Simulation of a Rail Crash For Various Element Formulation Values using Radioss

OBJECTIVE :

* To compare the results of Base simulation to the simulation with improved shell element properties at :
- Run time = 55 ms
- No. of animation steps : Min 25 ms, Max : 60 ms

Improvements to be made to Shell element properties :

- I_shell = 24
- I_smstr = 2
- I_sh3n = 2
- N = 5
- I_thick = 1
- I_plas = 1

Results to be compared between the simulations :
- Difference in Energy error and mass error
- Variation of Rigid wall forces, Internal energy, Hourglass energy, Contact energy, Total Energy

Theory :

For this simulation it is necessary to set the user profile to RADIOSS in Hypermesh.

Definitions :

* Energy Error :

- The difference in Energy level in the system between the starting of simulation to the energy level in system at each or final time step is called energy error.

- The value is always computed in percentage and

• If the error is negative, it means that some energy has been dissipated.
• If the error is positive, there is an energy creation with the system.

- The formula for computing Energy error is shown below :

* Mass Error :

- The difference in Mass in the system between the starting of simulation to the energy level in system at each or final time step is called energy error.

* Rigid Wall Forces :

- The force experienced by the Rigid wall due to the effect of the simulation is called Rigid Wall Forces.

- In this case the force experienced as a consequence of the impact of the tube on the wall is computed.

* Internal Energy :

- The energy in a system arising from the relative positions and interactions of its parts.

Study of Provided File :

* The provided system is made of 4 Components, they are

- 1D Rigid Body.

- 1D Springs defined by a P13_Spring_Beam Property Card Image which will act as Spot Welds.

- 2D Base surface called "Close Out Section" defined by a P1_Shell Property Card Image.

- 2D Top surface called "Hat Section" defined by a P1_Shell Property Card Image.

* A rigid wall is defined

* For the current simulation, it is required to set the Run time of simulation and control the number of animation steps as per provided criteria.

- To change the Run time of simulation, the value `T_(st op)` is changed in ENG_RUN Card as shown below :

- To control the number of animation steps, the value `T_(f req)` is changed in ENG__ANIM_DT Card as shown below :

* It is also mentioned that for the second case, the shell element properties should be changed.

- It can be seen that there are 3 properties are created in the provided Solver file, by clicking on the properies, we can see how they are defined (shown below).

- It is noted that 2 of the 3 properties created are with P1_SHELL Card Image and the other is a P13_SPR_BEAM Card Image. For the second case in this challenge it is required that we manipulate the properties of the shell elements with the values provided.

Running the simulation :

- To run a defined simulation, Go to Radioss under Analysis Radio Button and set the values as shown below before clicking on Radioss button again.

- The value "-nt 4" is used to limit the number of cores of the computer processor to be used by the solver and can be varied as per requirement

- The input file can also be changed which will make the solver create a duplicate of the file and save the results at the new location.

Post - processing :

Analysing the files through Explorer :

* Once the simulation is finished, it can be noted from explorer that, at the location of input file, lots of new files are created

.out Files

- The File_Name_0000.out and File_Name_0001.out files are the output files from the solver after processing the respective Input files of the same names.

- The .out files can be opened through notepad or any similar softwares.

- The _0000.out file contains inputs to the _0001 file and does not contain any results from the simulation. However the _0001.out file contains the energy datas and solver step datas as shown below :

The acceptable value for energy error is below -15% and the ideal value is -5%.

.h3d and Axxx Files

- The Axxx files (where xxx represent number of the animation steps from 001) and .H3D files contain information from all the Axxx files in one single files.

- Both these file types contain the animation of the simulation and can be opened through HyperView.

- The only difference between them is that each file in Axxx contains only 1 time step. To view the entire animation, all Axxx files must be kept in the same location and the first file should be opened through Hyperview, but in the case of an .h3d file, it is a standalone file which can be opened without any other files.

.T01 Files

- These files contain all numerical datas which are required for plotting purposes.

