| Aero-engine is an important component of modern aviation industry,which reflects a country’s comprehensive industrial level.Its performance directly affects the safe and reliable operation of aircraft.With the development of manufacturing industry,the position of aero-engines has become more prominent.It is of great significance to promote industrial technology innovation and development,improve industrial core competitiveness,and promote national defense construction,ensure national security.However,the performance improvement of aero-engine also faces many new challenges.As one of the critical components of aero-engine,turbine shafts mostly work under extreme working conditions such as high temperature,high pressure,and high rotational speed,and bear various loads,making it prone to fatigue failure,which can lead to serious safety accidents.Existing research shows that the failure of turbine shafts mainly occurs in the splines or keyways.Therefore,it is necessary to carry out optimization design method research on the weak links of turbine shafts to improve the overall performance of aero-engine.In addition,there are multiple sources of uncertainty factors such as geometry,material,and load in turbine shafts.Accurately quantifying uncertainty factors and establishing reasonable fatigue reliability analysis methods are prerequisites for turbine shaft optimization design.In this thesis,a certain type of aeroengine turbine shaft was taken as the research object,and fatigue reliability analysis and optimization design method research were carried out.The completed work is as follows:(1)The fatigue test conditions and results of a certain type of turbo-shaft were collected,and finite element analysis was carried out.An improved life prediction model was proposed and verified its feasibility with the test data of the same material as the turbo-shaft.A deterministic method for predicting the fatigue life of the turbo-shaft was established.(2)The multiple sources of uncertainty in the turbo-shaft were analyzed and quantified.The random variable sample data of the turbo-shaft were obtained through numerical simulation methods,and the fatigue life distribution of the turbo-shaft was calculated by combining finite element analysis and life prediction model.Furthermore,based on the stress-strength interference model,the fatigue reliability model of the turboshaft was constructed.The reliability evaluation was completed using its fatigue life simulation data,and the fatigue reliability function curve of the turbo-shaft was obtained.(3)Based on the active learning agent model,a parameterized model of the stress response of the turbine shaft was established to improve the efficiency of predicting the fatigue life and reliability analysis of the turbine shaft.The main factors affecting the fatigue life of the turbine shaft were efficiently obtained through Sobol’ global sensitivity analysis.(4)The design variables for optimizing the performance of the turbo-shaft were determined based on the analysis of factors affecting fatigue life,and its optimization design model was established.The algorithm flow of the fatigue reliability optimization design of the turbo-shaft was clarified,and the particle swarm optimization algorithm was used to solve the model.The life extension of the vortex shaft was realized. |
