Study On Mechanical Properties Of Self-piercing Riveted Joints Of Metal Foam Sandwich Structures And Strength Prediction Of Joints Based On Neural Network

With the concept of green design and manufacturing,lightweight structures are frequently used in different manufacturing fields.As structural parts,the metal foams are usually used in combination with traditional compact metals.Metal foam sandwich structures not only meet the stiffness and strength requirements of structural parts,but also meet the requirements of high-energy absorption characteristics.Such structures have been widely used in automobile manufacturing and aerospace industry.To study the effectiveness of the self-piercing riveting(SPR)technology to connect the metal foam sandwich structures,the forming quality,static and fatigue performance of the self-piercing riveted metal foam sandwich structures were investigated,and the microscopic failure mechanism of the joints were analyzed.Besides,the strength prediction of the SPR joint was studied based on the neural network,and the correlation between the strength of the sheet and the strength and failure mode of the SPR joints was summarized on the basis of the previous research works.The research works are mainly summarized as below:1.In the research on the SPR process.The metal sheets used in this study was 5052 aluminium alloy(AA5052)sheets with a dimension of 110mm×20mm×1.5mm,and nickel foam sheets and copper foam sheets with dimensions of 20mm×20mm×1.0mm and 20mm×20mm×1.5mm were used as the metal foam interlayers.The forming quality of the joints were evaluated,and the influence of the metal foam interlayer on the forming quality of the SPR joint under the same riveting parameters was studied,which provide reference for the connection of more kinds of metal foam and related process improvement.The research shows that the type and thickness of metal foam interlayer have a great influence on the forming quality of SPR joint,which directly affects the rivet spread and the remaining thickness of the bottom.2.In the research on the static performances of joints.Tensile-shear tests were conducted to each joint,and the static failure load and energy absorption performance of each joint were compared.The influence of metal foam interlayer on the static performance of the SPR joint was analyzed.In addition,the failure modes of difference joints were studied.Based on the typical fracture observed,which was examined using scanning electron microscopy,the static failure mechanism was analyzed.The results illustrate that the metal foam affects the static performance of SPR joint.The sandwich structure embedded with nickel foam exhibit the best mechanical properties.3.In the research on the fatigue properties of joints.In order to acquire the fatigue data at different load levels,the fatigue tests with single tensile-tensile mode were executed on each groups of the joints.Based on the three parameter empirical formula,S-N curve fitting method was used to draw the F-N curves of the joints.The influence of the metal foam interlayer on the fatigue properties of the SPR joint was analyzed.The fatigue failure fracture of the joints was observed by scanning electron microscope,and the fatigue failure mechanism was analyzed.The results show that the metal foam interlayer reduces the fatigue life of the SPR joint,and different metal foam has the difference influence on fatigue performance of SPR joint.4.In the research on the strength prediction of SPR joints.The orthogonal test of SPR was performed to AA5052 sheets.The influence degree of each parameter on the strength of the joint was analyzed by grey correlation degree.Based on BP neural network,the neural network prediction model of the riveting parameters and the strength of the joints were established.The network was trained by the experimental data,and the validity of the prediction model was verified by the samples which were not trained.The results show that the riveting pressure and the compressing pressure have the greatest influence on the joint strength.The riveting pressure plays a decisive role in the joint strength,and the joint strength can be improved by increasing the riveting pressure.The BP neural network optimized by LM algorithm has good generalization ability and prediction accuracy.The prediction accuracy of the untrained samples is high,the maximum relative error is 5.48%,and the prediction accuracy is over 90%.5.In the research on the relationship between the strength of the sheet and the strength and failure mode of SPR joints.Took SPR joints of light alloy as the research object,using the static tests to compare the static performance of the joints.The effect of sheet strength on the failure modes of the joints was analyzed,the microscopic failure mechanism was observed by scanning electron microscope.The results indicate that: the failure modes of the joints was affected by the strength of the sheet,when the gap of the yield strength between the upper and lower sheets is large,the failure modes of the joints are the sheet with low yield strength fractured;when it is small,the joints failure due to the separation of the rivet and the sheet.When the failure modes of the joints are the same,the strength of the sheet affects the static failure loads of the joints.