Shape sensitivity analysis using low- and high-order finite elements

A general procedure to perform the sensitivity analysis for the shape optimal design of elastic structures is proposed. The method is based on the implicit differentiation of the discretized equilibrium equations used in the finite element method (FEM). The so-called semianalytical approach is followed, that is, finite differences are used to differentiate the finite element matrices. The technique takes advantage of the geometric modeling concepts typical of the computer-aided design (CAD) technology used in the creation of a compact design model. This procedure is largely independent of the types of finite elements used in the analysis and has been implemented in ah-version andp-version finite element program. Very accurate and stable shape sensitivity derivatives were obtained from both programs over a wide range of finite difference step sizes. It is shown that the method is computationally efficient, general, and relatively easy to implement. Some classical shape optimal design problems have been solved using the CONLIN optimizer supplied with these gradients.     

Shape sensitivity analysis with design-dependent loadings—equivalence between continuum and discrete derivatives

The purpose of this paper is twofold: (1) showing equivalence between continuum and discrete formulations in sensitivity analysis when a linear velocity field is used and (2) presenting shape sensitivity formulations for design-dependent loadings. The equations for structural analysis are often composed of the stiffness part and the applied loading part. The shape sensitivity formulations for the stiffness part were well-developed in the literature, but not for the loading part, especially for body forces and surface tractions. The applied loads are often assumed to be conservative or design-independent. In shape design problems, however, the applied loads are often functions of design variables. In this paper, shape sensitivity formulations are presented when the body forces and surface tractions depend on shape design variables. Especially, the continuum–discrete (C–D) and discrete–discrete (D–D) approaches are compared in detail. It is shown that the two methods are theoretically and numerically equivalent when the same discretization, numerical integration, and linear design velocity fields are used. The accuracy of sensitivity calculation is demonstrated using a cantilevered beam under uniform pressure and an arch dam crown cantilever under gravity and hydrostatic loading at the upstream face of the structure. It is shown that the sensitivity results are consistent with finite difference results, but different from the analytical sensitivity due to discretization and approximation errors of numerical analysis.     

Computationally efficient, numerically exact design space derivatives via the complex Taylor's series expansion method

Within numerical design optimization, discrete sensitivity analysis is often used to estimate the derivative of an objective function with respect to the design variables. Discrete sensitivity analysis estimates these derivatives by taking advantage of additional derivative information available in an implicit computational fluid dynamics (CFD) solver of the discretized governing partial differential equations. The key benefits of steady-state discrete sensitivity analysis are its computational efficiency and numerical accuracy. More recently, the complex Taylor's series expansion (CTSE) method has been used to generate these design space derivatives to machine accuracy, by analyzing a complex perturbation of the objective function. For fortran codes, this method is quite easy to implement, for both implicit and explicit codes; unfortunately, the CTSE method can be quite time consuming, because it requires a complex solution of the governing partial differential equations. In this paper, the authors demonstrate that the direct formulation of discrete sensitivity analysis and the CTSE method solve the same iterative sensitivity equation, which sheds light on the most efficient use of the CTSE method. Finally, these methods are demonstrated via application to numerical simulations of one-dimensional and two-dimensional open-channel flows.     

A finite element method for thermoviscoelastic analysis of plane problems

A stress analysis for plane problems in linear thermoviscoelasticity using a finite element formulation is presented. The method employed is based on the assumptions that (1) the material is isotropic, homogeneous and linear, (2) the stress-strain laws are expressed in the hereditary integral form, and (3) the material is thermorheologically simple, which implies that the temperature-time equivalence hypothesis is valid. The associated computer program utilizes isoparametric plane elements.The element matrices that result in the equilibrium equations involve hereditary integrals, and these are approximated by a finite difference scheme for time marching. The solutions for two problems are compared with analytical results evaluated by the integral transform method.For approximate results which require less computer time an alternative form of equilibrium equations utilizing an iterative technique is presented and an example solution is included. Finally, the effect of incompressibility is considered for an axisymmetric numerical example.     

