APPROXIMATE SOLUTION OF PROBLEMS IN THE THEORY OF SMALL ELASTO-PLASTIC DEFORMATIONS
Unknown
Submitted 1957-01-01 | SovietRxiv: ru-195701.86681 | Translated from Russian

Abstract Generated abstract

The paper develops an approximate method for plane problems in the theory of small elastoplastic deformations by expanding stresses, displacements, and strains in a parameter about an axisymmetric zeroth-order state, with incompressibility and a power-law relation between stress and strain intensities. For each approximation order, the governing relations are reduced through a displacement potential and angular separation to a fourth-order radial differential equation with a known right-hand side determined by lower-order terms. The homogeneous equation is analyzed by deriving and factorizing its characteristic equation, including the special case of the first angular harmonic, which gives explicit roots. The resulting formulation is proposed for applications such as eccentric or elliptical tubes under pressure and thick plates with circular or elliptical holes, and is noted to extend analogously to symmetric torsion.

Full Text

THEORY OF ELASTICITY

D. D. IVLEV

APPROXIMATE SOLUTION OF PROBLEMS IN THE THEORY OF SMALL ELASTO-PLASTIC DEFORMATIONS

(Presented by Academician A. I. Nekrasov, 12 X 1956)

We shall seek the solution of the problem in the form of series in a certain parameter \(\delta\):

\[ \sigma_\rho=\sum_{n=0}\delta^n\sigma_\rho^{(n)},\ldots,\quad u=\sum_{n=0}\delta^n u^{(n)},\ldots,\quad e_\rho=\sum_{n=0}\delta^n e_\rho^{(n)},\ldots \tag{1} \]

If for \(\delta=0\) an axisymmetric stressed state occurs, then, neglecting compressibility, in the case of plane deformation we shall have

\[ u^0=-\frac{c}{\rho},\quad e_\rho^0=-e_\theta^0=\frac{c}{\rho^2},\quad e_i^0=\frac{2}{\sqrt{3}}\frac{c}{\rho^2},\quad c=\mathrm{const}. \tag{2} \]

Substituting the expansions (1) into the relations connecting stresses and deformations in the theory of small elasto-plastic deformations \((1)\):

\[ \sigma_\rho-\sigma_\theta=\frac{4}{3}\frac{\sigma_i}{e_i}e_\rho,\quad \tau_{\rho\theta}=\frac{1}{3}\frac{\sigma_i}{e_i}e_{\rho\theta}, \]

and assuming that \(\sigma_i=Ae_i^\chi\), using (2), we obtain:

\[ \sigma_\rho^{(n)}-\sigma_\theta^{(n)}=4B\chi\rho^p e_\rho^{(n)}+F_n,\quad \tau_{\rho\theta}^{(n)}=B\rho^p e_{\rho\theta}^{(n)}+\Phi_n, \tag{3} \]

where

\[ B=\frac{A}{3}\left(\frac{2c}{\sqrt{3}}\right)^{\chi-1},\quad p=2(1-\chi), \]

and the functions \(F_n\) and \(\Phi_n\) depend on components not higher than the \((n-1)\)-st approximation.

Assuming that the \((n-1)\)-st approximation has been determined, we determine the \(n\)-th approximation. Put

\[ u^{(n)}=-\frac{1}{\rho}\frac{\partial \Psi_n}{\partial\theta},\quad v^{(n)}=\frac{\partial \Psi_n}{\partial\rho},\quad \Psi_n=\rho R(\rho)\Theta(\theta). \tag{4} \]

From (3) and (4) we have

\[ \sigma_\rho^{(n)}-\sigma_\theta^{(n)}=-4B\chi\rho^p R'\dot{\Theta}+F_n, \]

\[ \tau_{\rho\theta}^{(n)} = B\rho^p \left[ \left(\rho R''+R'-\frac{R}{\rho}\right)\Theta -\frac{R}{\rho}\ddot{\Theta} \right] +\Phi_n, \tag{5} \]

where a prime above denotes differentiation with respect to \(\rho\), and a dot denotes differentiation with respect to \(\theta\).

