The closed-graph step for weak Sobolev spaces #
This file packages the successor step shared by the iterated weak Sobolev spaces. Given a
seminormed space X and a continuous linear map from X to an Lᵖ space of F-valued fields,
TauCeti.WeakDerivStep adjoins an Lᵖ weak Fréchet derivative of that field. The admissibility
condition is a closed subspace, so the resulting graph space is complete whenever X is. The
construction works on any real normed domain with measurable opens and a measure finite on compact
sets. Its intrinsic weak-derivative characterization additionally needs local finiteness of the
measure restricted to the open set, while finite dimensionality is needed only for extensionality
via uniqueness of weak derivatives. The constructor takes an existing weak derivative and needs no
local-finiteness assumption.
The construction is independent of the order of differentiation. It is iterated by Wkp,
which starts from the weak gradient and adjoins one weak derivative per order; the projections,
constructors, norm bounds, and completeness there specialize the results proved here.
The graph carries the Euclidean (WithLp 2) product norm
(∥x∥_X² + ∥Du∥_p²)¹⁄²,
whose exponent is two whatever p is. That is what makes the squared-norm identity
TauCeti.WeakDerivStep.norm_sq_eq_norm_prev_sq_add_norm_weakFDeriv_sq available for every p.
Main declarations #
TauCeti.weakDerivStepSubmodule: the closed graph of one weak-derivative step.TauCeti.mem_weakDerivStepSubmodule_iff_hasWeakFDerivOn: its intrinsic characterization.TauCeti.WeakDerivStep: the resulting complete seminormed space, with projectionsTauCeti.WeakDerivStep.prevandTauCeti.WeakDerivStep.weakFDeriv, constructorTauCeti.WeakDerivStep.mk, and extensionalityTauCeti.WeakDerivStep.ext.
References #
The iterated weak-derivative definition and the closed-graph completeness argument follow L. C. Evans, Partial Differential Equations, Chapter 5, §5.2.
The ambient graph space obtained by adjoining an Lᵖ candidate weak derivative to X,
carrying the Euclidean (WithLp 2) product norm of its two components. They are projected out
by WithLp.fst and WithLp.snd.
Equations
- TauCeti.WeakDerivStepJetLp mu Omega p X F = WithLp 2 (X × ↥(MeasureTheory.Lp (E →L[ℝ] F) p (mu.restrict ↑Omega)))
Instances For
The closed subspace in which the adjoined field is the weak derivative of base x.
Equations
- TauCeti.weakDerivStepSubmodule mu Omega p base = ⨅ (phi : TestFunction Omega ℝ ⊤), ⨅ (v : E), ClosedSubmodule.comap (TauCeti.weakDerivStepTestFunctional✝ base phi v) ⊥
Instances For
Membership in the weak-derivative step is the family of integration-by-parts identities.
A jet is in the closed graph exactly when its last field is the weak derivative of the
field selected by base.
One weak-derivative graph step over the field selected by base. It is complete when X is
complete.
Equations
- TauCeti.WeakDerivStep mu Omega p base = ↑(TauCeti.weakDerivStepSubmodule mu Omega p base)
Instances For
The continuous projection to the preceding graph space.
Equations
- TauCeti.WeakDerivStep.prevL base = WithLp.fstL 2 ℝ X ↥(MeasureTheory.Lp (E →L[ℝ] F) p (mu.restrict ↑Omega)) ∘SL (↑(TauCeti.weakDerivStepSubmodule mu Omega p base)).subtypeL
Instances For
The preceding graph-space component.
Equations
- TauCeti.WeakDerivStep.prev base u = (TauCeti.WeakDerivStep.prevL base) u
Instances For
The continuous projection to the adjoined weak Fréchet derivative.
Equations
- TauCeti.WeakDerivStep.weakFDerivL base = WithLp.sndL 2 ℝ X ↥(MeasureTheory.Lp (E →L[ℝ] F) p (mu.restrict ↑Omega)) ∘SL (↑(TauCeti.weakDerivStepSubmodule mu Omega p base)).subtypeL
Instances For
The adjoined Lᵖ weak Fréchet derivative.
Equations
- TauCeti.WeakDerivStep.weakFDeriv base u = (TauCeti.WeakDerivStep.weakFDerivL base) u
Instances For
Construct an element of a weak-derivative graph from its two components.
Equations
- TauCeti.WeakDerivStep.mk base x D h = ⟨WithLp.toLp 2 (x, D), ⋯⟩
Instances For
The adjoined field is the weak derivative of the field selected by base.
Two elements of a weak-derivative graph are equal when their preceding components and their adjoined weak derivatives are equal.
Two elements of a weak-derivative graph are equal when their preceding components are equal: uniqueness of the weak derivative then forces the adjoined components to agree.
The graph norm controls the preceding component.
The graph norm controls the adjoined weak derivative.
The squared graph norm is the sum of the squared component norms.
Convergence in a weak-derivative graph step is equivalent to convergence of the preceding component and of the adjoined weak derivative.
A weak-derivative graph step over a complete preceding space is complete because it is a closed subspace.