Optimization of airplane wing structures under gust loads

A methodology is presented for the optimum design of aircraft wing structures subjected to gust loads. The equations of motion, in the form of coupled integro-differential equations, are solved numerically and the stresses in the aircraft wing structure are found for a discrete gust encounter. The gust is assumed to be one minus cosine type and uniform along the span of the wing. In order to find the behavior of the wing structure under gust loads and also to obtain a physical insight into the nature of the optimum solution, the design of the typical section (symmetric double wedge airfoil) is studied by using a graphical procedure. Then a more realistic wing optimization problem is formulated as a constrained nonlinear programming problem based on finite element modeling and the optimum solution is found by using the interior penalty function method. A sensitivity analysis is conducted to find the effects of changes in design variables about the optimum point on the response quantities of the wing structure.     

Continuum shape sensitivity analysis for aeroelastic gust using an arbitrary lagrangian-eulerian reference frame

Continuum Sensitivity Analysis (CSA), a method to determine response derivatives with respect to design variables, is derived here for the first time in an arbitrary Lagrangian-Eulerian (ALE) reference frame. CSA differentiates nonlinear governing system of equations to arrive at a linear system of partial differential continuum sensitivity equations (CSEs), here, for fluid-structure interaction (FSI). The CSEs and associated sensitivity boundary conditions are derived here for the first time for FSI, using the boundary velocity formulation, carefully distinguishing design velocity from flow velocity and ALE mesh velocity. Whereas boundary conditions must be differentiated using the material (total) derivative, it is sometimes advantageous to derive the CSEs using local (partial) derivatives. The benefit is that geometric sensitivity, known as design velocity, may not be required in the domain. It is shown here that this advantage is realized when the ALE frame undergoes only the rigid body motion associated with the structure to which it is attached. It is further shown that the advantage is not realized when the ALE mesh deforms due to the flexible motion of the fluid-structure interface. The equations for the transient gust response of a two-dimensional airfoil in compressible flow, flexibly attached to a rigid body mass, are presented as a model problem to illustrate a detailed derivation.     

Shape design sensitivity analysis of nonlinear structural systems

A unified approach is presented for shape design sensitivity analysis of nonlinear structural systems that include trusses and beams. Both geometric and material nonlinearities are considered. Design variables that specify the shape of components of built-up structures are treated, using the continuum equilibrium equations and the material derivative concept. To best utilize the basic character of the finite element method, shape design sensitivity information is expressed as domain integrals. For numerical evaluation of shape design sensitivity expressions, two alternative methods are presented: the adjoint variable and direct differentiation methods. Advantages and disadvantages of each method are discussed. Using the domain formulation of shape design sensitivity analysis, and the adjoint variable and direct differentiation methods, design sensitivity expressions are derived in the continuous setting in terms of shape design variations. A numerical method to implement the shape design sensitivity analysis, using established finite element codes, is discussed. Unlike conventional methods, the current approach does not require differentiation of finite element stiffness and mass matrices.     

基于混合反射模型的SFS有限元方法的研究

提出了基于混合反射模型的由明暗恢复物体三维形状的有限元算法。用正方形面元逼近光滑曲面,把曲面表示为所有节点基函数的线性组合;基于既含有漫反射成分又有镜面反射成分的混合模型,结合节点基函数,将反射图线性化。考虑数字图像的特点,直接使用离散形式的SFS问题的亮度约束形式,用最小化方法得到高度满足的线性方程;使用Kaczmarz算法计算出表面三维形状。使用合成图像和实际图像验证该文算法的有效性,探讨了该算法的性能。     

An immersed boundary approach for shape and topology optimization of stationary fluid-structure interaction problems

This paper presents an approach to shape and topology optimization of fluid-structure interaction (FSI) problems at steady state. The overall approach builds on an immersed boundary method that couples a Lagrangian formulation of the structure to an Eulerian fluid model, discretized on a deforming mesh. The geometry of the fluid-structure boundary is manipulated by varying the nodal parameters of a discretized level set field. This approach allows for topological changes of the fluid-structure interface, but free-floating volumes of solid material can emerge in the course of the optimization process. The free-floating volumes are tracked and modeled as fluid in the FSI analysis. To sense the isolated solid volumes, an indicator field described by linear, isotropic diffusion is computed prior to analyzing the FSI response of a design. The fluid is modeled with the incompressible Navier-Stokes equations, and the structure is assumed linear elastic. The FSI model is discretized by an extended finite element method, and the fluid-structure coupling conditions are enforced weakly. The resulting nonlinear system of equations is solved monolithically with Newton’s method. The design sensitivities are computed by the adjoint method and the optimization problem is solved by a gradient-based algorithm. The characteristics of this optimization framework are studied with two-dimensional problems at steady state. Numerical results indicate that the proposed treatment of free-floating volumes introduces a discontinuity in the design evolution, yet the method is still successful in converging to meaningful designs.     

