Knowledge Based Cloud FE Simulation of Sheet Metal Forming Processes

The overall goal of the Knowledge Based Cloud FE Simulation Method is to combine specialties from different fields in a single portal to enhance the capability and accuracy of forming process simulations without significantly increasing the difficulty of using them. This method can help answer key questions in the sheet metal forming field, such a setting the parameters to forming components successfully, and predicting the lifetime of forming tools. A major strength of this methodology is that it enables advanced predictive models to be utilized with any FE simulation software without requiring any modifications to the models themselves.

Generally, individuals new to this method will be able to utilize it with very little difficulty because of the strain lie user interface. Visual demonstration will show how straightforward it really is to encompass advanced predictive models in conventional FE simulations. This section covers how KBC-FE computing can be used to predict formability in a hot stamping process.

Start a new project in the FE simulation software. Select the process as stamp hot forming and the solver type as PAM-AutoStamp when saving the project. Next, import the door inner die by first clicking on the import tools CAD, name the imported object as die, toggle the hot forming strategy to mesh the tools, then import and transfer the door inner IGS geometry file into the graphical interface.

Now, repeat the process to import the punch and blank holder. Next, under the set-up tab, click on blank. Then, in the blank editor, click add blank.

Set the new object as blank, and set the type as surface blank. Now, choose outline for the definition type, and import the blank shape by clicking on import from CAD file. Under mesh options, define the refinement as imposed level, and select level one.

Then, turn off automatic meshing and set the mesh size to four millimeters. Proceed with defining the material properties in the blank editor. Under the material tab, click on load a material, and select the AA-six-zero-eight-two material.

Set the rolling direction to X equals one. Set the blank thickness to two millimeters and the blank initial temperature to 490 degrees celsius. Next, go to the set-up tab, click on process, and select the plus icon to load a new macro.

Then, browse to the stamp hot forming folder and select the HF validation double action GPA file. In the customize dialogue, activate the blank, die, punch, and blank holder objects. Under the stages tab, activate gravity, holding, stamping, and quenching.

Now, set all parameters in the object's attributes under the set-up tab to correspond with the actual experimental set-up. Set the heat transfer coeeficients as a function of gap and contact pressure. Then, click the check icon to check the set-up for errors, and if there are none, click on the computation icon to start the simulation on the host computer.

After observing the results, run a script to export the major strain, minor strain, and temperature contour values for all elements from all the simulations states as ascii files, and then save the files. By exploiting the data in this way, no information about the geometry of the component is transferred to the online portal, thus protecting any confidential information. Now, access smart forming, which is the newly created portal for KBC-FE simulations, and log in to your own user profile.

Select the forming limit prediction module, and export the simulation result files to the cloud computer. Then, input the number of states in the simulation, manually input the simulation details and parameters, and start the computation. After the computation is made, download the results from the cloud computer for visualization in your FE simulation.

Then, load the final state of the FE simulation results, and under the coutours tab, click on imported, and then scalar values. Select ascii to display the forming limit prediction results. This section covers how KBC-FE computing can be used to predict tool life using an alternate forming process.

Create and name a new simulation project in the FE simulation software. Select the process as standard stamping, and the solver type as PAM AutoStamp when saving the project. Next, import the die geometry by clicking on the import tools CAD.

Then, import and transfer the u-shaped die IGS geometry file into the graphic interface. Select the validation strategy for meshing of tools, and set the mesh size to two millimeters, with a maximum angle of five. Name the imported object as die.

In the same manner, import the punch and blank holder. Now, under set-up, click on blank, and add blank in the blank editor. Set the new object as blank and then select the type as surface blank.

Then, choose four points for the definition type and set the blank size to 120 by 80 square millimeters. Ensure that automatic meshing is turned off and set the mesh size to one point five millimeters. Now, define material properties in the blank editor.

Click on load a material under the material tab and select the AA five-seven-five-four H one-one-one material as the material properties. Then set the blank thickness to one point five millimeters with an initial temperature of 20 degrees celsius. Continue by clicking on process under the set-up tab and select the plus icon to load a new macro.

Browse to the stamp feasibility folder and select the stamping only double action dot GPA file. In the customize dialogue, activate the blank, die, punch, and blank holder. Under stages, activate stamping.

Now, set all the parameters in the simulation to correspond with the actual experiment set-up. Next, click on check in the set-up to ensure no errors were made. Now, click on the computation icon and start the computation for an 11 state u-shape benting simulation on the host computer.

When the simulation is complete, run a script to export ascii files of the coordinate data and the contact pressure data for the die. Then, from the smart forming portal, select the tool life prediction module, manually input the simulation details, and the number of states in the simulation, and export the simulation result files to the cloud computer. Then, start the computation.

Once the computation is complete, download the results and view their final state in the FE simulation software. To do so, go to the contours tab, click on imported, and then scalar values. Then select ascii to display the tool life prediction results.

The initial blank shape adopted from a conventional cold stamping process was used in the KBC-FE simulation. Experimental results with this shape had large failure areas visible after the hot stamping. After one iteration of blank shape optimization, an almost fully successful panel was formed with much less necking.

It can be seen that there is still an indication of necking at the pockets in the top right and left corners of the panel. After further optimization, a blank shape with no visible necking on the panel was obtained. The optimized blank shape was verified by hot stamping trials conducted on a fully automated production line.

To investigate the effects of blank holding force on tool life, three blank holding forces were examined. At a constant forming speed of 250 millimeters per second, over 300 forming cycles, the remaining coding thickness decreased as the blank holding force increased. Graphing the pressure and remaining coding thickness along the curvilinear distance of the die showed that the wear of the coding occurred mainly at the die entry radius.

The two peak values of coding thickness reduction correspond with the peaks of the pressure. By developing at once the predictive models and implementing them as modules on the smart forming portal, the accuracy of commercional FE simulations can be enhanced immeasurably without the use of complicated sub-routines. While attempting this procedure, it is important to remember that different modules, while have to be calibrate according to the sheet metal alloy being simulated.

In addition to forming limit and tool life prediction, other features of forming processes can potentially be captured using this technique, such as marchistructural evolution and the prediction of post form strength. This implies that specializations of leading scientists from around the world can now be linked together by contributing the work in metal forming in the form of modules. The implications of this technique extend to big data.

Information of forming conditions from numerous processes can be collated for relevant analysis to guide future experimental work and model developments.