| The aero-engine is the "heart" of the aircraft,the source of impetus for the rapid development of the aircraft,and an important indicator of a country's comprehensive scientific and technological level and comprehensive national strength.As a key component of the aero-engine,the high-pressure turbine disk has a decisive influence on its performance and quality.The turbine disk structural design with high-precision and high-efficiency is a key step to improve the overall level of modern aero-engine.This thesis is devoted to reducing the weight of the high-pressure turbine disk,improving its performance,and proposing a structure design method considering the constraint of low-cycle fatigue for high-pressure turbine disk.The main work of this thesis consists of two parts: probabilistic fatigue life assessment of high-pressure turbine disk and research on structural optimization design method of high-pressure turbine disk,the main work is listed as follows:(1)Fatigue-creep life assessment of turbine disk.Firstly,the working conditions and loads of a certain type 1 high-pressure turbine disk are analyzed.Secondly,the local maximum stress/strain of each dangerous point of the high-pressure turbine disk is analyzed based on the stochastic finite element thermo-solid coupling method.then,the SWT model based on Walker correction is used to calculate the low cycle fatigue life at the center hole and the bolt hole;considering the effect of high temperature creep on the fatigue life of the turbine disk,the Larson-Miller creep damage model is introduced to evaluate the creep damage,and finally the fatigue-creep life of the turbine disk is estimated.(2)Fatigue-creep life reliability assessment of turbine disk based on BP neural network method and transfer learning.Firstly,the uncertainties of the turbine disk are quantified,according to the three-dimensional finite element anlysis models,the maximum stress/strain at the dangerous points are obtained.Further to improve the efficiency of the fatigue life assessment of the turbine disk,the two-dimensional axisymmetric finite element analysis models of the turbine disk are established,and also the maximum stress/strain at the dangerous points are analyzed.Then,based on the BP artificial neural network method combining with transfer learning method,different precision data is fused to build the maximum stress/strain prediction surrogate model of each dangerous points of the turbine disk.Finally,the Monte-Carlo method is used to fatigue-creep the turbine disk Variable reliability.(3)A topology optimization design method for the high-pressure turbine disk is proposed based on improved BESO method considering the constraint of fatigue life.Firstly,taking the ratio of volume to stiffness of turbine disk as objective function,the optimization design model for the turbine disk is formulated under the performance constraints of fatigue-creep life and rupture speed;secondly,the turbine disk topology stucture is optimizated based on improved BESO method,and in the optimization process,response surface method is employed to fit the constraints of turbine disk performance indexes.Finally,the performance of optimized design structure and the reference turbine disk are compared.(4)A strucute design optimization method for turbine disk considering uncertainty is proposed.Firstly,the fatigue reliability based topological optimization design model of turbine disk is established.Secondly,combining with the idea of probabilistic reliability optimization design,the structure optimization design and fatigue reliability estination are sequentially proceeded using Random Sampling Sensitivity Analysis-Reliability based Bi-Directional Evolutionary Structural Optimization(RSSA-BESO)method.Finally,the results from reliability based topology optimization design and the deterministic turbine disk optimization design are compared. |
PDF Full Text Request