Font Size: a A A

Study On Fatigue Reliability Modeling And Analysis Of Turbine Disk Of An Aero-Engine

Posted on:2020-08-12Degree:MasterType:Thesis
Country:ChinaCandidate:R C DingFull Text:PDF
GTID:2392330596976606Subject:Engineering
Abstract/Summary:PDF Full Text Request
The aero-engine turbine disks are subjected to high temperature and high rotational velocity for a long time,some special requirements for designing,i.e.the thrust-to-weight ratio,fuel efficiency and the overall performance of the aircraft engine,have been brought forward correspondingly.The life and reliability of the aero-engine turbine disk have been restricted the development of major mechanical equipment.During the service of turbine disks,it needs to be continuously started-up,normal operation and shut-down phases and superposition of various complicated tasks.In addition to the complicated working environment,the uncertainty factors affecting the life of the turbine disk become more complicated,which result in different failure modes and typically exhibit random behavior.Failure of a disk often affects the performance of the aero-engine and cause uncontained damage of aero-engine.The low cycle fatigue of a disk,generally,will reduce the safety and reliability of aero-engine,which will lead to catastrophic consequences.All of these give a new challenge to the precision of the prediction model for disk low cycle fatigue(LCF)life and reliability analysis.The advanced countries in aero engine,such as America,England,France and Russia,have set up the sophisticated system of disc LCF life prediction and experiment assessment.Although many domestic scholars and experts have gained a lot of research in this field,there are still many issues that needed to be solved.Therefore,it is of important theoretical significance and practical value to improve the fatigue life prediction accuracy of turbine disk and prolong its service life.The research on cycle fatigue life prediction of disk and reliability analysis under the influence of many uncertain factors is essential.In this dissertation,some key challenges and critical issues in life prediction and reliability analysis on disk are addressed.The main research work can be drawn as follows:(1)Uncertainty analysis in fatigue life prediction of turbine disk.Due to the harsh working environment and the complexity of structure of turbine disks,there are a large number of uncertainties in the fatigue life prediction.Based on this,by fitting the experimental data,the probabilistic model of fatigue life and probabilistic cyclic stress-strain model of turbine disk is established,which can quantify the influence of material parameters and load variations.(2)Probabilistic fatigue life prediction and fatigue reliability analysis of the disk.Firstly,the finite element analysis model of the turbine disk is established by ANSYS.Based on the basic ideas of Monte Carlo method and Latin hypercube sampling,the analytical model under the uncertainty of the turbine disk is established.Secondly,several fatigue life predictions are briefly reviewed.After comparison,the strain life prediction model based on Walker and SWT parameter models is selected for turbine disk life prediction.Again,the Kowen multiplier method is used to obtain a cycle life with a certain degree of reliability and confidence.Finally,the Miner linear cumulative damage rule is used to calculate the total life of the turbine disk,and the log-normal distribution and failure probability map are obtained.(3)Turbine disk fatigue reliability modeling based on neural network.In the fatigue life prediction and reliability analysis,the relationship between the load applied to the working environment of the high-pressure turbine disk and the stress-strain has no certain mathematical expression,and the traditional reliability analysis takes a long time.In view of this,since the neural network is a powerful general function approximation scheme,it has wide applicability in many engineering fields.In this paper,the neural network algorithm is introduced into the analysis of fatigue reliability of the turbine disk,and through the training of the network,Reliability analysis of the total fatigue life of the turbine disk is done.(4)Fatigue life prediction and reliability analysis of turbine disk based on cumulative damage.Combined with the Palmgren-Miner linear cumulative damage model,the one-to-one transformation of the probability density function and the previous statistical results,the probabilistic modeling of fatigue damage accumulation under single stress is established,and then the reliability analysis method under multi-stress is studied based on the transfer of cumulative distribution function.And the reliability analysis and the failure probability of the turbine disk is carried out.
Keywords/Search Tags:turbine disc, uncertainty quantification, reliability analysis, neural networks, probabilistic damage accumulation model
PDF Full Text Request
Related items