Skip to content
TheChatArchiveThe Chat Archive
GPT 5.1· openaiHugging Face Frontier Reasoning Traces

Solve a heat equation using Fourier transforms.

  1. User

    Solve a heat equation using Fourier transforms.

  2. Assistant

    None Consider the standard 1D heat equation on the whole line: [ \begin{cases} u_t = \kappa u_{xx}, & x\in\mathbb{R},\ t>0,\[3pt] u(x,0) = f(x), & x\in\mathbb{R}, \end{cases} ] where (\kappa>0) is the thermal diffusivity.

    We solve this using the Fourier transform in (x).


    1. Take the Fourier transform in space

    Define the Fourier transform (in (x)) by [ \widehat{u}(\xi,t) = \mathcal{F}xu = \int{-\infty}^{\infty} e^{-i\xi x} u(x,t),dx. ]

    Then:

    • (\mathcal{F}_x[u_x] = i\xi,\widehat{u}),
    • (\mathcal{F}x[u{xx}] = -\xi^2,\widehat{u}).

    Apply (\mathcal{F}_x) to both sides of the PDE: [ \mathcal{F}_x[u_t] = \kappa,\mathcal{F}x[u{xx}] \quad\Rightarrow\quad \frac{\partial}{\partial t}\widehat{u}(\xi,t) = -\kappa \xi^2 \widehat{u}(\xi,t). ]

    This is now an ODE in (t) for each fixed (\xi).

    The transformed initial condition is [ \widehat{u}(\xi,0) = \widehat{f}(\xi) = \int_{-\infty}^{\infty} e^{-i\xi x} f(x),dx. ]


    2. Solve the ODE in the transform domain

    Solve [ \frac{\partial}{\partial t}\widehat{u}(\xi,t) = -\kappa \xi^2 \widehat{u}(\xi,t),\quad \widehat{u}(\xi,0) = \widehat{f}(\xi). ]

    This is a first-order linear ODE with solution [ \widehat{u}(\xi,t) = e^{-\kappa \xi^2 t},\widehat{f}(\xi). ]


    3. Invert the Fourier transform

    Now take the inverse Fourier transform: [ u(x,t) = \mathcal{F}^{-1}\xi\widehat{u} = \frac{1}{2\pi}\int{-\infty}^{\infty} e^{i\xi x}\widehat{u}(\xi,t),d\xi = \frac{1}{2\pi}\int_{-\infty}^{\infty} e^{i\xi x} e^{-\kappa\xi^2 t}\widehat{f}(\xi),d\xi. ]

    Recognize that multiplying by (e^{-\kappa \xi^2 t}) in Fourier space corresponds to convolving with a Gaussian in physical space. The inverse transform of (e^{-\kappa\xi^2 t}) is [ \mathcal{F}^{-1}_\xie^{-\kappa \xi^2 t} = \frac{1}{\sqrt{4\pi \kappa t}}\exp!\Big(-\frac{x^2}{4\kappa t}\Big). ]

    Thus [ u(x,t) = (G_t * f)(x) = \int_{-\infty}^{\infty} G_t(x-y),f(y),dy, ] where [ G_t(x) = \frac{1}{\sqrt{4\pi \kappa t}} \exp!\Big(-\frac{x^2}{4\kappa t}\Big) ] is the heat kernel.

    So the solution is [ \boxed{ u(x,t) = \frac{1}{\sqrt{4\pi \kappa t}} \int_{-\infty}^{\infty} \exp!\Big(-\frac{(x-y)^2}{4\kappa t}\Big),f(y),dy } ]

    This solves the heat equation on (\mathbb{R}) with initial condition (f).

    If you have a specific domain (e.g. (0<x<L)) or boundary conditions (Dirichlet/Neumann), I can adapt the Fourier-transform method to that case as well.

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.