A fully Lagrangian mixed discrete least squares meshfree method for simulating the free surface flow problems

This paper presents a fully Lagrangian mixed discrete least squares meshfree (MDLSM) method for simulating the free surface problems. In the proposed method, the mass and momentum conservation equations are first discretized in time using the projection method. The resulting pressure Poisson equation is then re-written in the form of three first-order equations in terms of the pressure field and its first-order derivatives. The mixed discrete least squares meshless method is then used to solve this system of equations and simultaneously calculate the pressure field and its gradients. The advantage of the proposed Lagrangian MDLSM is twofold. First, the pressure gradients are directly computed and, therefore, they enjoy higher accuracy than those calculated in the conventional DLSM via a post-processing method. The more accurate pressure gradient will in turn lead to more accurate velocity field when used in the momentum equations. Second, the proposed Lagrangian MDLSM method can be more efficient than the corresponding original Lagrangian DLSM method, for the specific number of nodes, since the costly calculation of the shape function second derivatives required for solving the pressure Poison equation are avoided in each time step of the simulation. Several free surface problems are solved and the results are presented and compared to those of DLSM method. The results indicate the superior efficiency and accuracy of the proposed Lagrangian MDLSM method compared to those of the existing Lagrangian DLSM method in the literature.

     

A modification to the sensitivity analysis algorithm for known target displacements

The purpose of this note is to show that a secant approximation of the displacement state can be obtained by a simple modification of the sensitivity analysis algorithm for static displacements and frequency responses, if the displacement state of the target design is known. When applied to the discretized equilibrium equations obtained by the finite element method, this new approximation contains the target design exactly, if the system matrices are linear functions of the design variables and all components of the displacement vector are known at the target design. Therefore, a single application of a numerical search process to this approximation will reach the target design.The displacement state at the optimal design is not known in advance in design optimization. However, for special applications, such as model updates or the creation of simplified models from very large models, this new approximation could be useful in augmenting the straightforward application of the conventional structural optimization techniques.     

A Fourier analysis of spurious mechanisms and locking in the finite element method

A Fourier analysis technique is used to examine spurious mechanisms and the element locking phenomena engendered by different finite element discretizations. It is shown that these phenomena can be identified by examining the uncoupled discrete Fourier operators and corresponding characteristic equations and are caused, in terms of discrete filter concepts, by either a spurious mode carrier or by violating the unlocking condition specified in the paper. The analysis is performed for two simple problems: wave propagation in a bar and vibration of a Timoshenko beam which, when discretized by linear shape functions, succinctly manifest the two phenomena as being directly traceable to specific components of the finite element discretizations.     

Level set topology optimization considering damage

This paper provides a level set based topology optimization approach to design structures exhibiting resistance to damage. The geometry of the structures is represented by the level set method. The design domains are discretized by the extended finite element method allowing for fixed non conforming meshes. The mechanical model represents quasi-brittle materials. Undamaged material behavior is assumed linear elastic while a loss of stiffness is introduced through a non-local damage model. Small strains are assumed. The sensitivities are evaluated by an analytical derivation of the discretized governing equations of the system and considering the adjoint approach. As the damage process is irreversible, the structural responses are path-dependent and this dependency is accounted for in the sensitivity analysis. The optimization problems are solved by mathematical programming algorithms, in particular using the GCMMA scheme. The proposed approach is illustrated with two dimensional examples that highlight the influence of degradation on the optimized designs.     

A modal approach for the evaluation of the response sensitivity of structural systems subjected to non-stationary random processes

A method for the evaluation of the response sensitivity of both classically and non-classically damped discrete linear structural systems under stochastic actions is presented. The proposed approach requires the following items: (a) a suitable modal expansion of the response; (b) the derivation in analytical form of the equations governing the evolution of the derivatives of the response (the so-called sensitivity equations) with respect to the parameters that define the structural model; (c) an extensive use of the Kronecker algebra for determining the analytical expressions of the sensitivity of the structural response statistics to non-stationary random input processes. Moreover, a step-by-step integration scheme able to solve the sensitivity equations is also studied. Handy expressions for the cross-correlations between the input process and the response sensitivities are also derived. A numerical application shows that the proposed approach is suitable to cope with practical problems of engineering interest.     

