Documentation

LeanPool.HighContrastHomogenization.Provider.Regularity.AffineGradientQuotientInverse

High-contrast homogenization: Provider.Regularity.AffineGradientQuotientInverse #

Imported from the Apache-2.0 HighContrastHomogenization development at commit 7a13dbcd8d6609264a713373f5c69ceeac870472.

Inverse law for affine gradient quotient pullback #

The quotient transport is inverted on the same fixed source and target domains by the inverse coordinate matrix and inverse value action.

If V is the inverse-image of U under L, then U is the inverse-image of V under L⁻¹.

noncomputable def HCPolySupport.HighContrast.affineGradientQuotientPushforward {d : ℕ} {L : Mat d} (hL : IsUnit (Matrix.det L)) {U V : Set (Vec d)} (hU : MeasurableSet U) (hV : V = matImage L⁻¹ U) (A : Mat d) :

Push a gradient quotient from the affine inverse-image domain back to the original domain.

Equations
Instances For
    theorem HCPolySupport.HighContrast.coeFn_affineGradientQuotientPushforward {d : ℕ} {L : Mat d} (hL : IsUnit (Matrix.det L)) {U V : Set (Vec d)} (hU : MeasurableSet U) (hV : V = matImage L⁻¹ U) (A : Mat d) (F : ↥(HilbertVectorL2 V)) :
    ↑↑(affineGradientQuotientPushforward hL hU hV A F) =ᵐ[volumeMeasureOn U] fun (x : Vec d) => HilbertVec.ofVec (matVecMul A (↑↑F (matVecMul L⁻¹ x)).toVec)

    The quotient pushforward has the literal inverse-coordinate representative almost everywhere on the original domain.