FINITE ELEMENT ANALYSIS OF SHELLS - EARLY ACCESS 
Section 4
Expanding to a full plate element solver
21. Section overview - Expanding to a full plate element solver
01:28 (Preview)
22. Procedurally generating a rectangular mesh
24:30
23. Defining plate constraints
11:08
24. Defining the self-weight force vector
10:35
25. Building the structure stiffness matrix
10:05
26. Solving the system and extracting reaction forces
28:13
27. Plotting the plate displacements
18:10
28. Building an evaluation grid for stress resultants
10:31
29. Calculating the moments and shears
22:00
30. Visualising the plate bending moments
14:13
31. Extracting shear forces
29:04
32. Visualising the plate shear forces
12:21
33. Adding strip and edge masking to the shear plot
26:04
34. Adding magnitude clipping to the shear plot
10:40
35. Building an interpolation utility function
09:53
66. The substitute transverse shear strain matrix - part 1
Eliminating zero-energy displacements
Please log in or enroll to access resources

Summary

In this lecture, we'll cover the following:

  • Revisit the assumed transverse shear strain field and its simplified polynomial form that avoids shear locking.
  • Transform transverse shear strains from Cartesian to natural coordinates using the Jacobian.
  • Sample shear strains at midpoints of element edges to determine the unknown field coefficients.
  • Derive a relationship between sampled shear strains and the polynomial coefficients via matrix transformations.

In this lecture, we begin the derivation of a substitute transverse shear strain matrix by first restating the assumed shear strain field in a simplified polynomial form that avoids shear locking. We express these strains in the natural coordinate system and use the Jacobian to relate them back to Cartesian coordinates. We then rewrite the shear strain field in a compact matrix form, where it is defined by a set of unknown coefficients that we aim to determine.

To find these coefficients, we introduce the concepts of natural directions along element edges, their associated angles, and the corresponding shear strain components. By sampling shear strains at the midpoints of each element edge, where higher-order locking terms vanish, we construct a system of equations that links these sampled values to the unknown coefficients. Combining these relationships, we derive a matrix expression that allows us to compute the coefficients directly from the sampled shear strains, establishing a key step towards building the substitute shear strain matrix.

Next up

In the next lecture, we will continue the derivation by expressing the interpolated shear strains in terms of nodal degrees of freedom and assembling the full substitute shear strain matrix.

Tags

transverse shear strainshear lockingnatural coordinatesstrain sampling

Please log in or enroll to continue

If you've already enrolled, please log in to continue.

Finite Element Analysis of Plate and Shell Structures: Part 1 - Plates

An analysis pipeline for thick and thin plate structures, a roadmap from theory to toolbox

After completing this course...

  • You will understand how Reissner-Mindlin theory enables us to accurately capture both thin and thick plate behaviour.
  • You will understand how to turn the fundamental mechanics of plate behaviour into a custom finite element solver written in Python.
  • You will have developed meshing workflows that utilise the powerful open-source meshing engine, GMSH.
  • In addition to using your own custom finite element code, you will be comfortable validating your results using OpenSeesPy and Pynite.
Next Lesson
67. The substitute transverse shear strain matrix - part 2