No label defined (Q4944619)

The axially moving strings have been subject of much engineering research. They occur in wire drawing, band saws, and perhaps with surprising applications in modern physics. The author provides an extensive list of references. In this article, the author studies the mechanics of a string pulled at constant speed through two eyelets at constant \(x\)-distance from each other, where \(x\) is the axis parallel to the direction of the pulling force. One of the eyelets is fixed, the other can move in the \(y\) direction, where \((x,y)\) are Cartesian coordinates. The author uses Mote's equation, which has been accepted by several other authors: \[ y_{tt}(x, t)+ 2a\nu y(x,t)= [(1- (a\nu)^2+ 3\{\textstyle{{1\over 2}}by^2_x(x, t)\}] y_{xx}(x, t),\quad x\in [0,1],\quad t\geq 0. \] \(y\) is the transversal displacement, \(a\) and \(b\) are positive constants, while \(\nu\geq 0\) is proportional to the speed of the string passing through an eyelet. Physical arguments imply \(a\nu< 1\). The tension \(T\) in the string is not constant. In fact \(T= 1+{1\over 2}by^2_x\). The control force applied to the moving eyelet is given by the stabilizing effect: \(u(1,t)= -ky_t(1,t)\), which offers us the boundary condition: \[ -ky_t(1,t)= [1- (a\nu)^2+ \textstyle{{1\over 2}}by^2_x(1, t)] y_x(1,t)\;\forall t\geq 0. \] The purpose of this article is to analyze the effects of the stabilizing control force \(u(1,t)\). The Lyapunov function \[ V(t)= E(t)+ \gamma \int^1_0 [xy_t(x, t)y_x(x, t)+ a\nu y^2_x(x,t)] dx \] is introduced, where \[ E(t)= \int^1_0 [y^2_t(x, t)+ (1- (a\nu)^2) y^2_x(x, t)] dx+ \textstyle{{1\over 8}}\displaystyle{b\int^1_0 y^4_x(x, t) dx}. \] Now, it is not very hard to show that for specific positive values of \(K_1\) and \(K_2\) the following inequality is true: \(0\leq K_1E(t)\leq V(t)\leq K_2E(t)\). The author computes \(dE/dt\), and shows that the functions \(V\) and \(E\) both tend to zero together as \(t\to\infty\), and proves that \(V(t)\leq V(0)\exp\{-\gamma t/K_2\}\). Now it follows easily that the assumed control \(u(t)\) does indeed stabilize the motion of the string. This is a useful information confirming that the nonlinear system described here behaves in this respect just like some thoroughly analyzed linear vibrating systems. Perhaps the following improvements could have been incorporated into the author's analysis. Without saying so the author assumes that the deflection function \(y(x,t)\) has all kind of classical derivatives (which are elements of \(L^2\), \(L^3\), \(L^4\)). This causes the string to avoid shapes that are not sufficiently smooth for such requirements. However, such string may not know about it and may prefer a less perfect (but certainly continuous) shape that would improve the value of the Lagrangian action integral. The reviewer predicts that nothing would really change in the principal result if all the functions mentioned in this review were assigned appropriate Sobolev spaces, and weak solutions were examined first. But this prediction needs to be confirmed.