Physlib

PhyslibAlpha.Electromagnetism.Distributional.WireJunction

Junctions of thin wires

Let `A` be a distributional electromagnetic potential and `J` a distributional Lorentz current density such that Maxwell's equations hold, `IsExtrema 𝓕 A J`. Taking the divergence of Ampère's law and using Gauss's law gives the continuity equation `∂ₜ ρ + ∇ ⬝ J = 0` as an equation of distributions.

Distributions allow currents to be carried by thin wires. A junction is modelled by finitely many straight, semi-infinite wires leaving the origin in the directions `u k`, carrying steady currents `I k`. The divergence of such a current density is `(∑ k, I k) δ₀`: the wires end at the origin, where the currents must come from somewhere. If no charge accumulates, the continuity equation forces `∑ k, I k = 0`.

This is motivated by Kirchhoff's current law, of which it is a special case: a single node with steady currents, rather than the branch currents at every node of a lumped-element circuit.

Main results

- `Electromagnetism.DistElectromagneticPotential.continuityEquation` : the continuity equation for distributions. - `Electromagnetism.DistLorentzCurrentDensity.IsWireJunction` : the current density of a junction of thin wires. - `Electromagnetism.DistLorentzCurrentDensity.IsWireJunction.distSpaceDiv_currentDensity` : the divergence of the current density of a junction. - `Electromagnetism.DistLorentzCurrentDensity.IsWireJunction.sum_currents_eq_zero` : the currents leaving a junction of thin wires sum to zero.

Contents

- A. Analysis on `Space` - A.1. Time derivatives and the divergence of distributions - A.2. Schwartz functions along a ray - A.3. Integrating test functions over time - B. The continuity equation - C. Junctions of thin wires - C.1. The current density of a junction - C.2. The currents at a junction sum to zero

A. Analysis on `Space`

A.1. Time derivatives and the divergence of distributions

A.2. Schwartz functions along a ray

The ray from the origin in the direction `w` is parametrized by `s ↦ s • w`.

A.3. Integrating test functions over time

B. The continuity equation

C. Junctions of thin wires

C.1. The current density of a junction

The `k`-th wire is the ray `s ↦ s • u k`, `s > 0`. A current `I k` along it has current density `I k` times the unit tangent times the arc-length measure on the ray; in the parameter `s` this is `I k • u k` times the Lebesgue measure, whatever the length of `u k`. The currents are steady, so the pairing with a test function `η` only sees its time integral.

C.2. The currents at a junction sum to zero

10 declarations

theorem

∂t(∇⋅f)=∇⋅(∂tf)\partial_t (\nabla \cdot \mathbf{f}) = \nabla \cdot (\partial_t \mathbf{f}) for Distributions

Let f\mathbf{f} be a vector-valued distribution defined on the spacetime domain Time×Space d\text{Time} \times \text{Space } d. The temporal derivative ∂t\partial_t and the spatial divergence operator ∇⋅\nabla \cdot commute, such that: ∂∂t(∇⋅f)=∇⋅(∂f∂t)\frac{\partial}{\partial t} (\nabla \cdot \mathbf{f}) = \nabla \cdot \left( \frac{\partial \mathbf{f}}{\partial t} \right) where ∇⋅f\nabla \cdot \mathbf{f} is the distributional spatial divergence and ∂∂t\frac{\partial}{\partial t} is the distributional temporal derivative.

theorem

The restriction of a Schwartz function to a line through the origin is a Schwartz function

For any dimension dd, let χ∈S(Space d,R)\chi \in \mathcal{S}(\text{Space } d, \mathbb{R}) be a Schwartz function and w∈Space dw \in \text{Space } d be a non-zero vector. Then there exists a Schwartz function g∈S(R,R)g \in \mathcal{S}(\mathbb{R}, \mathbb{R}) such that for all s∈Rs \in \mathbb{R}, g(s)=χ(s⋅w)g(s) = \chi(s \cdot w). Here, s⋅ws \cdot w represents the parametrization of a line through the origin in the direction of ww.

