Math 2.0

Numerical analysis Definition

Finite element method

Builds on Galerkin methodWeak solutionHilbert spaceBilinear formSobolev space

Definition

Let VV be a real Hilbert space of functions on a bounded polygonal domain Ω⊂ℝd\Omega \subset \R^d, for example the Sobolev space H01(Ω)H^1_0(\Omega). Let a ⁣:V×V→ℝa \colon V \times V \to \R be a bounded bilinear form and ℓ ⁣:V→ℝ\ell \colon V \to \R a bounded linear functional. The weak form of the problem is: find u∈Vu \in V with a(u,v)=ℓ(v)a(u, v) = \ell(v) for all v∈Vv \in V. The finite element method consists of:

  1. A mesh 𝒯h\mathcal{T}_h: a partition of Ω‾\overline{\Omega} into finitely many closed simplices of diameter at most hh, meeting face to face.
  2. A finite-dimensional space Vh⊆VV_h \subseteq V of functions that are polynomials of degree at most kk on each element of 𝒯h\mathcal{T}_h, with a locally supported basis.
  3. The discrete problem: find uh∈Vhu_h \in V_h with a(uh,vh)=ℓ(vh)a(u_h, v_h) = \ell(v_h) for all vh∈Vhv_h \in V_h.

Remarks

In a basis φ1,…,φN\varphi_1, \ldots, \varphi_N of VhV_h, the discrete problem is the linear system Ac=bA c = b with Aij=a(φj,φi)A_{ij} = a(\varphi_j, \varphi_i) and bi=ℓ(φi)b_i = \ell(\varphi_i), and uh=∑jcjφju_h = \sum_j c_j \varphi_j. If aa is coercive, a(v,v)≥α∥v∥2a(v, v) \ge \alpha \norm{v}^2, and ∣a(u,v)∣≤M∥u∥∥v∥\abs{a(u, v)} \le M \norm{u} \norm{v}, then uhu_h exists, is unique, and Céa's lemma gives ∥u−uh∥≤(M/α)min⁡vh∈Vh∥u−vh∥\norm{u - u_h} \le (M / \alpha) \min_{v_h \in V_h} \norm{u - v_h}. Example: −u′′=f-u'' = f on (0,1)(0, 1) with u(0)=u(1)=0u(0) = u(1) = 0, V=H01(0,1)V = H^1_0(0, 1), a(u,v)=∫01u′v′a(u, v) = \int_0^1 u' v', nodes xi=ihx_i = ih with h=1/(N+1)h = 1/(N+1), and hat functions φi\varphi_i with φi(xj)=δij\varphi_i(x_j) = \delta_{ij}. Then ∫φi′2=2/h\int \varphi_i'^2 = 2/h and ∫φi′φi±1′=−1/h\int \varphi_i' \varphi_{i \pm 1}' = -1/h, so A=h−1tridiag⁡(−1,2,−1)A = h^{-1} \operatorname{tridiag}(-1, 2, -1). Non-example: the Galerkin method with VNV_N spanned by sin⁡(πx),…,sin⁡(Nπx)\sin(\pi x), \ldots, \sin(N \pi x) satisfies clause 3 but not clause 2. These functions are not piecewise polynomials on a mesh, so this is a spectral method.

Reading path

To understand this page from the ground up, read these 6 pages in order. The foundations come first.

Show the path as a graph
Each page uses the pages above it. The colour of the bar on a box shows how strong the evidence is, and the box also gives it as text. The list has the same pages.
  • no evidence level
  1. Bilinear form concept
  2. Galerkin method concept
  3. Hilbert space concept
  4. Sobolev space concept
  5. Weak solution concept
  6. Finite element method definition

Further reading

  1. Web https://en.wikipedia.org/wiki/Finite_element_method
About this page
TagCP0TQ5ZWSP
URIm2://CP0TQ5ZWSP
Kinddefinition
WritingW0 stub
MSC65N30, 65M60
DefinesFinite element method
UsesGalerkin method, Weak solution, Hilbert space, Bilinear form, Sobolev space
Changed2 November 2026

Markdown · JSON · Edit on GitHub

What this page needs for W1

  • a source with a locator (0 of 1)

The writing levels say how complete the text of a page is. They do not measure its truth.

What links here

No page links to this page yet.