Studies On Low Cycle Fatigue Life Prediction Of Steam Turbine Blades Based On Mean Stress Relaxation At Blade Roots

As the core components of the steam turbine,the blade root is subjected to large alternating centrifugal loads during service,which is prone to low cycle fatigue failure.Due to the plastic yield,the mean stress relaxation occurs at blade root.Therefore,it brings new challenges to the anti-fatigue design and life prediction of the blades.In this paper,combined with the anti-fatigue design requirements of a new blade root,the low-cycle fatigue test of the blade material and the blade root simulation was carried out.Based on the local stress-strain method,a low-cycle fatigue life prediction method for blade root can be developed by considering the mean stress relaxation behavior of the blade root notch.Ultimately,a more accurate life prediction method is provided for the anti-fatigue design and service safety assessment of the blade.Firstly,a low-cycle fatigue test was carried out on the smooth sample of the root material(OCrXXX)with a strain ratio of R=0,and the cyclic hardening/softening characteristics and low-cycle fatigue characteristics of the material were obtained.The results showed that when the strain amplitude was large,the root material first hardens and then softens.When the strain amplitude was small,the material hardens first,then slowly softens,and finally tends to be stable.Combined with the cyclic stress response of the root material,the mean stress relaxation behavior of the material under different strain amplitudes was further studied.The results showed that the mean stress relaxation of the blade material existed under asymmetric fatigue loading,and the relaxation rate increased rapidly with the increase of strain amplitude.Based on the Landgraf average stress relaxation model,an average stress relaxation description model for the root material of the fatigue process was obtained.In order to determine the local strain response of the root notch,the strain test,the elastic-plastic finite element and the approximate calculation were used to study the deformation of the blade root.The results showed that there were serious stress concentrations in the three notches of the blade root.The most wide gap in the neck section was the most dangerous part and needs attention.By comparing the approximate calculation results of the blade root notch with the strain test results and the elasto-plastic finite element calculation results,it was further proved that the "unified method"approximate calculation of the local strain-strain was basically consistent with the results of elasto-plastic finite element analysis.Therefore,this formula can be applied to the local stress-strain calculation of the blade root notch.In this paper,a low-cycle fatigue test under pulsating loads was performed for the blade root specimens,and the low-cycle fatigue properties and crack initiation life of the blade root specimens were obtained.The results showed that the fatigue cracks of blade root were firstly produced at the widest notch of the neck section,and the crack initiation life accounted for about 78%-93%of the total blade root life.There was a linear relationship between the fatigue life of the blade root and the cyclic load in double logarithmic coordinates,which could be quantitatively described by the Basquin equation.Based on the low-cycle fatigue mechanical behavior of blade material and the true deformation response of blade root notch,a new method for low-cycle fatigue life prediction considering the mean stress relaxation behavior of blade root notch was established by using local strain method.By combined with the low-cycle fatigue test results of the blade root simulation specimens,the prediction accuracy of the above method was verified.The results showed that the prediction method of the fatigue life established in this paper was within 2 times of the dispersion band between the predicted result and the experimental result.Compared with the life prediction results that did not consider the mean stress relaxation behavior,the prediction accuracy was significantly improved.Therefore,the above life prediction method can be applied to the anti-fatigue design and fatigue life prediction of steam turbine blades.