- They can be opened through Hypergraph 2D

Animating in Hyperview :

- As mentioned above, open the .h3d file or .A001 file in Hyperview

- With the Hyperview UI, the animation can be played, paused, the playback speed can be controlled.

- The Hyperview UI enables us to plot the values of stress, displacement etc while also enabling us to activate dynamic max-min which shows the node locaiton where the maximum and minimum of applied contour is.

The plotting can be done using the "Contour" option and enabling and disabling the Dynamic max-min comes under "Measures" Option as shown below :

Plotting in Hypergraph :

- Similarly, as mentioned above, open the .T01 file in Hypergraph 2D

- Select the parameters along x and y axis respectively and click on apply to generate the graph, the UI is shown below

SIMULATION :

Case (i) : Base simulation :

* Card Manipulation : As mentioned above, the ENG_RUN card and ENG_ANIM_DT cards are modified as per provided criteria

* Running the simulation : Since the first case does not require for any other data/property manipulation, the simulation is run as shown above

* Perforning Energy error and Mass Error Test : The output file Filename_0001.out file is opened and the final value of energy error and mass errror is checked to see if it is under accepatable range :

It is observed that the values of energy and mass errors are within acceptable ranges, hence the simulation is qualified for post-processing.

* Post Processing the data - Generating animations and Plotting Required graphs :

Graph 1 : Rigid wall Forces vs Time :

Graph 2 : Internal Energy/ Kinetic Energy/ Hourglass Energy/ Total Energy vs Time

Graph 3 : Contact Energy vs Time :

Animation 1 : Displacement vs Time

Animation 2 : Von-Misses Stress vs Time

All datas obtained from graph and Animations have been noted down in a report and will be discussed in detial after running the simulation for second case.

Case (ii) : Improved shell element Properties:

* Creating a new folder location : The initial files provided are copied to a new location in explorer and the file name is changed for identification purposes. (To identify which model is the modified one)

* Card Manipulation : As mentioned above, the ENG_RUN card and ENG_ANIM_DT cards are modified as per provided criteria

* Shell element properties manipulation : As mentioned above, the properties with P1_SHELL card image are modified as told. The modified card image is shown below :

* Running the simulation : Once the other Shell element is also modified, the simulation is run as shown above.

* Perforning Energy error and Mass Error Test : The output file Filename_0001.out file is opened and the final value of energy error and mass errror is checked to see if it is under accepatable range :

It is observed that the values of energy error is almost negligible and the mass error remains same and within acceptable range. Hence the simulation is qualified for post-processing.

* Post Processing the data - Generating animations and Plotting Required graphs :

Graph 1 : Rigid wall Forces vs Time :

Graph 2 : Internal Energy/ Kinetic Energy/ Hourglass Energy/ Total Energy vs Time

Graph 3 : Contact Energy vs Time :

Animation 1 : Displacement vs Time

Animation 2 : Von-Misses Stress vs Time

REPORT :

- A report was Compiled with the data obtained from the simulation.

- A page of the report is shown below and the full report is attached at the end.

Inference from Simulation :

There is indeed a difference in the results between the two cases and It is evident from the Graphs, output files and animations that the Rail simulation with the modified Shell element properites reflects a much more realistic scenario than Case 1.

- One of the predominant reasons for the change in results would be that the effect of Hourglassing is significant in CASE 1 such that by viewing the animation we can see the elements getting deformed in ways physically impossible in Case 1 even at the beginning of the simulation.

- The other reason would be due to the vast difference in contact energy between the two cases which lead to variations accross the simulation and might be the reason for the significant energy error in Case 1.

CONCLUSION :

Thus the results of Base simulation and simulation with improved shell element properties were compared and a report was compiled with the differences highlighted.

All relevant files are attached below.

https://drive.google.com/file/d/1_eKJ4rcRjqGWPNAAPNIfjueF1u0EWeV-/view?usp=sharing