The Finite Element Method (FEM) is a powerful computational technique used
to solve complex problems in engineering and physics. The basic algorithm
for solving problems using FEM can be broken down into several key steps:
1. Problem Definition and Discretization
• Define the Problem: Clearly define the physical problem, including
the geometry, material properties, boundary conditions, and the
governing differential equations.
• Discretize the Domain: Divide the problem domain into a finite
number of smaller, simpler parts called elements. This process is
known as meshing. Each element is connected at points called nodes.
2. Selection of Element Type and Shape Functions
• Choose Element Type: Depending on the problem, select an appropriate
type of element (e.g., line elements for 1D problems, triangular
or quadrilateral elements for 2D problems, and tetrahedral or
hexahedral elements for 3D problems).
• Define Shape Functions: For each element, choose shape functions
(also called interpolation functions) that approximate the solution
within the element. These functions are usually polynomial
functions of the spatial coordinates.
3. Derive Element Equations
• Formulate Element Stiffness Matrix and Load Vector: Using the
governing differential equations and the shape functions, derive
the element stiffness matrix (which relates nodal displacements to
forces) and the element load vector.
• Apply Numerical Integration: Perform numerical integration (e.g.,
Gaussian quadrature) to evaluate the integrals involved in the
stiffness matrix and load vector formulations.
4. Assemble Global Equations
• Assemble Global Stiffness Matrix and Load Vector: Combine the
stiffness matrices and load vectors of all elements to form the
global stiffness matrix and global load vector. This step involves
summing the contributions from each element to the corresponding
entries in the global matrices.
5. Apply Boundary Conditions
• Impose Boundary Conditions: Modify the global stiffness matrix and
load vector to incorporate boundary conditions (e.g., fixed
supports, prescribed displacements, and applied loads). This step
ensures that the solution satisfies the physical constraints of the
problem.
6. Solve the System of Equations
• Solve for Nodal Displacements: Use numerical methods (such as
Gaussian elimination, LU decomposition, or iterative solvers) to
solve the resulting system of linear algebraic equations for the
nodal displacements.
7. Post-Processing
• Compute Element Strains and Stresses: Using the nodal
displacements, calculate the strains and stresses within each
element. This often involves differentiating the shape functions
and applying constitutive relations (e.g., Hooke's law for linear
elasticity).
• Visualize Results: Generate plots and visualizations of the
displacement field, strain field, and stress field to interpret and
analyze the results. This can involve contour plots, deformation
plots, and vector plots.
8. Validation and Verification
• Verify the Solution: Check the numerical results for convergence
and accuracy. This might involve refining the mesh, comparing with
analytical solutions (if available), or validating against
experimental data.
• Perform Sensitivity Analysis: Analyze the sensitivity of the
solution to changes in input parameters (e.g., material properties,
boundary conditions) to ensure robustness and reliability.
