Integral representation of the invariant hull

About 2 years old · traced to

Suppose the functional II has the integral form

I(u)=∫ΩW(∇u(x)) dxI(u)=\int_\Omega W(\nabla\mathbf{u}(\boldsymbol{x}))\,d\boldsymbol{x}

for a continuous integrand

W(F):Rm×N→R.W(\mathbf F):\mathbb{R}^{m\times N}\to\mathbb{R}.

Let IiI_i be its invariant hull with respect to an invariant class D0\mathcal D_0, and define

Wi(F)=inf⁡X∈R+N×Ndet⁡X W(1det⁡XFadj⁡XT),W_i(\mathbf F)=\inf_{\mathcal X\in\mathbb{R}^{N\times N}_+} \operatorname{det}\mathcal X\,W\left(\frac{1}{\operatorname{det}\mathcal X}\mathbf F\operatorname{adj}\mathcal X^T\right),

where R+N×N={X∈RN×N:det⁡X>0}\mathbb{R}^{N\times N}_+=\{\mathcal X\in\mathbb{R}^{N\times N}:\operatorname{det}\mathcal X>0\}. Integral representation conjecture. If IiI_i is itself an integral functional, then Ii≡Ii∗I_i\equiv I_i^*, namely

Ii(u)=∫ΩWi(∇u(x)) dx.I_i(u)=\int_\Omega W_i(\nabla\mathbf{u}(\boldsymbol{x}))\,d\boldsymbol{x}.

The proposition preceding this statement gives Ii∗≤Ii≤II_i^*\leq I_i\leq I, so the assertion identifies the invariant hull with the simpler pointwise invariant density whenever the hull admits an integral representation. The source provides no resolution of this claim.

References

Primary source

Pablo Pedregal, “Invariant Hulls and Geometric Variational Principles”, arXiv:2411.04552 (2024).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.