Inverse heat conduction problems in three-dimensional anisotropic functionally graded solids (Q695561)

This paper presents a local equation formulation in the time domain using a meshless method based on the local Petrov-Galerkin approach to solving inverse transient three-dimensional heat conduction problems in anisotropic, continuously inhomogeneous solids. This method can be employed to determine the unspecified heat transfer coefficients. The Heaviside step function is used as the test function in the local weak form, leading to the derivation of local boundary-domain integral equations. In contrast to conventional boundary integral equation methods, all the integrands in the present formulation are regular. Thus, no special integration techniques are required to evaluate the integrals. Nodal points are randomly distributed in the domain analyzed, and each node is surrounded by a spherical subdomain in which a local integral equation is applied. The moving least-squares (MLS) approximation is used in the considered analysis. After performing spatial integrations, a system of ordinary differential equations for certain nodal unknowns is obtained. A backward finite-difference method is used for the approximation of the diffusive term in the heat conduction equation. A truncated singular-value decomposition is applied to solve the ill-conditioned linear system of algebraic equations at each time step. The numerical results indicate that the present method is accurate and stable with respect to perturbations in the input data. Simple geometry and boundary conditions are considered for many examples to enable a comparison with numerical results. The main advantage of the present method is its simplicity and general applicability. The proposed method can be further extended to nonlinear problems, where meshless approximations may have certain advantages over conventional domain-type discretization approaches. A similar conclusion can be reached for the comparison of the proposed method with the conventional boundary element method (BEM). The method is particularly promising for problems which cannot be solved by the conventional BEM when fundamental solutions are not available. However, in its current implementation, the computation time of the proposed method is large since the number of nodes is huge and the shape functions in the MLS approximation are significantly more complex than in BEM or the finite element method which uses simple polynomials. Intensive research to reduce the CPU requirements is currently in progress.