theorem

Fundamental Theorem of Calculus for Schwartz Functions along a Ray

Let ψ\psi be a Schwartz function on Rd\mathbb{R}^d (denoted by ψ∈S(Rd,R)\psi \in \mathcal{S}(\mathbb{R}^d, \mathbb{R})) and let w∈Rdw \in \mathbb{R}^d be a non-zero vector. The integral along the ray (0,∞)(0, \infty) in the direction of ww of the directional derivative of ψ\psi in the direction ww is equal to the negative value of ψ\psi at the origin: ∫0∞Dψ(sw)[w] ds=−ψ(0)\int_{0}^{\infty} D\psi(sw)[w] \, ds = -\psi(0) where Dψ(x)[w]D\psi(x)[w] denotes the Fréchet derivative of ψ\psi at the point xx evaluated in the direction ww. This result is a version of the fundamental theorem of calculus applied to a function restricted to a ray.

theorem

∑iui∫0∞∂iψ(su) ds=−ψ(0)\sum_i u_i \int_0^\infty \partial_i \psi(su) \, ds = -\psi(0)

Let ψ:Rd→R\psi: \mathbb{R}^d \to \mathbb{R} be a Schwartz function and u∈Rdu \in \mathbb{R}^d be a non-zero vector with components uiu_i relative to the standard orthonormal basis {ei}\{e_i\}. Then the sum over the components of uu multiplied by the integral along the ray s↦sus \mapsto su of the partial derivatives of ψ\psi equals −ψ(0)-\psi(0): ∑i(∫0∞∂ψ∂xi(su) ds)ui=−ψ(0)\sum_i \left( \int_0^\infty \frac{\partial \psi}{\partial x_i}(su) \, ds \right) u_i = -\psi(0) where ∂ψ∂xi\frac{\partial \psi}{\partial x_i} denotes the directional derivative of ψ\psi in the direction of the basis vector eie_i.

theorem

Integration over Time Commutes with Spatial Differentiation: ∫∂η∂xidt=∂∂xi∫ηdt\int \frac{\partial \eta}{\partial x_i} dt = \frac{\partial}{\partial x_i} \int \eta dt

Let η∈S(Time×Space d,R)\eta \in \mathcal{S}(\text{Time} \times \text{Space } d, \mathbb{R}) be a Schwartz function on spacetime. For any index i∈{0,…,d−1}i \in \{0, \dots, d-1\} of the standard orthonormal basis {ei}\{e_i\} of Space d\text{Space } d and any point x∈Space dx \in \text{Space } d, the integral over time of the partial spatial derivative of η\eta is equal to the partial spatial derivative of the time integral of η\eta: ∫Time∂η∂xi(t,x) dt=∂∂xi(∫Timeη(t,x) dt)\int_{\text{Time}} \frac{\partial \eta}{\partial x_i}(t, x) \, dt = \frac{\partial}{\partial x_i} \left( \int_{\text{Time}} \eta(t, x) \, dt \right) where ∂η∂xi\frac{\partial \eta}{\partial x_i} denotes the Fréchet derivative of η\eta at (t,x)(t, x) evaluated in the direction of the spatial basis vector (0,ei)(0, e_i).

theorem

Existence of a Schwartz function with ∫η(t,0) dt≠0\int \eta(t, 0) \, dt \neq 0

There exists a Schwartz function η\eta defined on spacetime Time×Rd\text{Time} \times \mathbb{R}^d such that its integral over time at the spatial origin 0∈Rd0 \in \mathbb{R}^d is non-zero, i.e., ∫Timeη(t,0) dt≠0.\int_{\text{Time}} \eta(t, 0) \, dt \neq 0.

theorem

Distributional Continuity Equation ∂tρ+∇⋅j=0\partial_t \rho + \nabla \cdot \mathbf{j} = 0 from Maxwell's Equations

