| Crankshaft is one of the important parts of the engine,which suffers from various impact loads during their working period,the accumulated fatigue damage in the stress concentration areas such as the fillet of the connect rod or the oil hole is the main contributor to the final failure of the parts.Meanwhile,the increasing explosion pressure and the lightweight in actual application will also result in higher demand in fatigue strength.Up to now,the traditional technique in crankshaft research such as the experimental technique or the fatigue analysis software approach always has the flaws of the high cost or the low accuracy.Besides,for some of the fatigue damage models,the application range and applicability has not been clearly defined yet.As a result of this,suitable fatigue damage model chosen and accurate fatigue strength prediction of the crankshaft in its design stage has an important influence in actual engineering application.In order to solve the problems above,this paper will do these jobs below:(1)Crankshaft finite element analysis and experimental verification.In this section,the load period of each crankpin is analyzed and a simplified model is proposed based on this and the structural features of the crankshaft to be taken into the following research instead of the original whole model.Besides,a comparative study of different crankshaft experiment and corresponding rationality is conducted to lay the foundation for the following work.(2)Notch fatigue factor application in crankshaft fatigue limit load prediction.In this research,the crankshaft is considered to be the notched bar and suitable notch fatigue model is chosen.Then the stress concentration factors based on different bending section definitions and corresponding notch fatigue factors are calculated by analyzing the stress condition of a crankshaft under its limit load.Based on this,the fatigue limit loads of other crankshafts with the same material property but different structures are determined.Finally the applicability of these methods can be evaluated by comparing the experimental data with the prediction and the influence of the stress calculation error on the fatigue limit load prediction based on different bending section definitions.(3)The theory of multiaxial fatigue and its application in crankshaft fatigue limit load prediction.In this research,first the stress state of the crankshaft under a bending moment is analyzed,the suitable multiaxial fatigue model is chosen by comparing the stress-strain condition and the crack propagation path.Based on this,the fatigue limit loads of other crankshafts with the same material property but different structures are determined.By comparing the fatigue limit load prediction based on this approach with the experimental data and the influence of the stress and strain calculation error,the applicability of this approach can be evaluated.(4)The indirect defined TCD and its application in crankshaft structural equivalent research.In this section,the length of the critical distance of a given crankshaft is obtained by analyzing the stress distribution under its limit load based on the traditional indirect definition of TCD.After this the limit load of crankshafts with the same material and different structures are determined and corresponding modification method is proposed.By comparing the prediction based on different approaches with the experimental results and the influence of the model parameter calculation error,the applicability of the approach and the modification can be determined.(5)The direct defined TCD and its application in crankshaft structural equivalent research.In this chapter,first the value of the critical distance of a crankshaft is obtained by analyzing the stress distribution under its limit load and the crack-modelling technique,then suitable fatigue criterions and stress distribution are chosen to determine the fatigue limit value of crankshafts with the same material and different structures.By comparing the prediction based on different approaches with the experimental results and the influence of the model parameter calculation error,the applicability of the corresponding methods can be determined.From the work above,these conclusion can be obtained:(1)Crankshaft finite element analysis and experimental verification.In this section,the load period of each crankpin is analyzed and a crankpin is proposed based on this and the structural features of the crankshaft to be taken into the following research instead of the original whole model.Besides,the finite element analysis results shows that clear mesh convergence could be observed where the size changes from 2 mm to 0.5 mm and the error between these results and the experimental data is less than 10%so the model can be taken into the following research.(2)Notch fatigue factor application in crankshaft fatigue limit load prediction.In this section,first the Peterson notch fatigue model is chosen based on the high cycle fatigue and stress concentration property,then the stress concentration factors and notch fatigue factors based on different bending section definitions are calculated,finally the fatigue limit load of crankshafts with the same material and different structure is predicted and verified.Comparison between the prediction and the experimental data shows that most of the errors based on different bending section definitions are less than 10%,so this approach has the certain value in actual engineering application.Besides,among the three different bending section definitions,the prediction based on the third definition has the best accuracy and least sensitivity to the stress calculation error,so it's most suitable to be taken into the final calculation.(3)The theory of multiaxial fatigue and its application in crankshaft fatigue limit load prediction.In this section,first the damage model of the crankshaft under a bending moment is determined to be the shear damage model,then two corresponding shear damage models are chosen:the KBM and Mc Diarmid model.The fatigue limit load prediction of the crankshafts with the same material and different structure shows that in the application of the Mc Diarmid model,the shear stress has the more obvious influence on the final prediction.While for the KBM model,the shear strain has the most obvious influence.Beside,the second model has the best accuracy and least sensitivity to the stress and strain calculation error,so it's most suitable to be taken into the final calculation.(4)The indirect defined TCD and its application in crankshaft structural equivalent research.In this section,first the third and fourth strength criterions are chosen based on the multiaxial fatigue analysis above and a stress distribution function is proposed based on a stress gradient approach.After this the limit load of crankshafts with the same material and different structures are determined.Comparison between the prediction and the experimental data shows that the traditional indirect defined TCD will result in bigger error in the crankshaft fatigue limit load prediction.Besides,among the main model parameters,the largest stress of the crankshaft has the most obvious influence on the prediction.Then a modification approach based on the stress gradient is proposed to modify the length of the critical distance.Validation between the predictions and experimental data shows that compared with the original model,the modified model has the better accuracy in crankshaft fatigue limit load prediction,and is more sensitive to some model parameters such as the relative stress gradient.(5)The direct defined TCD and its application in crankshaft structural equivalent research.In this section,first the length of the critical distance is determined based on the crack-modeling technique,then the limit load of crankshafts with the same material and different structures are determined based on this direct defined TCD.Validation between the predictions and experimental data shows that compared with the critical point method,the critical line method will apply a better accuracy in the fatigue limit load prediction.Besides,this approach is less sensitive to the value of the critical distance,so it's more suitable to be taken into the actual application. |
PDF Full Text Request