Eight(y) mathematical questions on fluids and structures (Q2281285)

The authors present eight open questions concerning turbulence problems especially in the case of aeroelastic phenomena for suspended bridges. They start with a historical description of turbulence problems, going back to Leonardo da Vinci. They present the d'Alembert paradox that they formalize as a mathematical result concerning the unforced incompressible Euler equations \(u_{t}+(u\cdot \nabla )u+\nabla p=0\), \(\nabla \cdot u=0\) posed in \( \Omega \times (0,\infty )\) where \(\Omega \) is the exterior of a smooth and compact domain \(D\subset \mathbb{R}^{3}\). They then consider the Navier-Stokes equations \(u_{t}-\nu \Delta u+(u\cdot \nabla )u+\nabla p=0\), \( \nabla \cdot u=0\) posed in \(\Omega \times (0,\infty )\) where \(\Omega =\{(-L,L)^{2}\times (0,\Lambda )\}\setminus \{\overline{K}\times (0,\Lambda )\}\) with \(L,\Lambda >0\) and \(K\) is a 2D obstacle. Considering the associated Stokes equation \(-\nu \Delta u+\nabla p=0\), \(\nabla \cdot u=0\) posed in \(\Sigma =(-L,L)^{2}\setminus \overline{K}\subset \mathbb{R}^{2}\) and the boundary conditions \((u,v)=(U,V)\) on the outer boundary \(\Gamma \) of \(\Sigma \) and \((u,v)=(0,0)\) on the inner boundary of \(\Sigma \), they recall the result proved by the two last authors in a forthcoming preprint and concerning symmetry properties of the solution, if \((U,V)\) satisfies the compatibility condition \(\int_{-L}^{L}[U(L,y)-U(-L,y)]dy+ \int_{-L}^{L}[V(x,L)-V(x,-L)]dx=0\). They also quote from this preprint the existence of a unique weak solution \((u_{1},u_{2},p)\in H^{1}(\Sigma )\times L_{0}^{2}(\Sigma )\) to the 2D steady Navier-Stokes problem \(-\nu \Delta u+(u\cdot \nabla )u+\nabla p=0\), \(\nabla \cdot u=0\), assuming that \((U,V)\in H^{1/2}(\Gamma )\). This weak solution satisfies the same symmetry properties. Assuming that the boundary \(\partial K\) of the obstacle \(K\) is \( C^{2}\), \textit{C. Foias} and \textit{R. Temam} proved in [Commun. Pure Appl. Math. 30, 149--164 (1977; Zbl 0335.35077)] the existence of a dense open set \(\mathcal{O}\subset \{(U,V)\in H^{3/2}(\Gamma ):(U,V)\) satisfies the preceding compatibility condition\(\}\) such that if \((U,V)\in \mathcal{O}\) the number of solutions of the above 2D Navier-Stokes problem is finite. Moreover, on any connected component of \(\mathcal{O}\) the number of solutions of this Navier-Stokes problem is constant. The authors raise the questions of the existence of multiple steady flows past any symmetric body for large Reynolds number and of multiple symmetric steady flows. In Section 4 of their paper, the authors analyze the flow behavior around a cylinder. They start with the presentation of experimental results which prove the importance of Reynolds number. They here raise the questions of the existence of a critical threshold of the flow velocity which initiates the lift and of a deterministic law for the dependence of the lift on the flow. Considering the evolution Navier-Stokes problem \(u_{t}-\nu \Delta u+((u-V(t))\cdot \nabla )u+\nabla p=0\), \(\nabla \cdot u=0\) posed in \(\Omega \times (0,\infty )\) where \(\Omega =\mathbb{R}^{3}\setminus D\), the obstacle \( D\) moving in a rigid motion with velocity \(V\in H^{1}(0,T)\) which is supposed to be \(T\)-periodic, with the boundary