Math 2.0

Numerical analysis Ordinary differential equations Proposition

Exponential integrator

Builds on Initial value problemMatrix exponentialStiff equation

Statement

Let AA be a real d×dd \times d matrix, let N ⁣:ℝd→ℝdN \colon \R^d \to \R^d be continuous, and consider the initial value problem y′=Ay+N(y)y' = A y + N(y), y(t0)=y0y(t_0) = y_0. Put φ1(z)=(ez−1)/z\varphi_1(z) = (e^z - 1)/z for z≠0z \ne 0 and φ1(0)=1\varphi_1(0) = 1, so that φ1(hA)=∑k≥0(hA)k(k+1)!\varphi_1(hA) = \sum_{k \ge 0} \frac{(hA)^k}{(k+1)!}.

  1. The exact solution satisfies the variation-of-constants formula y(t0+h)=ehAy0+∫0he(h−s)AN(y(t0+s)) ds,y(t_0 + h) = e^{hA} y_0 + \int_0^h e^{(h - s)A} N(y(t_0 + s))\, ds, where ehAe^{hA} is the matrix exponential.
  2. The exponential Euler method is the step yn+1=ehAyn+h φ1(hA) N(yn).y_{n+1} = e^{hA} y_n + h\, \varphi_1(hA)\, N(y_n). If NN is constant, it gives the exact solution at every step, for every h>0h > 0.

Remarks

Clause 1 follows by differentiating e−tAy(t)e^{-tA} y(t). For clause 2, if N≡bN \equiv b, the integral in clause 1 equals ∫0he(h−s)Ab ds=hφ1(hA)b\int_0^h e^{(h-s)A} b\, ds = h \varphi_1(hA) b, which is the method. In general, the exponential Euler method replaces N(y(t0+s))N(y(t_0 + s)) by N(yn)N(y_n) in the integral; it has order 11 when NN is Lipschitz and the solution is smooth. Example: for the scalar equation y′=−100y+1y' = -100 y + 1 the method reproduces y(t)=1/100+(y0−1/100)e−100ty(t) = 1/100 + (y_0 - 1/100) e^{-100 t} exactly, whatever the step size. Higher-order exponential Runge–Kutta and Rosenbrock-type methods use the functions φk(z)=∑j≥0zj/(j+k)!\varphi_k(z) = \sum_{j \ge 0} z^j / (j + k)!.

Going deeper

For a student

Exponential integrators are a tool for stiff equations where the stiffness sits in a known linear operator, such as the discretised Laplacian of a semilinear parabolic equation. They are explicit in NN and do not need the solution of nonlinear equations. They need products of φk(hA)\varphi_k(hA) with vectors, computed with Krylov methods, Padé approximants or contour integrals. Their error analysis for parabolic problems uses that the bounds do not depend on the norm of AA.

Reading path

To understand this page from the ground up, read these 4 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. Initial value problem concept
  2. Matrix exponential concept
  3. Stiff equation concept
  4. Exponential integrator statement

Further reading

  1. Web https://en.wikipedia.org/wiki/Exponential_integrator
About this page
TagH0MZB0DPP2
URIm2://H0MZB0DPP2
Kindstatement, proposition
WritingW0 stub
MSC65L05, 65F60, 65L04
StatusThis copy of the site has no build data. Run m2 status H0MZB0DPP2.
AboutMatrix exponential
UsesInitial value problem, Matrix exponential, Stiff equation

Markdown · JSON · Edit on GitHub

What this page needs for W1

  • an inventory row: the row of the inventory that the node accounts for (0 of 1)
  • 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.