Sensitivity analysis of structural response to position of external applied load: in plane stress condition

Procedures for sensitivity analysis of the structural responses, i.e., nodal displacement, mean compliance and local stresses within an element, with respect to the location of an external applied load are developed. This is mainly because the external loads are often of some freedom to change their application positions in the structural preliminary design. Apart from the structural response evaluation, the finite element method is employed in this work for the sensitivity analysis implementation of a plane stress continuum structure. First, an external load is transformed into the equivalent nodal forces such that the influence of an external load shift is represented completely by the magnitude variation of the associated nodal forces, upon which the first- and second-order derivatives of an external load to its location change are performed respectively in a closed form by the aid of the features of the element shape functions. Then, the relevant sensitivities of the structural responses aforementioned are formulated readily with the discrete method. Finally, two typical examples are provided to demonstrate the validity of the sensitivity formulations presented, and the numerical results show a perfect accuracy of calculation of the response sensitivity.     

On the enhancement of computation and exploration of discretization approaches for meshless shape design sensitivity analysis

A new implementation of Reproducing Kernel Particle Method (RKPM) is proposed to enhance the process of shape design sensitivity analysis (DSA). The acceleration process is accomplished by expressing RKPM shape functions and their derivatives explicitly in terms of kernel function moments. In addition, two different discretization approaches are explored elaborately, which emanate from discretizing design sensitivity equation using the direct differentiation method. Comparison of these two approaches is made, and the equivalence of these two superficially different approaches is demonstrated through two elastostatics problems. The effectiveness of the enhanced RKPM is also verified by comparison of consumption of computer time between the classical method and the improved method.     

Aerodynamic design optimization on unstructured grids with a continuous adjoint formulation

A continuous adjoint approach for obtaining sensitivity derivatives on unstructured grids is developed and analyzed. The derivation of the costate equations is presented, and a second-order accurate discretization method is described. The relationship between the continuous formulation and a discrete formulation is explored for inviscid, as well as for viscous flow. Several limitations in a strict adherence to the continuous approach are uncovered, and an approach that circumvents these difficulties is presented. The issue of grid sensitivities, which do not arise naturally in the continuous formulation, is investigated and is observed to be of importance when dealing with geometric singularities. A method is described for modifying inviscid and viscous meshes during the design cycle to accommodate changes in the surface shape. The accuracy of the sensitivity derivatives is established by comparing with finite-difference gradients and several design examples are presented.     

On energy conservation for finite element approximation of flow-induced airfoil vibrations

In this paper the numerical approximation of a two-dimensional fluid–structure interaction problem is addressed. The fully coupled formulation of incompressible viscous fluid flow interacting with a flexibly supported airfoil is considered. The flow is described by the incompressible system of Navier–Stokes equations, where large values of the Reynolds number are considered. The Navier–Stokes equations are spatially discretized by the finite element method and stabilized with a modification of the Galerkin Least Squares (GLS) method; cf. [T. Gelhard, G. Lube, M.A. Olshanskii, J.-H. Starcke, Stabilized finite element schemes with LBB-stable elements for incompressible flows, Journal of Computational and Applied Mathematics 177 (2005) 243–267]. The motion of the computational domain is treated with the aid of Arbitrary Lagrangian Eulerian (ALE) method and the stabilizing terms are modified in a consistent way with the ALE formulation.     

Large deflection analysis of plates and shells by discrete energy method

The discrete energy method—a special form of finite difference energy approach—is presented as a suitable alternative to the finite element method for the large deflection elastic analysis of plates and shallow shells of constant thickness. Strain displacement relations are derived for the calculation of various linear and nonlinear element stiffness matrices for two types of elements into which the structure is discretized for considering separately energy due to extension and bending and energy due to shear and twisting. Large deflection analyses of plates with various edge and loading conditions and of a shallow cylindrical shell are carried out using the proposed method and the results compared with finite element solutions. The computational efforts required are also indicated.