| Metal-net-rubber material is a kind of elastic porous dry friction damping material.It is pressed by wire mesh woven by wire mesh.It consumes energy by orderly stacking and interleaving wire deformation inside the wire mesh and sliding dry friction between each other.It has good elasticity and damping characteristics,so it can play a good role in vibration isolation and cushioning.In the field of vibration isolation,its impact resistance under strong vibration such as high speed and heavy load must be considered.However,due to the nonlinear mechanical properties of Metal-net-rubber materials,there are few studies on its impact resistance.Therefore,the research on the impact resistance of Metal-net-rubber has a wide application prospect and research value.In this paper,the mechanical properties of Metal-netrubber under different impact conditions were studied.The impact resistance and dynamic constitutive model were studied by means of experiment and neural network fitting.The research contents are as follows:1.Cylindrical through-hole Metal-net-rubber specimens were prepared,and quasi-static compression hysteresis tests were carried out on Metal-net-rubber specimens with different relative densities.The test force displacement curve of the loading section in the quasi-static compression test of metal mesh rubber is analyzed by segments.The effects of relative density,compression and loading speed on the stiffness of Metal-net-rubber during deformation were studied.The damping characteristics of Metal-net-rubber were analyzed by using energy dissipation coefficient.The stress-strain curves under different static compression conditions are obtained.2.Improve the Hopkinson bar device to realize the medium strain rate compression test of metal mesh rubber material.The metal mesh rubber specimens with different relative densities were subjected to repeated medium strain rate direct impact tests,and the impact test process,test force displacement response and stiffness change were analyzed.The dynamic mechanical properties of Metal-net-rubber under the changes of relative density and impact times were discussed.3.Based on the split Hopkinson compression bar test device,an improved Hopkinson bar device with preload is designed.The incident wave is shaped by changing the shape of the impact bar and waveform shaping technology to meet the impact test requirements of metal mesh rubber materials with high strain rate.The high strain rate dynamic compression test of Metal-net-rubber was carried out to study the effects of its relative density,strain rate and preload on its dynamic mechanical properties.4.Based on the highly nonlinear characteristics of Metal-net-rubber and the fact that its mechanical properties are difficult to accurately describe and predict,artificial BP neural network is used to predict the constitutive model of Metal-net-rubber.The metal mesh rubber compression test data are analyzed and sorted,and a three-layer BP neural network with "4-7-1" structure is established.The neural network model is optimized by genetic algorithm,and the neural network prediction model is trained by using the training set in the test data.Comparing the predicted value with the experimental data,the error is small,and a neural network model with high accuracy is obtained.It is proved that this method can better predict the constitutive relationship of Metal-net-rubber. |
