PARALLEL COMPUTING. Tayfun E. Tezduyar. Ahmed Sameh

Size: px
Start display at page:

Download "PARALLEL COMPUTING. Tayfun E. Tezduyar. Ahmed Sameh"

Transcription

1 1 FINITE ELEMENT METHODS: 1970 s AND BEYOND L.P. Franca, T.E. Tezduyar and A. Masud (Eds.) c CIMNE, Barcelona, Spain 2004 PARALLEL COMPUTING Tayfun E. Tezduyar Mechanical Engineering Rice University MS Main St Houston, TX 77005, USA tezduyar@rice.edu Ahmed Sameh Computer Science Purdue University West Lafayette, IN 47907, USA sameh@cs.purdue.edu Abstract. We provide an overview of the role of parallel computing in finite element computations, with emphasis on computational fluid mechanics and moving boundaries and interfaces. The class of problems with moving boundaries and interfaces, which includes fluid object and fluid structure interactions as well as free-surface and two-fluid flows, offers a number of computational challenges. These challenges need to be addressed in such a way that 3D computation of complex applications can be carried out efficiently on parallel computers. This requirement becomes an important factor in designing various components of the overall solution strategy, such as solution techniques for the discretized equations and mesh update methods for handling the changes in the spatial domain occupied by the fluid. Our overview includes description of how these types of challenges are addressed. Key words: Parallel computing, mesh update, preconditioned iterative algorithms. 1 INTRODUCTION Within the past decade parallel computing has had a major and indispensable role in 3D finite element computation of complex engineering problems. Together with advanced algorithms capable of efficiently simulating complex, real-world problems, advanced parallel computing platforms have brought simulation and modeling to a new level. Also within the past decade, we witnessed many changes in parallel computing platforms from Thinking Machines CM 5 to CRAY T3E, from IBM SP to PC-clusters. Currently the future of supercomputing in the US is being assessed, and there are calls for revitalizing high performance computing in the country. We provide additional comments on this subject in Section 5. Parallel computations can be carried out on shared or distributed memory systems, or a combination of the two. Parallel implementations on shared memory systems, which are relatively simpler, are based on the processors using the same

2 Finite Element Methods: 1970 s and beyond central memory. Parallel implementations on distributed memory systems require more elaborate work, but within the past decade many interfaces have been developed to make the task easier for computational engineers. For example, the user-friendly Connection Machine software libraries developed by Johnsson and his associates 1 3 at Thinking Machines encouraged several finite element researchers to implement their formulations on the Connection Machine systems. Some of the earlier and landmark contributions in parallel finite element computations were made by Hughes and his coworkers. 4 6 Examples of other earlier contributions in finite element computations are those by Tezduyar, 7 9 Farhat, Shephard, 13 and their associates. It is not our intention here to provide an exhaustive review of the parallel finite element computation activities of the past decade, but to give examples of how we have addressed some of the computational challenges in 3D finite element computation of complex engineering problems on parallel computing platforms. We focus on flow problems with moving boundaries and interfaces, such as fluid object and fluid structure interactions and free-surface and two-fluid flows. Specifically we describe how the mesh is updated with sufficient parallel efficiency as the spatial domain occupied by the fluid changes its shape during the computation of this class of problems. We also focus on how we can employ effective iterative schemes for the solution of large, coupled equation systems generated by the spatial and temporal discretizations. In particular, we present advanced, special preconditioners that are ideally suited for implementation on high-end parallel architectures. In the computation of moving boundaries and interfaces, flow conditions determine whether we should use an interface-tracking or interface-capturing technique or a combination of the two. In an interface-tracking technique the interface computation requires meshes that track the interfaces. The mesh needs to be updated as the interface evolves. In an interface-capturing technique the interface computation does not require that the mesh be updated as the interface evolves. Instead, the interface is captured within the resolution of the finite element mesh covering the area where the interface is. The interface-tracking and interface-capturing techniques we have developed over the years (see ) are based on stabilized formulations. The stabilized methods are the streamline-upwind/petrov-galerkin (SUPG) and pressure-stabilizing/petrov-galerkin (PSPG) 14,23 formulations. An earlier version of the pressure-stabilizing formulation for Stokes flows was reported in. 24 These stabilized formulations prevent numerical instabilities in solving problems with high Reynolds or Mach numbers and shocks or thin boundary layers, as well as when using equal-order interpolation functions for velocity and pressure. These stabilized formulations also substantially improve the convergence rate in iterative solution of the large, coupled nonlinear equation system that needs to be solved at every time step of a flow computation. Such nonlinear systems are typically solved with the Newton Raphson method, which involves, in each iteration step, solution of a large linear system of equations. It was pointed out in 8 that using a good stabilized method makes substantial difference in the speed of convergence of iterative schemes for solving those inner linear systems. Preconditioned iterative schemes that perform well even in the absence of such stabilizations have also been developed. 25,26 We will discuss those briefly in later sections. In our interface-tracking techniques the core method is the Deforming-Spatial-

