Please give some input if u can.
I m getting higher value of natural frequency of shell structure following TSDT than that with FSDT using Finite element method. It should be the other way. I double checked my code written in MATLAB but could not find any fault or error.
I think shear locking contributes to some damping reducing the first natural frequency values computed using FSDT which cannot get rid of shear locking. TSDT is better that way especially in laminated materials.
Dear Padmanabhan Sir I am getting higher value of first natural frequency using Third order theory than FSDT.
Here are a few things I would check:
1) Make sure that you have benchmarked your code against known analytical solutions for simple cases, such as a clamped or simply supported flat plate/cylindrical shell.
2) If your code accurately predicts the analytical solutions then vary the mesh size and perform a convergence study for your current shell geometry.
3) Take a look at the mode shapes corresponding to each natural frequency for clues.
4) How much difference is there between the two results? Your code might be suffering from numerical round-off or floating point rounding errors. Consider the precision of your routines.
5) What is the thickness of your shell element? If the elements are not thin enough you could be in a regime where both theories perform poorly and are inaccurate. In the case of analytical solutions, shell theories typically require a length to thickness ratio of greater than 20. (L/H > 20)
- Badges
- Science topic
- Similar topics
- Engineering
- Mechanical Engineering