Study of different shell element properties using Radioss

OBJECTIVE:

To compare the results with base simulation and improved shell element properties as per the steps given below:

1. To change the run time to 55 ms

2. To change the number of animation steps during simulation to a minimum of 25 and a maximum of 60

3. To run the base simulation without modification in elemental properties

4. To determine the energy and mass error after simulation and check whether they are in the acceptable range

5. To plot the rigid wall forces, internal energy, hourglass energy, contact energy and kinetic energy of the simulation

6. To change the shell element properties to the recommended properties and to run the simulation again by saving this file separately. The recommended properties are given below:

7. To follow steps 5 and 6

8. To create a brief report by comparing the results obtained from step 5 and 6 for both simulations.

9. To comment the change in results and to justify why the changes in results is observed.

PROCEDURE:

Case 1: Simulation without changing shell element properties

1. Open Hypermesh software>>Select Radioss as user profile

2. Go to Import Solver Deck>>Open the file Element_Formulation-Shell-3_assignment_0000.rad>>Import.

3. In the components view tab go to Cards>>ENG_RUN card>>T_Stop = 55. In this step the run time has been changed to 55ms. This is shown in the figure below.

4. Go to ENG_ANIM_DT card and change Tstart to 25 and set Tfreq to 1.0. In this step the animation steps during simulation has been changed to 25 ms and frequency has been set to 1. This is shown in the figure below.

5. Ensure that in the ENG_ANIM_ELEM card all the necessary energy parameters required are set. This is shown in the figure below.

6. Go to Analysis>>Radioss>>Select the target folder and assign the file name to be saved>>Radioss. This is shown in the figure below.

7. After running the simulation, the Hyperworks Solver View shows the results obtained from the simulation.

8. The energy error and mass error is obtained from either the solver or the Filename_0001.out file in the destination folder. It is shown in the figure below.

9. From the above results it is cleared that the energy error is -10.3% and mass error is 0.016%. The acceptable value of energy error must be in the range of 5% to -15% and mass error is 2% to 3%. Therefore, in this simulation the accepted errors are within the accepatable values.

10. In the Hyperworks Solver View go to Results to run the animation file. However, the animation file can also be opened by changing Hypermesh to Hyperview and loading the .h3d file. Further, the file is loaded and curves are generated as shown in the figure below.

11. From the above shown image, click on contour to obtain the displacement, velocity and Von-Misses stress contour plots for the animation. This is shown in the figure below.

12. Now go to Hypergraph 2D to generate the necessary plots for the animation. Load the T01 from the destination folder to get the plots.

13. The different energy which needs to be plotted are chosen>>Apply. The generated plots for different time vs energy values is shown in the figure below. 

14. Similarly, the graphs are generated for rigid wall forces. This is shown in the figure below.

Case 2: Simulation with improved shell element properties

1. Repeat the steps 1-5 as given in case 1.

2. Go to Property card>>change the shell element properties as given. The property has to be changed for both Hat-Section and Close-Out Panel. This is shown in the figure below.

3. Repeat steps 6-7 as given in case 1. The energy error and mass error obtained is shown in the figure below. These results are obtained after running the Filename_0001.out file.

4. The energy and mass errors are obtained as -0.1% and 0.016% respectively which are well within the accepted values.

5. Repeat step 10. The contour plots are shown in the figure below.

6. Repeat step 12, 13 as given for case 1. The generated plots are shown in the figure below.

 RESULTS AND CONCLUSION:

 Based on the comparison shown in the above table, following can be concluded:

1. The use of improved shell elements has resulted in simulation with elimination of hourglass energy. This is a direct inference from the graph. (Hourglass modes are element distortions with zero strain energy). Also, there is a reduction of energy error to -0.1% when an improved shell element is used.

2. There is a slight increase observed in maximum displacement. Also, there is a reduction in velocity observed in the nodes near the wall.

3. The Von-Mises stress is reduced in the improved shell element simulation as listed in the table above.

4. The total resultant force has a sharp spike as observed from the graph of improved shell element simulation. This is because of buckling observed in the middle portion of the beam.

DETAILS ON HOURGLASS ENERGY AND MODES:

Hourglass modes are non-physical modes of deformation that occur in under-integrated elements and produce no stress. This is caused mainly due to 

1. Solid elements with a single integration point.

2. Shell or Tshell elements with a single in-plane integration point.

The reduced integration can lead to non-physical zero energy modes, called hourglass modes. The energy associated with these modes is called hourglass energy.

There are two methods to control the formation of hourglassing:

Viscous Method: (Ishell = 1,2,3 or 4 (Q4))

In this method the hourglass is detected with the relative motion of the nodes. A force is applied on the nodes to stabilize the deformation. This force is defined in accordance with the element stiffness. This force introduces an artificial energy known as hourglass energy.

Stiffness Method: (24, QEPH)

Since the hourglass energy depends on natural yield, the material level changes are done to reduce the hourglass energy. Hourglass loading stiffness depends on natural tangent modulus.

Drive Link: https://drive.google.com/file/d/1nXwpdOYqyoHpejlK0Je0ofKUrMyK4VWc/view?usp=sharing