Abstract Generated abstract
This paper develops a generalization of Mindlin’s method for elasticity problems in a half-space by using analytic continuation of elastic solutions through a plane boundary. It formulates reflection laws for body forces that allow boundary-value problems with either vanishing displacements or vanishing stresses on the plane boundary to be represented as whole-space elasticity problems with an explicitly constructed system of reflected and additional body forces. The resulting theorem gives formulae, in terms of Poisson’s ratio, reflected force components, divergences, derivatives, and certain integrals, for the equivalent force distributions in the full space. The paper notes that some integrals must be interpreted in the Hadamard or generalized-function sense when ordinary convergence fails.
Full Text
Reports of the Academy of Sciences of the USSR
1970. Volume 194, No. 5
UDC 539.311
THEORY OF ELASTICITY
A. P. SHMAKOV
ON THE LAWS OF REFLECTION OF BODY FORCES IN THE THEORY OF ELASTICITY THROUGH A PLANE BOUNDARY
(Presented by Academician L. I. Sedov, 15 VI 1970)
It is known that some boundary-value problems of the theory of elasticity can be solved by superposition of Kelvin solutions. These include the problems of Boussinesq, Cerruti \(\left({}^{1}\right)\), and Mindlin \(\left({}^{2}\right)\). The most general of them is Mindlin’s problem on the action of a concentrated force inside an elastic half-space whose boundary is free of external actions. In the present paper a further development and generalization of Mindlin’s method is given by realizing a simple and general idea. This idea is based on the analytic continuation of solutions of the equations of the theory of elasticity through a plane boundary. Knowledge of the laws of analytic continuation (reflection) makes it possible to construct solutions of boundary-value problems in a general form and to obtain reflection laws for body forces. The problem of analytic continuation through a plane boundary was considered in works \(\left({}^{3,4}\right)\).
In work \(\left({}^{4}\right)\), analytic continuation was used to solve problems of the theory of elasticity with mixed boundary conditions. Some of the results of that work are presented in the monograph \(\left({}^{5}\right)\).
The main results of the present paper are formulated in the following theorem.
Theorem. Let in the half-space \(x_3>0\), on whose boundary \(x_3=0\) either a) displacements or b) stresses are absent, body forces \(\dot F_i\) act, with \(\dot F_i(x_1,x_2,x_3)=0\) for \(x<0\) \((i=x_1,x_2,x_3)\). Then the displacements that arise in the elastic half-space \(x_3>0\) from the action of these forces can be regarded as the solution of an elastic problem for the whole space, in which the following system of body forces acts:
\[ \begin{aligned} \text{a)}\quad F_i={}&(\dot F_i-\dot F_i') +\frac{2(1-\nu)}{(1-2\nu)(3-4\nu)}\,x_3 \left[ 2\,\frac{\partial \dot F_3'}{\partial x_i} -\frac{1}{2(1-\nu)}\,x_3 \times \right.\\ &\left. \times \frac{\partial}{\partial x_i}(\operatorname{div}\dot{\mathbf F})' \right] +\frac{2}{(1-2\nu)(3-4\nu)}\,\delta_{i3} \left[\dot F_3'-2\nu x_3(\operatorname{div}\dot{\mathbf F})'\right] \quad (i=x_1,x_2,x_3); \end{aligned} \]
\[ \begin{aligned} \text{b)}\quad F_i={}&(\dot F_i+\dot F_i') -\frac{4(1-\nu)}{1-2\nu}\,x_3\frac{\partial \dot F_3'}{\partial x_i} +4(1-\nu)\frac{\partial}{\partial x_i} \int_{x_3}^{+\infty}\dot F_3'\,dx_3 \\ &+\frac{1}{1-2\nu}x_3^2\frac{\partial}{\partial x_i} (\operatorname{div}\dot{\mathbf F})' +2(1-\nu)x_3\frac{\partial}{\partial x_i} \int_{x_3}^{+\infty}(\operatorname{div}\dot{\mathbf F})'\,dx_3 \\ &-2(1-\nu)\int_{x_3}^{+\infty}x_3(\operatorname{div}\dot{\mathbf F})'\,dx_3 \quad (i=x_1,x_2), \end{aligned} \]
\[ F_3=\dot F_3-\frac{1+2\nu(3-4\nu)}{1-2\nu}\dot F_3' -\frac{4(1-\nu)}{1-2\nu}x_3\frac{\partial \dot F_3'}{\partial x_3} +\frac{4\nu}{1-2\nu}x_3(\operatorname{div}\dot{\mathbf F})' +\frac{1}{1-2\nu}x_3^2\frac{\partial}{\partial x_3}(\operatorname{div}\dot{\mathbf F})' +2\nu\int_{x_3}^{+\infty}(\operatorname{div}\dot{\mathbf F})'\,dx_3 . \]
Here \(\nu\) is Poisson’s ratio, \(\delta_{i3}\) is the Kronecker symbol, and the prime operation denotes replacing \(x_3\) by \((-x_3)\). We note that the integrals may turn out to be divergent in the ordinary sense; therefore they should be understood in the sense of Hadamard or, equivalently, in the sense of the theory of generalized functions \((^6)\).
Moscow State University
named after M. V. Lomonosov
Received
6 VI 1970
REFERENCES
\(^1\) A. Lyav, Mathematical Theory of Elasticity, Moscow, 1935.
\(^2\) R. D. Mindlin, Physics, 7, 195 (1936); Collected Translations: Mechanics, 4, 118 (1952).
\(^3\) R. J. Duffin, J. Rat. Mech. and Analysis, 5, No. 5, 939 (1956).
\(^4\) E. I. Obolashvili, Rev. Roumaine Math. pures et appl., 11, No. 8, 965 (1966).
\(^5\) V. D. Kupradze, T. G. Gegelia et al., Three-Dimensional Problems of the Mathematical Theory of Elasticity, Tbilisi, 1968.
\(^6\) Ch. Solver, in: Problems of Mechanics, 3, Moscow, 1961, p. 7.