Weak-strong uniqueness of incompressible magneto-viscoelastic flows (Q2308275)

The author proves a weak-strong uniqueness property of solutions to a hydrodynamic system that models the dynamics of incompressible magneto-viscoelastic flows with periodic conditions. He considers the problem \(v_{t}+(v\cdot \nabla )v-\mu \Delta v+\nabla p=\nabla \cdot (FF^{T})-\nabla \cdot (\nabla ^{T}M\nabla M)+\nabla ^{T}H_{\mathrm{ext}}M\), \(\nabla \cdot v=0\), \(F_{t}+(v\cdot \nabla )F=\nabla vF\); \(M_{t}+(v\cdot \nabla )M=\Delta M-\frac{1}{\alpha ^{2}}(\left\vert M\right\vert ^{2}-1)M+H_{\mathrm{ext}}\), posed in \(\mathbb{T}^{d}\times (0,T)\), \(d=2,3\), where \(v\) is the velocity field, \(p\) the pressure, \(F\) the deformation gradient, \(M\) the magnetization vector and \(H_{\mathrm{ext}}\) the external effective magnetic field. The fluid viscosity \(\mu \) is supposed to be a positive constant and \(\alpha >0\) is a parameter that controls the strength of penalization on the deviation of \(\left\vert M\right\vert \) from 1. The author introduces the strain tensor \(E=F-I\) , where \(I\) is the identity matrix, and he rewrites the above problem with \(E\). Initial data \(v_{0},E_{0},M_{0}\) are imposed on \(v\), \(E\) and \(M\). The author recalls the energy law for a classical solution to this problem. He introduces the notion of weak solution as a triple \((v,E,M)\) such that \(v\in L^{\infty }(0,T;L^{2}(\mathbb{T}^{d}))\cap L^{2}(0,T;H^{1}(\mathbb{T}^{d}))\), \(E\in L^{\infty }(0,T;L^{2}(\mathbb{T}^{d}))\), \(M\in L^{\infty}(0,T;H^{1}(\mathbb{T}^{d}))\cap L^{2}(0,T;H^{2}(\mathbb{T}^{d}))\) which satisfies a variational formulation of the problem for all test functions \(\varphi ,\psi ,\eta \in C^{1}([0,T];\mathcal{D}(\mathbb{T}^{d}))\), with \(\nabla \cdot \varphi =0\). The main result of the paper proves that if \((v,E,M)\) is a strong solution of the problem with initial data \(v_{0},E_{0},M_{0}\) respectively belonging to \(H^{s}(\mathbb{T}^{d}),H^{s}(\mathbb{T}^{d})\) and \(H^{s+1}(\mathbb{T}^{d})\), with \(s\geq 3\) and if \((\overline{v},\overline{E},\overline{M})\) is a weak solution of the problem with the same initial data \(v_{0},E_{0},M_{0}\), then \((v,E,M)=(\overline{v}, \overline{E},\overline{M})\). The author quotes from one of his preceding papers [Discrete Contin. Dyn. Syst. 38, No. 9, 4637--4655 (2018; Zbl 1403.35249)] the existence of strong solutions to this problem. The main tool of the proof is based on the energy associated to the difference between the strong and weak solutions and on estimates of the different terms of this energy. The author here uses direct computations, Sobolev embeddings and Gronwall's lemma.