On the theory of elastic stability
- 1 April 1925
- journal article
- Published by The Royal Society in Proceedings of the Royal Society of London. Series A, Containing Papers of a Mathematical and Physical Character
- Vol. 107 (744) , 734-760
- https://doi.org/10.1098/rspa.1925.0053
Abstract
The object of the present paper is to derive equations that are adequate to decide questions of the stability under stress of thin shells of isotropic elastic material. Equations for the same purpose have been given by R. V. Southwell, who used a method that is closely followed in a part of this paper. Such equations must contain terms that may be, and are, neglected in applications of the theory of elasticity to problems in which the stability of configurations is not considered. The truth of Kirchhoff's uniqueness theorem, which has reference to the ordinary equations of elasticity, in which powers of the displacement co-ordinates above the first are neglected, is sufficient proof of this statement. In practice it is generally sufficient to retain only the first and second order terms, and no terms of higher order are considered here. To obtain such equations an extended form of Hooke's Law is necessary; the extension made by Southwell is used in this paper. There are then two methods available for the derivation of the equations. Either we may obtain the three conditions for the equilibrium of an elementary volume of the substance by considering the forces acting upon it, or we may calculate the energy of strain correct to the third order of displacement co-ordinates, and deduce the equations by variation of this function. The first method has been used in one place here, as it would appear to be the simpler in the particular case of a plane plate, in which only one of the equations, and that the simplest, is required. However, the stability equations for a cylindrical shell are also obtained, and then all three equations are necessary. The derivation by the first method of each one of these is a laborious matter, while using the second method there is only one calculation, that of the strain energy function, to be made. Consequently, for this purpose, as in general, the second method seems to be preferable.This publication has 0 references indexed in Scilit: