fatigue life distribution of the turbine shaft.The optimal distribution of turbine shaft fatigue life is determined by hypothesis testing methods such as Chi square,Jacques Bella,etc.The distribution parameters of turbine shaft fatigue life are obtained by Maximum likelihood estimation,and verified by K-S test. |
PDF Full Text Request