Differential Equation · aiwiki.page

A differential equation is an equation involving an unknown function and one or more of its derivatives. Unlike an algebraic equation whose unknowns are numbers, it generally seeks a function satisfying a relationship throughout an interval or region. Differential equations connect local rates of change with overall behavior and form a central subject of calculus and applied mathematics. They describe processes such as growth, oscillation, diffusion, and wave propagation. (ocw.mit.edu)

Classification and notation

An ordinary differential equation (ODE) involves derivatives with respect to a single independent variable. For an unknown function $y(t)$ , a general equation of order $n$ can be written

F(t,y,y',\ldots,y^{(n)})=0.

Its order is the highest derivative order appearing in the equation: $y'=ky$ is first order, whereas $y''+\omega^2y=0$ is second order. The independent variable need not represent time. (ocw.mit.edu)

A partial differential equation (PDE) concerns an unknown function of several independent variables and involves partial derivatives. For example, the one-dimensional heat equation

\frac{\partial u}{\partial t} =\alpha\frac{\partial^2u}{\partial x^2}, \qquad \alpha>0,

relates temporal change to spatial curvature. It is second order because its highest-order derivative is the second spatial derivative. (ocw.mit.edu)

An equation is linear when the unknown function and its derivatives occur linearly, with coefficients depending only on the independent variables. A linear ODE has the form

a_n(t)y^{(n)}+\cdots+a_1(t)y'+a_0(t)y=g(t).

It is homogeneous when $g=0$ , and nonhomogeneous otherwise. Terms such as $y^2$ , $yy'$ , or $\sin y$ make an equation nonlinear. For a homogeneous linear equation, any linear combination of solutions is another solution—the superposition principle. (ocw.mit.edu)

Solutions and supplementary conditions

A classical solution has the required derivatives and satisfies the equation pointwise on a specified domain. A formula is therefore incomplete without an interval or region on which it is valid. Solutions need not have expressions in elementary functions: a definite integral can define an exact solution even when no elementary antiderivative exists. (ocw.mit.edu)

Differential equations commonly admit families of solutions. For

y'=ky,

the general solution is $y(t)=Ce^{kt}$ , with arbitrary constant $C$ . An initial value problem adds data at one point, such as $y(t_0)=y_0$ , selecting

y(t)=y_0e^{k(t-t_0)}.

A boundary value problem instead prescribes conditions at different points or along a region’s boundary. For PDEs, initial data may specify an entire spatial profile rather than a single number. (ocw.mit.edu)

Systems describe several interacting unknown functions. A linear first-order system can be expressed as

\mathbf{x}'=A(t)\mathbf{x}+\mathbf{b}(t),

where $A(t)$ is a matrix. Higher-order ODEs in normal form can be converted into first-order systems by treating successive derivatives as additional unknowns. This connects differential equations with linear algebra and provides a common framework for analytical and numerical methods. (ocw.mit.edu)

Existence, uniqueness, and qualitative behavior

Finding a formula differs from establishing whether a solution exists or is unique. The Picard–Lindelöf theorem gives local existence and uniqueness for $y'=f(t,y)$ when $f$ is continuous and satisfies suitable local Lipschitz continuity in $y$ . Continuity alone does not generally guarantee uniqueness. (ocw.mit.edu)

Local existence also does not imply existence for all time. The initial value problem

y'=y^2,\qquad y(0)=1

has solution $y(t)=1/(1-t)$ on its maximal interval containing zero, $(-\infty,1)$ ; the solution becomes unbounded as $t$ approaches $1$ . (ocw.mit.edu)

Qualitative analysis examines equilibria, stability, oscillations, and long-term behavior without necessarily deriving explicit solutions. Direction fields visualize the slope prescribed by a first-order ODE. For PDEs, well-posedness requires existence, uniqueness, and continuous dependence on the specified data. These properties depend on the equation, domain, conditions, and function spaces involved. (ocw.mit.edu)

Analytical and numerical methods

Analytical techniques exploit particular structures. Separation of variables solves certain first-order equations; integrating factors handle first-order linear equations. Constant-coefficient linear equations can be reduced to characteristic equations. Other approaches use power series, variation of parameters, or integral transforms such as the Fourier transform. No single method provides elementary formulas for every differential equation. (ocw.mit.edu)

Numerical methods approximate solutions at discrete points. For $y'=f(t,y)$ , Euler’s method uses

y_{n+1}=y_n+h f(t_n,y_n), \qquad t_{n+1}=t_n+h.

Runge–Kutta methods combine multiple slope evaluations per step to obtain higher-order approximations. Accuracy depends on the method, step size, and solution behavior; numerical stability determines whether errors are controlled or amplified. (ocw.mit.edu)

PDE theory also employs weak solutions, which satisfy an integrated formulation rather than requiring every derivative to exist classically. This permits treatment of less regular functions and is important in existence theory and variational formulations. (math.ucdavis.edu)

Modeling applications

In classical mechanics, differential equations describe motion and oscillating systems, including forced and damped vibrations. The heat equation models heat conduction and diffusion, while wave equations describe propagating disturbances. In quantum mechanics, the Schrödinger equation governs wave-function evolution. A model’s solutions depend not only on the differential equation but also on its parameters, domain, initial or boundary data, and assumptions about the physical system. (ocw.mit.edu)