Let dd be the spatial dimension and F\mathcal{F} be a free space environment with speed of light cc. Suppose AA is a distributional electromagnetic potential and JJ is a distributional Lorentz current density such that AA is an extremum of the electromagnetic Lagrangian density (i.e., Maxwell's equations are satisfied). Then the distributional charge density ρ\rho and spatial current density j\mathbf{j} satisfy the continuity equation. Specifically, for any Schwartz test function ε\varepsilon on spacetime Time×Spaced\text{Time} \times \text{Space}^d: ∂ρ∂t(ε)+(∇⋅j)(ε)=0\frac{\partial \rho}{\partial t}(\varepsilon) + (\nabla \cdot \mathbf{j})(\varepsilon) = 0 where ρ=1cJ0\rho = \frac{1}{c} J^0 is the distributional charge density and j\mathbf{j} is the distributional spatial current density.

definition

JJ is a junction of thin wires with currents IkI_k and directions u⃗k\vec{u}_k

A distributional Lorentz current density JJ is a **junction of thin wires** with directions u⃗k∈Rd\vec{u}_k \in \mathbb{R}^d and steady currents Ik∈RI_k \in \mathbb{R} (where kk ranges over a finite index set ι\iota) if, given the speed of light cc, its spatial current density distribution j⃗\vec{j} satisfies: ⟨j⃗,η⟩=∑k∈ι(Ik∫0∞ηˉ(su⃗k) ds)u⃗k\langle \vec{j}, \eta \rangle = \sum_{k \in \iota} \left( I_k \int_0^\infty \bar{\eta}(s \vec{u}_k) \, ds \right) \vec{u}_k for all Schwartz test functions η∈S(Time×Spaced,R)\eta \in \mathcal{S}(\text{Time} \times \text{Space}_d, \mathbb{R}), where ηˉ(x)=∫Timeη(t,x) dt\bar{\eta}(x) = \int_{\text{Time}} \eta(t, x) \, dt is the time-integrated test function. Physically, this defines JJ such that each wire kk is a semi-infinite ray s↦su⃗ks \mapsto s \vec{u}_k for s>0s > 0 carrying a constant current IkI_k flowing away from the origin.

theorem

∇⋅j=(∑Ik)δ0\nabla \cdot \mathbf{j} = (\sum I_k) \delta_0 for a Junction of Thin Wires

Let dd be the spatial dimension and cc be the speed of light. Suppose JJ is a distributional Lorentz current density representing a junction of thin wires with non-zero direction vectors u⃗k∈Rd\vec{u}_k \in \mathbb{R}^d and steady currents Ik∈RI_k \in \mathbb{R} for kk in a finite index set ι\iota. For any Schwartz test function η\eta defined on spacetime Time×Spaced\text{Time} \times \text{Space}_d, the spatial divergence ∇⋅j\nabla \cdot \mathbf{j} of the spatial current density j\mathbf{j} satisfies: ⟨∇⋅j,η⟩=(∑k∈ιIk)∫Timeη(t,0) dt\langle \nabla \cdot \mathbf{j}, \eta \rangle = \left( \sum_{k \in \iota} I_k \right) \int_{\text{Time}} \eta(t, \mathbf{0}) \, dt where 0\mathbf{0} is the origin in Spaced\text{Space}_d. This identifies the divergence as a point source at the origin with a strength equal to the total current leaving the junction.

theorem

The sum of steady currents ∑Ik\sum I_k at a distributional wire junction is zero

Consider a junction of thin wires in a dd-dimensional space, where each wire kk in a finite set ι\iota has a non-zero direction vector u⃗k∈Rd\vec{u}_k \in \mathbb{R}^d and carries a steady current Ik∈RI_k \in \mathbb{R}. Let JJ be the distributional Lorentz current density associated with this junction. Suppose there exists a distributional electromagnetic potential AA such that AA is an extremum of the electromagnetic Lagrangian density in free space F\mathcal{F} (implying Maxwell's equations hold). If the distributional charge density ρ=1cJ0\rho = \frac{1}{c} J^0 satisfies the condition of no charge accumulation, expressed as the vanishing of its distributional time derivative ∂tρ=0\partial_t \rho = 0, then the sum of the currents leaving the junction is zero: ∑k∈ιIk=0\sum_{k \in \iota} I_k = 0