Documentation

LeanPool.AnomalousDiffusion.AVenhance.Infra.Section5.RelativeError.TraceInstanceFrozen

Stream-function and diffusivity bounds parameter instance #

theorem AVenhance.Infra.Section5.RelativeError.TraceInstanceFrozen.e44_later_scale_comparisons {e p δ R : ℝ} (he : 0 < e) (he1 : e ≤ 1) (hδ : 0 < δ) (hlater : e ^ (p - δ) ≤ R) :
e ^ p ≤ R ∧ e ^ (-δ) ≤ R / e ^ p

The later-scale hypothesis gives the two radius comparisons used by the parameter absorption and the cutoff recombination.

Derivative cutoff order used to control the E44 trace tail.

Equations
Instances For

    Amplitude required for the E44 trace frequency cutoff.

    Equations
    • One or more equations did not get rendered due to their size.
    Instances For

      Fixed positive-word trace constant generated from the diffusivity bounds lower constant. Once c is chosen from l_recurse β C₀, this constant is independent of the Ingredients instance.

      Equations
      • One or more equations did not get rendered due to their size.
      Instances For
        theorem AVenhance.Infra.Section5.RelativeError.TraceInstanceFrozen.e44_trace_rpow_le_nat_power {ε p δ : ℝ} (hε : 0 < ε) (hε1 : ε ≤ 1) (n : ℕ) (hn : p ≤ δ * ↑n) :
        ε.rpow (-p) ≤ ε.rpow (-δ) ^ n
        theorem AVenhance.Infra.Section5.RelativeError.e44_trace_numeric_absorption {β c ε : ℝ} (hβ : 1 < β) (hβ' : β < 4 / 3) (hc : 0 < c) (hε : 0 < ε) (hε1 : ε ≤ 1) :
        theorem AVenhance.Infra.Section5.RelativeError.TraceInstanceFrozen.e44_natural_cutoff_data {r R A : ℝ} (hr : 0 < r) (hrR : r ≤ R) (hr1 : r ≤ 1) (hA : 32 ≤ A) :
        ∃ (M : ℕ), 0 < M ∧ 2 * Real.pi * √2 * ↑M ≤ A / r ∧ A / (2 * r) ≤ 2 * Real.pi * √2 * ↑M ∧ A / r ≤ 2 * Real.pi * √2 * (↑M + 1) ∧ 1 ≤ R / 2 * (↑M + 1)

        The floor cutoff realizes the two-sided Fourier radius estimates needed for the low-frequency trace and the analytic tail.

        theorem AVenhance.Infra.Section5.RelativeError.e44_streamVelocity_spatialDerivative_bound_of_A3 {β : ℝ} (I : Ingredients β) {Φ : ℕ → ℝ → Homogenization.Vec 2 → ℝ} (hΦ : IsStreamSeq I Φ) {m : ℕ} (hm : 2 ≤ m) (hA3 : ∀ (j : ℕ), 1 ≤ j → ∀ (t : ℝ) (n : ℕ), 2 ≤ n → barNorm n (2 ^ 8 * (epsilon β I.Λ j)⁻¹) (Φ j t) ≤ ENNReal.ofReal (2 ^ 5 * AVenhance.a β I.Λ j * epsilon β I.Λ j ^ 2 * ((↑n + 2) ^ 2 / (↑n + 1) ^ 3))) (t : ℝ) (x : Homogenization.Vec 2) :
        ‖Flow.jointSpatialFDeriv (streamVel fun (s : ℝ) (y : Homogenization.Vec 2) => Φ (m - 1) s y) t x‖ ≤ 2 ^ 22 * a β I.Λ (m - 1)

        The stream-function estimates second-jet bound gives the actual drift gradient bound used by the low-mode trace estimate.

