InteriorEnergyBridge #
Ported from scottnarmstrong/Superdiffusion at 7f20476385d4af418f110303085ad0740c9eba46. Module imports, support namespace and local option directives adapted for Lean Pool.
The energy leg of the interior clause: child frame versus parent frame #
The coarse-graining energy term of the child-frame composition carries CoarseGraining's coefficient-energy norm at the child's own frame,
‖∇u_tilde_x‖_{ã, L²(□_n)} = h1EnergyNormOnCube □_n ã u_tilde_x , u_tilde_x = u(· + x) ,
localizedCoeffEnergyValue ((x−z) + □_n) ã_{L,n+2} u_tilde_z , u_tilde_z = u(· + z) .
The two agree exactly. Two facts, both identities:
- the flux increment is antisymmetric, so the quadratic form of the shifted
coefficient is the quadratic form of
a_L, i.e. exactlyν |ξ|²(AntisymmetricShiftCutoff.vecDot_matVecMul_fluxCorrectedField_self,NuIdentification); the same is true of the re-based family, whose subtracted constant is the parent's increment. So both sides areν ⨍ |∇u|²on their own window; - only the volume average has to move, and Lebesgue measure is translation
invariant (
TranslationTransportNorms), so⨍_{(x−z)+□_n} |∇u(·+z)|²is⨍_{□_n} |∇u(·+x)|².
No constant, no hypothesis, no good event.
References #
- ABK26, (the
ν|∇u|²energy reading).
1. The ν-identification at the re-based family #
The symmetric part of the re-based child-frame coefficient is exactly
ν Id: the subtracted constant is the parent's antisymmetric increment.
The ν-identification at the re-based family. On any window, the
localized coefficient energy is exactly ν times the normalized gradient L²
square.
The coefficient-energy norm of the re-based family on the child cube.
2. The volume average moves #
The gradient energy average transports between the two frames.
The normalized average of |∇u(· + z)|² over (x−z) + □_n is the normalized
average of |∇u(· + x)|² over □_n. Exact.
3. The energy leg #
The energy leg of the interior clause.
‖∇u_tilde_x‖_{ã, L²(□_n)} = ( ⨍_{(x−z)+□_n} ∇u_tilde_z · ã_{L,n+2} ∇u_tilde_z )^{1/2} .