Modeling and Optimization of Anisotropic Viscoelastic Porous Structures Using CQBEM and Moving Asymptotes Algorithm

Abstract

The aim of this paper is to develop a novel numerical modeling optimization technique and implement it with the boundary element method (BEM) based on convolution quadrature method for studying optimization of anisotropic viscoelastic porous permeable structures. After applying the quadrature rule to the discretized boundary integral equation, a time stepping procedure based on the use of the linear multistep method is obtained. To obtain anisotropic fundamental solutions, the calculation of a double integral should be performed, but it increases the total computation time in the BEM. In order to overcome this problem and improve the computational efficiency of the formulation, our new proposed BEM technique is implemented. The method of implicit differentiation with respect to design variables has been implemented to calculate the displacements and pore pressure design sensitivities, and the effects of viscosity on these sensitivities are discussed. The resulting topology optimization problem has been solved using the method of moving asymptotes algorithm with adjoint variable method to optimize material distribution and find the influence of viscosity on the optimal design. The validity, efficiency and accuracy of the proposed BEM technique were confirmed by comparing obtained results for elliptical sandwich structure and multi-well reservoir with the corresponding results of finite difference method, lattice Boltzmann method and finite element method, which are special cases of our general and complex problem.