Contents

1. Overview
2. Etymology
3. Cultural Impact

Class of problems for partial differential equations

In the mathematical theory of partial differential equations (PDEs) a class of problems refers to a systematic way of formulating, analysing, and solving equations that involve partial derivatives of an unknown function with respect to more than one independent variable. The classification of PDE problems is a fundamental step that determines which theoretical tools, existence‑uniqueness theorems, and computational techniques are applicable. This article surveys the principal categories, the contexts in which they arise, the salient features that distinguish them, and the principal solution strategies that have been developed. All internal Wikipedia links are retained in their original Markdown form text and no external links are introduced.

Scope

The taxonomy of PDE problems cuts across several domains of human inquiry, reflecting the breadth of phenomena that can be modelled by such equations. The principal fields are listed below; each field is linked to its dedicated Wikipedia entry so that readers can navigate to the relevant background material.

Natural sciences – physics, chemistry, biology, and earth‑system studies frequently employ PDEs to describe continuum phenomena.
Engineering – fluid dynamics, heat transfer, electromagnetism, and control theory rely heavily on PDE models.
Astronomy – radiative transfer and gravitational potentials are modelled with PDEs.
Physics – quantum mechanics, relativistic field theory, and statistical mechanics use PDEs extensively.
Chemistry – reaction‑diffusion systems and transport phenomena are described by PDEs.
Biology – pattern formation, diffusion of chemicals, and population dynamics are often modelled with PDEs.
Geology – seismology and heat flow in the Earth’s interior are governed by PDEs.

The classification also intersects with Applied mathematics , a broader domain that unites analytical, numerical, and computational approaches to solving real‑world problems.

Sub‑disciplinary domains

Within applied mathematics, PDEs are coupled with several specialised areas:

Continuum mechanics – solid and fluid mechanics.
Chaos theory – the study of deterministic nonlinear dynamics that exhibit sensitive dependence on initial conditions.
Dynamical systems – qualitative analysis of solution behaviour over time.

These sub‑fields are often treated as separate headings in the taxonomy because they bring distinct conceptual frameworks and technical toolkits to bear on PDE problems.

Classification

The classification of PDEs is multi‑dimensional, reflecting the way in which the equations are structured, the nature of their solutions, and the methods used to study them. The principal dimensions are outlined below.

By equation type

Ordinary – equations involving derivatives with respect to a single independent variable; contrasted with partial derivatives.
Partial – equations involving partial derivatives with respect to two or more independent variables.
Differential‑algebraic – systems that mix differential and algebraic equations.
Integro‑differential – equations that contain both derivatives and integrals of the unknown function.
Fractional – equations involving derivatives of non‑integer order.
Linear – equations in which the unknown function and its derivatives appear linearly.
Non‑linear – equations in which the unknown function or its derivatives appear non‑linearly.

By variable type

Dependent and independent variables – the distinction between quantities that are solved for and those that are given.
Autonomous – equations that do not explicitly depend on the independent variables.
[Coupled / Decoupled] – systems where equations are interlinked versus those that can be solved independently.
Exact – equations that can be expressed as the total differential of a function.
Homogeneous / Non‑homogeneous – distinctions based on whether the equation equals zero or a non‑zero forcing term.

By order and structure

Order – the highest derivative order present in the equation.
Operator – the formal differential operator that acts on the unknown function.
Notation – conventions for representing derivatives.

Features

The qualitative behaviour of solutions to PDEs is encoded in a suite of concepts that are essential for both theoretical analysis and practical computation. Each feature is linked to its Wikipedia entry for deeper exploration.

Order – determines the number of initial or boundary conditions required for a well‑posed problem.
Operator – often written as (L[u] = f) where (L) denotes a linear differential operator.
Notation – includes symbols such as (\partial) for partial derivatives and (\nabla) for gradient.

Boundary and initial conditions

Initial conditions – specifications of the solution and its derivatives at a single point in time.
Boundary values – specifications on the boundary of the spatial domain.
Dirichlet – conditions that prescribe the solution itself on the boundary.
Neumann – conditions that prescribe the normal derivative on the boundary.
Robin – a linear combination of Dirichlet and Neumann conditions.
Cauchy problem – a unified formulation that may involve either initial or boundary data, depending on context.
Wronskian – a determinant used to assess linear independence of solutions.
Phase portrait – a geometric representation of trajectories in the state space.
Lyapunov / Asymptotic / Exponential stability – criteria for stability of solutions.
Rate of convergence – how quickly numerical or analytical approximations approach the exact solution.
Series – formal expansions of solutions in powers of the independent variables.
[Integral solutions] – representations of solutions via integral operators.
Numerical integration – quadrature techniques used to approximate integrals that arise in analytical solutions.
Dirac delta function – a distribution used to model point sources in PDEs.