conditions \(u=V(t)\) on \( \partial D\) and \(u(x)\rightarrow 0\) and \(\left\vert x\right\vert \rightarrow \infty \) the authors quote from \textit{G. P. Galdi} and \textit{A. L. Silvestre} [Pac. J. Math. 223, No. 2, 251--267 (2006; Zbl 1109.76016)] the existence of at least one \(T\)-periodic weak solution to this problem. In Section 5, the authors focus on the behavior of floating bridges, starting with the Tacoma Narrows Bridge which collapsed in 1940. They recall the observations and discussions following this collapse. They introduce the mathematical framework associated to such floating bridges considering a domain \(D=(-l,l)\times (-d,d)\times (0,\Lambda )\) with \(d\ll l<\Lambda \), \(D_{\ast }=(-l,l)\times (0,\Lambda )\) and \(\Omega =(-L,L)^{2}\times (0,\Lambda )\setminus \overline{D}\). They consider the problem \(u_{tt}+\delta u_{t}+\Delta ^{2}u-S[\int_{D_{\ast }}u_{z}^{2}]u_{zz}=g(\xi ,t)\) in \(D_{\ast }\times (0,T)\), with the boundary conditions \(u=u_{zz}=0\) on \((-l,l)\times \{0,\Lambda \}\) and \( u_{xx}+\sigma u_{zz}=u_{xxx}+(2-\sigma )u_{xzz}\) on \(\{-l ,l\}\times (0,\Lambda )\) and with initial conditions for \(u\) and \(u_{t}\). They define the notion of torsional stability for the source term \(g\) and they prove the existence of a threshold \(g_{0}\) such that if \(\lim \sup_{t\rightarrow \infty }\left\Vert g(t)\right\Vert _{L^{2}}<g_{0}\) then this problem is torsional stable for every \(\delta >0\) and the existence of a threshold \(\delta _{0}>0\) such that if \(\delta >\delta _{0}\) this problem is torsional stable whatever \(\lim \sup_{t\rightarrow \infty }\left\Vert g(t)\right\Vert _{L^{2}}\), see \textit{D. Bonheure}, \textit{F. Gazzola} and \textit{E. Moreira dos Santos} in [SIAM J. Math. Anal. 51, No. 4, 3052--3091 (2019; Zbl 1437.35452)]. In Section 6, the authors consider the fluid-structure interaction problem for a fluid flowing around a fixed or moving body and described as \(u_{t}-\nu \Delta u+(u\cdot \nabla )u+\nabla p=0 \), \(\nabla \cdot u=0\) posed in the fluid part \(\Omega _{t}\subset A\) with the boundary conditions \(u=0\) on \(\partial A\) and \(u=h^{\prime }(t)-\omega (t)\wedge n\) on \(\partial B_{t}\), where the position \(h(t)\) of the center of the ball \(B_{t}\) satisfies \(Mh^{\prime \prime }(t)=-\int_{\partial B_{t}}\sigma n\) and its angular velocity \(\omega (t)\) satisfies \(J\omega ^{\prime }(t)=\int_{\partial B_{t}}n\wedge \sigma n\).\ Initial conditions are added for \(u\), \(h^{\prime }\) and \(\omega \). They quote from the papers by \textit{C. Conca} et al., [Commun. Partial Differ. Equations 25, No. 5--6, 1019--1042 (2000; Zbl 0954.35135)] or by \textit{D. Gerard-Varet} and \textit{M. Hillairet} in [Commun. Pure Appl. Math. 67, No. 12, 2022--2076 (2014; Zbl 1307.35193)] the existence of a local in time weak solution to this problem, under smoothness hypotheses on the data. One further has either \(T_{0}<\infty \) or \( \lim_{t\rightarrow T_{0}}\mathrm{dist}(B_{t},\partial A)=0\). In the final section 7, the authors present results concerning numerical simulations for suspended bridges and they end with the last open questions. Throughout the whole paper, the authors quote texts from papers which have been published by different scientists along the centuries. The paper ends with a long list of references.