Fatigue Life Prediction And Fatigue Reliability Analysis Of Aero-engine Turboshaft

Aero-engine is the power source of aircraft flight.The condition of the engine will directly affect the performance of the aircraft.The rotating parts of the engine are the key parts for the normal operation of the engine.The quality of the rotating parts will directly affect the working state of the engine.Therefore,in order to better develop the technical level of the engine,it is necessary to deeply study the structure of the rotating parts and the factors affecting the service life.The rotating parts of an engine are composed of turbine blades,turbine disks,turbine shafts and bearings,etc.Among them,the turbine shafts are an important part of the rotor system,whose main function is to transfer power and circulate working loads and support other parts.When running on the aircraft,the engine is running at high speed.At this time,the turbine shaft is subject to a variety of alternating loads.Under such a complex working environment,it is easy to be damaged.Once the turbine shaft fails,the engine will not work normally and the supply of flight power will be blocked,which may cause a serious plane crash.Therefore,in order to ensure the safe and reliable operation of aero-engine,it is necessary to study the fatigue life of turbine shaft in depth.In this thesis,the fatigue reliability of turbine shaft of a certain aero-engine is studied,and the research contents are as follows:(1)The structural characteristics and load bearing of the turbine shaft are analyzed,the constraints and load conditions of the turbine shaft in the working process are defined,and the possible failure modes of the turbine shaft are analyzed,so as to have a comprehensive understanding of the turbine shaft.The turbine shaft model is imported into ANSYS Workbench,and the finite element simulation analysis is carried out on the turbine shaft to obtain the stress-strain distribution of each part of the turbine shaft under various working conditions,and the dangerous parts of the turbine shaft were found out.(2)In this thesis,the related theories of high and low cycle composite fatigue life prediction are summarized to understand the existing composite life prediction methods.The damage mechanics method which can describe the damage evolution process during fatigue is selected to solve the problem.First of all,the theory of damage mechanics is introduced from the concept of damage variable,principle of strain equivalence and law of damage evolution.Then,a high and low cycle fatigue damage model is established based on damage mechanics theory.Finally,the material parameters involved in the model are fitted and solved,and the feasibility of the model is verified.(3)Fatigue reliability analysis methods are briefly introduced,and the advantages and disadvantages of each method are analyzed,so the neural network method is derived.Through the learning of neural network model,I have a preliminary understanding of the structure of neural network.I study several common neural networks in depth.I understand the advantages and disadvantages of each network.The back propagation neural network and orthogonal basis neural network are selected to analyze the reliability of the turbine shaft respectively.I select the nine random variables as neural network input layer,the fatigue life of turbine shaft for the output layer.The Monte Carlo simulation method is used to sample in ANSYS and solve the data obtained by MATLAB to train the neural network,and then 10,000 sets of data obtained by Monte Carlo simulation are input into the trained network to calculate the corresponding life,so as to realize the fatigue reliability analysis of the turbine shaft.