        theorem AVenhance.Infra.Section5.RelativeError.e44_initialTrace_of_A3_A5_data {β : ℝ} (I : Ingredients β) {Φ : ℕ → ℝ → Homogenization.Vec 2 → ℝ} (hΦ : IsStreamSeq I Φ) {N m : ℕ} (hm : 2 ≤ m) (hmN : m ≤ N) {κ : ℝ} (c C : ℝ) (hc : 0 < c) (_hcC : c < C) (hA3 : ∀ (j : ℕ), 1 ≤ j → ∀ (t : ℝ) (n : ℕ), 2 ≤ n → barNorm n (2 ^ 8 * (epsilon β I.Λ j)⁻¹) (Φ j t) ≤ ENNReal.ofReal (2 ^ 5 * AVenhance.a β I.Λ j * epsilon β I.Λ j ^ 2 * ((↑n + 2) ^ 2 / (↑n + 1) ^ 3))) (hA5 : ∀ (j : ℕ), 1 ≤ j → j < N → c * (AVenhance.a β I.Λ j * epsilon β I.Λ j ^ (2 + gamma β)) ≤ I.kappaAt κ j (N - j) ∧ I.kappaAt κ j (N - j) ≤ C * (AVenhance.a β I.Λ j * epsilon β I.Λ j ^ (2 + gamma β))) {g : Homogenization.Vec 2 → ℝ} {u : ℝ → Homogenization.Vec 2 → ℝ} {R : ℝ} (hsol : IsClassicalSol (streamVel fun (t : ℝ) (y : Homogenization.Vec 2) => Φ (m - 1) t y) (I.kappaSeq κ N (m - 1)) (fun (x : ℝ) (x_1 : Homogenization.Vec 2) => 0) g u) (hθ : IsThetaAnalytic R g) (hmean : ∫ (x : Homogenization.Vec 2) in unitCube, g x = 0) (hκ1 : I.kappaSeq κ N (m - 1) ≤ 1) (hLater : epsilon β I.Λ (m - 1) ^ (1 + gamma β / 2 - delta β) ≤ R) (A : ℝ) (hA1 : 1 ≤ A) (hA5scale : 2 ^ 24 ≤ c * A ^ 2) (Mcut : ℕ) (hMcut : 0 < Mcut) (hKup : 2 * Real.pi * √2 * ↑Mcut ≤ A / epsilon β I.Λ (m - 1) ^ (1 + gamma β / 2)) (hKlower : A / (2 * epsilon β I.Λ (m - 1) ^ (1 + gamma β / 2)) ≤ 2 * Real.pi * √2 * ↑Mcut) (hKupperTail : A / epsilon β I.Λ (m - 1) ^ (1 + gamma β / 2) ≤ 2 * Real.pi * √2 * (↑Mcut + 1)) (hTailScale : 1 ≤ R / 2 * (↑Mcut + 1)) (hNumeric : (√c)⁻¹ * (epsilon β I.Λ (m - 1)).rpow (-(β + gamma β) / 2) * Real.exp (-e44TailExponent * A * (epsilon β I.Λ (m - 1)).rpow (-delta β)) ≤ 1) :

        Concrete stream-function and diffusivity bounds trace instance. hNumeric is the explicit large-Λ absorption threshold; the constants and cutoff radius stay visible in the statement so the later-scale and profile-radius checks cannot be hidden.

        Identify the classical slice-energy integral with the space-time gradient norm used by the initial-trace premise.

        theorem AVenhance.Infra.Section5.RelativeError.e44_initialTrace_of_explicit_A3_A5 {β : ℝ} (I : Ingredients β) {Φ : ℕ → ℝ → Homogenization.Vec 2 → ℝ} (hΦ : IsStreamSeq I Φ) {N m : ℕ} (hm : 2 ≤ m) (hmN : m ≤ N) {κ : ℝ} (c C : ℝ) (hc : 0 < c) (hcC : c < C) (hA3 : ∀ (j : ℕ), 1 ≤ j → ∀ (t : ℝ) (n : ℕ), 2 ≤ n → barNorm n (2 ^ 8 * (epsilon β I.Λ j)⁻¹) (Φ j t) ≤ ENNReal.ofReal (2 ^ 5 * AVenhance.a β I.Λ j * epsilon β I.Λ j ^ 2 * ((↑n + 2) ^ 2 / (↑n + 1) ^ 3))) (hA5 : ∀ (j : ℕ), 1 ≤ j → j < N → c * (AVenhance.a β I.Λ j * epsilon β I.Λ j ^ (2 + gamma β)) ≤ I.kappaAt κ j (N - j) ∧ I.kappaAt κ j (N - j) ≤ C * (AVenhance.a β I.Λ j * epsilon β I.Λ j ^ (2 + gamma β))) {g : Homogenization.Vec 2 → ℝ} {u : ℝ → Homogenization.Vec 2 → ℝ} {R : ℝ} (hsol : IsClassicalSol (streamVel fun (t : ℝ) (y : Homogenization.Vec 2) => Φ (m - 1) t y) (I.kappaSeq κ N (m - 1)) (fun (x : ℝ) (x_1 : Homogenization.Vec 2) => 0) g u) (hθ : IsThetaAnalytic R g) (hmean : ∫ (x : Homogenization.Vec 2) in unitCube, g x = 0) (hκ1 : I.kappaSeq κ N (m - 1) ≤ 1) (hLater : epsilon β I.Λ (m - 1) ^ (1 + gamma β / 2 - delta β) ≤ R) :
        1 ≤ e44FrozenTraceConstant β c ∧ ∀ (w : List (Fin 2)), 1 ≤ w.length → √(∫ (x : Homogenization.Vec 2) in unitCube, Section4.classicalWordDerivative w g x ^ 2) ≤ √(I.kappaSeq κ N (m - 1)) * √(spaceTimeGradNormSq fun (t : ℝ) => spaceGrad (u t)) * (↑w.length.factorial / (epsilon β I.Λ (m - 1) ^ (1 + gamma β / 2) / e44FrozenTraceConstant β c ^ 2) ^ w.length)

        The same trace construction with the diffusivity bounds constants supplied explicitly. This is the family-level form: callers may choose c,C before the Ingredients instance and provide the corresponding stream-function and diffusivity bounds.