Relation to processes

PDEs are closely linked to a variety of discrete and stochastic analogues that capture the same underlying physical phenomena from different conceptual angles. These relationships are formalised through the following correspondences.

Difference (discrete analogue) – finite‑difference approximations of PDEs.
Stochastic – PDEs that incorporate random fluctuations.
Stochastic partial – PDEs that are coupled with stochastic processes such as Brownian motion.
Delay – equations in which the derivative at a given time depends on the solution at earlier times.

Understanding these connections helps to transfer analytical insights across discrete, deterministic, and stochastic frameworks.

Solution

Existence and uniqueness

The foundational theoretical results that guarantee that a PDE problem possesses a solution, and that the solution is unique under prescribed conditions, are summarised below. Each theorem is associated with its Wikipedia entry.

Picard–Lindelöf theorem – provides conditions for existence and uniqueness of solutions to ordinary differential equations; a version exists for certain classes of PDEs.
Peano existence theorem – guarantees existence of solutions under weaker hypotheses.
Carathéodory’s existence theorem – extends existence results to more general settings, including certain nonlinear PDEs.
Cauchy–Kowalevski theorem – a cornerstone result stating that analytic data yield a unique analytic solution in a neighbourhood of the initial surface.

These theorems are typically invoked after specifying the appropriate functional spaces and regularity assumptions.

General solution topics

Initial conditions – often encoded as Cauchy data on a hypersurface.
Boundary values – may be Dirichlet, Neumann, Robin, or mixed.
Dirichlet – prescribing the solution itself.
Neumann – prescribing the normal derivative.
Robin – a weighted combination of the two.
[Cauchy problem] – a unified formulation that may involve a mixture of initial and boundary specifications.
Wronskian – used to test linear independence of a set of solutions.
Phase portrait – visualises the qualitative behaviour of solutions in phase space.
Lyapunov / Asymptotic / Exponential stability – criteria for assessing stability of equilibrium states.
Rate of convergence – quantifies how quickly iterative or numerical schemes approach the exact solution.
Series – includes power‑series and integral representations of solutions.
Numerical integration – techniques such as Gaussian quadrature that are employed to evaluate integrals appearing in analytical solutions.
Dirac delta function – used to model point sources and impulsive forces.

Solution methods

A rich arsenal of analytical and numerical techniques has been developed to tackle the diverse classes of PDEs encountered in science and engineering. The principal families of methods are listed below, each linked to its Wikipedia page.

Inspection – ad‑hoc identification of a solution by recognizing a known pattern.
Method of characteristics – reduces first‑order PDEs to ordinary differential equations along curves in the domain.
Euler – a basic explicit time‑stepping scheme for numerical integration.
Exponential response formula – used for linear time‑invariant systems.
Finite difference – discretises derivatives on a grid; linked variants include Crank–Nicolson .
Finite element – a variational approach that constructs piecewise polynomial approximations.
Infinite element – an extension of finite elements to infinite domains.
Finite volume – conserves fluxes across control volumes, popular in computational fluid dynamics.
Galerkin – a weighted‑residual technique that yields weak solutions.
Petrov–Galerkin – a variant that employs distinct test and trial spaces.
Green’s function – a fundamental solution that propagates influence from a point source.
Integrating factor – transforms a non‑exact equation into an exact one.
Integral transforms – such as Laplace or Fourier transforms, convert PDEs into algebraic equations.
Perturbation theory – expands solutions in a small parameter.
Runge–Kutta – a family of high‑order explicit time‑stepping schemes.
Separation of variables – assumes a product form of the solution to reduce PDEs to ODEs.
Method of undetermined coefficients – guesses a form for particular solutions.
Variation of parameters – generalises undetermined coefficients for non‑homogeneous equations.

People

The development of the theory of PDEs has been shaped by many eminent mathematicians. Their contributions are recorded in the following list, each linked to the corresponding Wikipedia biography.

These figures are further organized in a navigable list that can be accessed via the Wikipedia navigation templates.

Cauchy Problem