Variational methods in nonconservative phenomena (Q1801231): Difference between revisions

The book is devoted to variational problems that arise in connection with nonconservative systems. The authors declare in the preface: ``We hope that this book will serve to dispel the notion that nonconservative mechanics have no Lagrangian structure. We also hope that the reader will believe, as we do, that there is a common basis for both conservative and nonconservative Lagrangian dynamics''. Another basic sentence in order to understand the purpose of the book is contained in the introduction of Chapt. 2: ``The methods by which approximate solutions to nonlinear nonconservative systems can be estimated by means of a variational principle are virtually the subject of this book.'' Let us follow the steps indicated by the authors to attain the purpose. Chapt. 1 contains a synthesis of the variational foundations of the analytical dynamics, namely the D'Alembert's principle in variational form, and the Hamilton's principle. Chapt. 2, starts with the consideration that for conservative systems a Lagrangian \(L=T-\pi\) can be introduced, with T and \(\pi\), respectively kinetic and potential energy; on the contrary for nonconservative systems this construction is not directly useful for finding Lagrangians. However also for nonconservative systems Lagrangians are available, although they generally have no energy interpretation. In addition it must be remarked that to a given dynamical system not only one, but more Lagrangians can be associated. Then the problem arises of constructing Lagrangians for a given system, that is the inverse problem of the calculus of variations. General methods for solving these problems are discussed: for systems with one degree of freedom a plurality of solutions can be obtained in any case; for systems with many degrees of freedom necessary and sufficient conditions are given for which the system is derivable from a Lagrangian. Several applications having mathematical, physical and engineering interest are developed and discussed. In Chapt. 3, the longest and may be the most ardous of the book, conservative and nonconservative dynamical systems with a finite number of degrees of freedom are studied. Starting from the concept of simultaneous and nonsimultaneous variations, infinitesimal transformations, absolute and gauge invariance, the reader is conducted to the conditions of invariance of the Hamilton action integral, to the Noether theorem, to the Killing equations, to the differential variational principles of Gauss and Jourdain, and to the conservative laws of Schul'gin and Painlevé. Two simple but expressive examples are: for the linear damped oscillator mẍ\(+\mu \dot x+cx=0\), the following conservation law is obtained: \(((\dot x^ 2/2)+ (\omega^ 2x^ 2/2)+ kx\dot x)e^{2kt}=\)const., where \(\omega^ 2=c/m\) and \(2k=\mu /m.\) For the Emden-Fowler equation \(t\ddot x+2\dot x+at^ nx^{2n+3}=0\) (a const.) the conservative law \(t^ 2\dot x^ 2+t^ 3x\dot x+(a/n+1)t^{n+2}x^{2n+4}=const.\) Chapt. 4 is devoted to a study of the motion of conservative and nonconservative dynamical system by means of field theory. Firstly the classical Hamilton-Jacobi field method is discussed for finding the motion of Hamiltonian systems. In the second part a field method is established for studying nonconservative, holonomic dynamical systems that do not posses Hamilton's action integral, and the Hamilton-Jacobi field equation cannot be employed. Namely the following system is considered: \[ (*)\quad \dot x_ 1=X_ 1(t,x_ 1,...,x_ n),...,\dot x_ n=X_ n(t,x_ 1,...,x_ n). \] One coordinate of the system, say \(x_ 1\), is selected as the basic field, depending on time t and the rest of the coordinates: \(x_ 1=\Phi (t,x_ 2,...,x_ n)\). Differentiating this expression totally with respect to time the following basic field equation can be obtained: \[ (\partial \Phi /\partial t)+(\partial \Phi /\partial x_{\alpha})X_{\alpha}(t,\Phi,x_ 2,...,x_ n)-X_ 1(t,\Phi,x_ 2,...,x_ n)=0\quad (\alpha =2,3,...,n), \] which plays the same role as the classical Hamilton-Jacobi partial differential equation. In a similar way, for the differential equation of motion written in nonconservative Hamiltonian form: \[ \dot x^ i=\partial H/\partial p_ i,\quad \dot p_ i=-(\partial H/\partial x^ i)+Q_ i(t,x^ 1,....,x^ n,p_ 1,...,p_ n), \] the corresponding basic field equation is obtained by a variational technique. The general properties and the complete solutions of the basic field equations are investigated. For the same system (*) also the so called potential method of Arzlianik is presented, developed and discussed for many applications. After this first part of the book, devoted to systems with a finite number of degrees of freedom, a second part follows with attention to the variational description of nonconservative (or irreversible) phenomena whose physical manifestations are described by means of partial differential equations and appropriate initial and boundary conditions. Chapt. 5 is devoted to variational principles with vanishing parameters, starting from the consideration that the classical parabolic Fourier equation for heat conduction is modified by many authors with an additional term involving the second time derivative: \(\tau \partial^ 2T/\partial t^ 2+\partial T/\partial t=a\nabla^ 2T\), in order to remove the paradox of an infinite velocity for the thermal disturbance. This equation, multiplied by exp(t/\(\tau\)) can be obtained from a Hamilton variational principle with Lagrangian: \(L=[(\tau/2)(\partial T/\partial t)^ 2-(a/2)\sum^{3}_{i=1}(\partial T/\partial x_ i)^ 2]e^{t/\tau}\). This general procedure applies also to other problems by means of defining a Lagrangian depending on a small parameter \(\tau\), then applying a variational principle, dividing by a factor depending on \(\tau\), and finally performing the limit as \(\tau\to 0\). Many examples are developed and discussed, also with application to boundary layer theory of incompressible flows. Chapt. 6 deals with variational principles with noncommutative rules and their application to nonconservative phenomena. Said principles are characterized by the fact that the well-known rule ``The variation of velocity is equal to the derivative of the variation'' is no more true. The applications refer to the Bogolyubov-Krylov-Mitropolsky method in nonlinear vibration analysis, and to the heat conduction in solids. Finally, Chapt. 7 is devoted to the application of Gauss's principle of last constraint to nonconservative phenomena, as transient, nonlinear heat conduction through prism-like infinite bodies, melting or freezing of a semi-infinite solid, a nonconservative convective problem. As a conclusion we recommend the lecture of this book owing to the interest, both theoretical and applicative, for the variety and to the clearness of the treated topics, in the prospect of enlarging the ambit in which, as a rule, the classical variational methods are acting.