Alphabeta Math
CorollaryStatement: Literature-sourcedProof: AI-adaptedPipeline-generatedprecheck pass
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

A midpoint-quadrature P1 solve has an explicit Strang error budget

Statement

Assume Countable Choice. Put V:=H01(0,1;R) with its H1 norm, and let Vh⊆V be a finite-dimensional conforming piecewise-linear space on an interval partition with maximum cell width h. Let u∈C2([0,1])∩V and assume its nodal interpolant Ihu belongs to Vh. Let A∈C2([0,1]) satisfy 0<α≤A(x)≤M, and write CA:=∥A′′∥L∞(0,1). Define a(v,w)=∫01Av′w′ and let ah use one midpoint quadrature sample per cell. Let F and Fh be bounded real linear exact and discrete loads, with F(v)=a(u,v) for every v∈V, and put δF,h:=sup⁡wh≠0∣F(wh)−Fh(wh)∣/∥wh∥H1, with δF,h=0 if Vh={0}. Then the exact and discrete solutions satisfy

∥u−uh∥H1≤(1+2M/α)2h∥u′′∥∞+2α(CAh224(∥u′∥2+h∥u′′∥∞)+δF,h).

The explicit δF,h is the data/source oscillation term; no rate or vanishing of it is asserted without an additional load approximation hypothesis. Here coercivity follows directly because every sampled coefficient satisfies A(mK)≥α; this special argument does not infer stability from quadrature exactness.

Facts & Assumptions

Given: Countable Choice; V=H01(0,1); a conforming finite-dimensional P1 space Vh; u∈C2∩V with Ihu∈Vh; A∈C2 with α≤A≤M; and bounded loads F,Fh.

[F1]

The Sobolev space H1 is a Hilbert space and Zero-boundary Sobolev space as a norm closure: under Countable Choice, H01(0,1) is a Hilbert space for its H1 inner product.

[F2]

Zero-boundary Poincaré inequality on the unit interval: ∥v∥2≤∥v′∥2 for v∈H01(0,1).

[F3]

One-dimensional P1 interpolation has a first-derivative error bound: ∥(u−Ihu)′∥2≤h∥u′′∥∞.

[F4]

Midpoint quadrature defect for one-dimensional P1 stiffness: ∣a(vh,wh)−ah(vh,wh)∣≤CAh2∥vh′∥2∥wh′∥2/24, with ∥vh′∥22=∑KhK∣vh′∣K2 and the analogous identity for every wh∈Vh.

[F5]

First Strang estimate for conforming perturbed Galerkin problems: error is bounded by best approximation plus the form and load defects divided by discrete coercivity.

[A1]

The exact and discrete equations are a(u,v)=F(v) and ah(uh,wh)=Fh(wh), and δF,h is as in the Statement.

Proof

1.1F1F2F6givenalgebra

By [F2], ∥v∥H12=∥v∥22+∥v′∥22≤2∥v′∥22. Holder [F6] gives ∣a(v,w)∣≤M∥v′∥2∥w′∥2≤M∥v∥H1∥w∥H1, while a(v,v)≥α∥v′∥22≥(α/2)∥v∥H12. Thus a is bounded and coercive in the inherited H1 norm.

1.2F2F4F6givenalgebra

The norm identity in [F4] and finite Cauchy–Schwarz [F6] give, for vh,wh∈Vh, ∣ah(vh,wh)∣≤M∑KhK∣vh′∣K∣wh′∣K≤M∥vh′∥2∥wh′∥2≤M∥vh∥H1∥wh∥H1, so ah is bounded. Also A(mK)≥α on each cell, so ah(vh,vh)=∑KhKA(mK)∣vh′∣K2≥α∥vh′∥22. By [F2], ∥vh∥H12≤2∥vh′∥22, hence ah(vh,vh)≥(α/2)∥vh∥H12 and αh≥α/2.

1.3F2F3givenalgebra

By [F3] and [F2], ∥u−Ihu∥H1≤2∥(u−Ihu)′∥2≤2h∥u′′∥∞. Also the triangle inequality gives ∥(Ihu)′∥2≤∥u′∥2+h∥u′′∥∞.

2.1F4F5A1step 1.3

Apply [F5] with comparison function Ihu. Its best-approximation term is bounded by step 1.3; [F4] bounds its form defect by CAh2(∥u′∥2+h∥u′′∥∞)/24 because ∥wh′∥2≤∥wh∥H1. The load defect is exactly δF,h by [A1].

3.1step 1.1step 1.2step 2.1algebra∎

Substitute M from step 1.1 and αh≥α/2 from step 1.2 into the first-Strang bound in step 2.1. This gives the displayed estimate and retains the full load defect.

Depends on

Used by

Nothing in the library uses this result yet.

Dependency tree · two levels

75 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources