Documentation

TauCeti.Analysis.Green.WindingWeight.Convergence

L¹ convergence of winding weights #

Uniform convergence of closed absolutely continuous curves, together with L¹ convergence of their derivatives, gives global L¹ convergence of their winding weights. The estimate uses translation continuity of a compactly truncated complex reciprocal kernel.

theorem TauCeti.Green.tendsto_integral_abs_windingWeight_sub {γ : ℝ → ℂ} {a b : ℝ} {γn : ℕ → ℝ → ℂ} (hab : a ≤ b) (hγn : ∀ (n : ℕ), AbsolutelyContinuousOnInterval (γn n) a b) (hγ : AbsolutelyContinuousOnInterval γ a b) (hγnclosed : ∀ (n : ℕ), γn n a = γn n b) (hγclosed : γ a = γ b) (hconv : TendstoUniformlyOn γn γ Filter.atTop (Set.Icc a b)) (hderiv : Filter.Tendsto (fun (n : ℕ) => ∫ (t : ℝ) in a..b, ‖deriv (γn n) t - deriv γ t‖) Filter.atTop (nhds 0)) :
Filter.Tendsto (fun (n : ℕ) => ∫ (z : ℂ), |windingWeight (γn n) a b z - windingWeight γ a b z|) Filter.atTop (nhds 0)

Uniform convergence of closed absolutely continuous curves and L¹ convergence of their derivatives imply global L¹ convergence of their winding weights.

theorem TauCeti.Green.tendsto_integral_abs_sum_windingWeight_sub {ι : Type u_1} {s : Finset ι} {γ : ι → ℝ → ℂ} {γn : ℕ → ι → ℝ → ℂ} {a b : ι → ℝ} {w : ι → ℤ} (hab : ∀ i ∈ s, a i ≤ b i) (hγ : ∀ i ∈ s, AbsolutelyContinuousOnInterval (γ i) (a i) (b i)) (hγn : ∀ (n : ℕ), ∀ i ∈ s, AbsolutelyContinuousOnInterval (γn n i) (a i) (b i)) (hclosed : ∀ i ∈ s, γ i (a i) = γ i (b i)) (hclosedn : ∀ (n : ℕ), ∀ i ∈ s, γn n i (a i) = γn n i (b i)) (hconv : ∀ i ∈ s, TendstoUniformlyOn (fun (n : ℕ) => γn n i) (γ i) Filter.atTop (Set.Icc (a i) (b i))) (hdconv : ∀ i ∈ s, Filter.Tendsto (fun (n : ℕ) => ∫ (t : ℝ) in a i..b i, ‖deriv (γn n i) t - deriv (γ i) t‖) Filter.atTop (nhds 0)) :
Filter.Tendsto (fun (n : ℕ) => ∫ (z : ℂ), |∑ i ∈ s, ↑(w i) * windingWeight (γn n i) (a i) (b i) z - ∑ i ∈ s, ↑(w i) * windingWeight (γ i) (a i) (b i) z|) Filter.atTop (nhds 0)

Per-curve uniform and derivative L¹ convergence imply L¹ convergence of the actual total real winding weight. No uniform bound on the indices is assumed.