Setting \(\Theta=\cos m\theta\) or \(\sin m\theta\), from (5) and the equilibrium equations we obtain:

\[ \rho^4 R^{\mathrm{IV}}+2(p+3)\rho^3R'''+\bigl[p^2+6p+5+2m^2(1-2\chi)\bigr]\rho^2R'' \]
\[ +\rho R'\bigl[(p^2-1)-2m^2(2\chi-1)(p+1)\bigr] +R\bigl[m^4+(1-p^2)-(2-p^2)m^2\bigr]=U_n, \tag{6} \]

where the right-hand side of equation (6) is a known function of the radius \(\rho\).

We find the general solution of the homogeneous equation (6). Setting \(R=\rho^k\), we obtain from (6):

\[ k^4+2pk^3+\left[p^2-2-2(2\varkappa-1)m^2\right]k^2-2p\left[1+(2\varkappa-1)m^2\right]k+ \]
\[ +\left[(1-p^2)-(2-p^2)m^2+m^4\right]=0. \tag{7} \]

Fig. 1

Fig. 1

Equation (7) admits a factorization:

\[ (k^2+pk+a+ib)(k^2+pk+a-ib)=0, \tag{8} \]

where

\[ a=-\left[1+(2\varkappa-1)m^2\right],\qquad a^2+b^2=(1-m^2)^2+p^2(m^2-1); \]

\[ b^2=4\left\{m^4\varkappa(1-\varkappa)+m^2\left[(1-\varkappa)^2-\varkappa\right]-(1-\varkappa)^2\right\}. \]

The factorization of equation (7) into factors (8) may occur under certain relations between \(\varkappa\) and \(m\). In Fig. 1 the graph for \(b=0\) is presented; the factorization (8) takes place in the zone of complex roots of equation (7). The roots of equation (8) are

\[ k_{1,2,3,4} =\frac12\left\{-p \pm \sqrt{\frac12\left[\sqrt{(p^2-4a)^2+16b^2}+(p^2-4a)\right]} \right. \]
\[ \left. \pm i\sqrt{\frac12\left[\sqrt{(p^2-4a)^2+16b^2}-(p^2-4a)\right]} \right\}. \tag{9} \]

For \(m=1\), equation (7) takes the form

\[ k\left[k^3+2pk^2+\left[p^2-2-2(2\varkappa-1)\right]k-2p\left[1+(2\varkappa-1)\right]\right]=0. \tag{10} \]

The roots of equation (10) (see also \({}^{(2)}\)) are

\[ k_1=0,\qquad k_2=-2(1-\varkappa),\qquad k_3=2\varkappa,\qquad k_4=-2. \tag{11} \]

Using the values of the roots (9) and (11), it is easy to write down the general solution of the homogeneous equation (7). Finding a particular solution of equation (7) in concrete problems presents no difficulty. Using the general solution of equation (7), one can obtain solutions of problems on an eccentric tube under the action of external and internal pressures, on an elliptical tube, on biaxial stretching of a thick plate with a circular or elliptical hole, etc.

We note that, analogously, we arrive at equation (7) if for \(\delta=0\) symmetric torsion takes place:

\[ \sigma_\rho^0=\sigma_\theta^0=0,\qquad \tau_{\rho\theta}^0=\frac{\tau_0}{\rho^2},\qquad \rho\geqslant \alpha\ne 0. \]

Moscow State University
named after M. V. Lomonosov

Received
10 X 1956

References Cited

\({}^{1}\) A. A. Ilyushin, Plasticity, 1948. \({}^{2}\) L. M. Kachanov, Izv. AN SSSR, OTN, no. 9 (1956).

Submission history

APPROXIMATE SOLUTION OF PROBLEMS IN THE THEORY OF SMALL ELASTO-PLASTIC DEFORMATIONS