3 T.E. Tezduyar and A. Sameh / PARALLEL COMPUTING Domain/Stabilized Space Time (DSD/SST) formulation, where the finite element formulation of the problem is written over its space time domain. As the spatial domain occupied by the fluid changes its shape in time, the mesh is updated with an automatic mesh moving method, 7,17 where the motion of the nodes is governed by the equations of elasticity, and full or partial remeshing (i.e., generating a new set of elements, and sometimes also a new set of nodes) as needed. The mesh update issues, including their relevance to parallel efficiency, is discussed in more detail in Section 2. We also note that the stabilized space time formulations were used earlier by other researchers to solve problems with fixed spatial domains (see for example 27 ). In our interface-capturing techniques (see 17,18 ) the core method consists of the stabilized formulations of the Navier Stokes equations and an advection equation governing the time-evolution of an interface function marking the interface location. For comparable levels of spatial discretization, interface-capturing methods yield less accurate representation of the interface. By accurate representation of the interface we mean not only accurately locating the interface, but also accurately resolving the boundary or interface layers that might be present near the interface. However, because the interface-capturing techniques do not require the mesh to be updated as the interface evolves, they can be used as practical alternatives when a free-surface or two-fluid interface might be too complex or unsteady to track while keeping the frequency of remeshing at an acceptable level. The accuracy of an interface-capturing technique in representing an interface can be increased in a number of ways, one of which is the Enhanced-Discretization Interface-Capturing Technique (EDICT). 17,18,28 In the EDICT, a second-level, more refined mesh is built over regions of the first-level mesh which contain the interface. The finite element function spaces are based on a combination of first- and second-level meshes. How the second-level mesh is constructed and updated, and how the computational and parallel efficiency are maintained, are discussed in more detail in Section 2. The large coupled, nonlinear equation systems that need to be solved in each time step in a time-accurate computation (or every pseudo-time step in a steady-state computation) are solved using iterative methods. 7,17,29 34 The main components of these iterative methods are: a) formation of the residual vector of the linear equation system (that needs to be solved in each iteration of a Newton Raphson sequence used in solving the coupled, nonlinear equations); b) designing effective preconditioners for the Jacobians in the Newton Raphson iterations; and c) updating the solution vector in an optimal way. More details on the iterative solution techniques are discussed in Sections 3 and 4. In Section 5 we provide brief comments on parallel computing platforms. We present some numerical examples in Section 6, and end with concluding remarks in Section 7. 2 MESH UPDATE METHODS In interface-tracking techniques, how the mesh is updated depends on several factors, such as the complexity of the interface and overall geometry, how unsteady the interface is, and how the starting mesh was generated. In general, the mesh update could have two components: moving the mesh for as long as it is possible, and full or partial remeshing (i.e., generating a new set of elements, and sometimes also a new set of nodes) when the element distortion becomes too high. In mesh moving strategies, the only rule the mesh motion needs to follow is that at the interface the

4 Finite Element Methods: 1970 s and beyond normal velocity of the mesh has to match the normal velocity of the fluid. Beyond that, the mesh can be moved in any way desired, with the main objective being to reduce the frequency of remeshing. In most 3D applications, remeshing requires calling an automatic, unstructured-mesh generator, and therefore reducing the cost of automatic mesh generation becomes a major incentive for reducing the frequency of remeshing. Since the parallel efficiency of most automatic mesh generators is substantially lower than that of most flow solvers, one has to strive to reduce the frequency of remeshing. For example, remeshing every ten time steps would sufficiently reduce the influence of remeshing in terms of its added cost and lack of parallel efficiency. In most of the complex flow problems we computed in the past, the frequency of remeshing was far less. In our current parallel computations on a PC-cluster, typically we perform remeshing on one of the nodes, which, with its 2 GigaBytes of memory, is powerful enough to generate very large meshes. If remeshing does not consist of (full or partial) regeneration of just the element connectivities but also involves (full or partial) node regeneration, we need to project the solution from the old mesh to the new one. This involves a search process, which can be carried out in parallel. Still, the computational cost involved in this search process as well as the projection errors introduced by remeshing, add more incentives for reducing the frequency of remeshing. In the automatic mesh moving technique introduced in, 7 the motion of the internal nodes is determined by solving the equations of elasticity. As boundary condition, the motion of the nodes at the interfaces is specified to match the normal velocity of the fluid at the interface. Similar mesh moving techniques were used earlier by other researchers (see for example 35 ). In 7 the mesh deformation is dealt with selectively based on the sizes of the elements, and based on the deformation modes in terms of shape and volume changes. Mesh moving techniques with comparable features were later introduced in. 36 In the technique introduced in, 7 selective treatment of the mesh deformation based on shape and volume changes is attained by adjusting the relative values of the Lamé constants of the elasticity equations. The objective would be to stiffen the mesh against shape changes more than we stiffen it against volume changes. Selective treatment based on element sizes, on the other hand, is attained by altering the way we account for the Jacobian of the transformation from the element domain to the physical domain. In this case, the objective is to stiffen the smaller elements, which are typically placed near solid surfaces, more than the larger ones. We developed a number of additional mesh moving features around this core method for the purpose of reducing the frequency of remeshing. We refer the interested readers to. 8,18,37,38 In the EDICT, a subset of the elements in the first-level mesh are identified as those at or near the interface. The second-level mesh is constructed by patching together the more refined meshes generated over each element in this subset. The interpolation functions for velocity, pressure and interface function will all have two components each: one from the first-level mesh and the other from the second-level mesh. We re-define the subset of the first-level mesh over which we build the secondlevel mesh not every time step but with sufficient frequency to keep the interface enveloped. We need to avoid this envelope being too wide or too narrow. If the envelope is too wide, although we would keep the interface enveloped in for a large number of time steps without reconstructing the second-level mesh, we would also

5 T.E. Tezduyar and A. Sameh / PARALLEL COMPUTING have at a given time enhanced discretization mostly where it is not needed. This would pointlessly increase the computational cost for the flow solver. On the other hand, if the envelope is too narrow, then the frequency of reconstructing the secondlevel mesh would increase. This would in turn increase the computational overhead involved in redefining the location of the second-level mesh and generating it. This would also adversely affect the parallel efficiency of the computations. Our target would be to maintain the frequency of reconstructing the second-level mesh at about every ten time steps. This would keep the computational cost for the flow solver at a reasonable level and at the same time would sufficiently reduce the effect of mesh reconstruction in terms of its added cost and harm to parallel efficiency. 3 ITERATIVE SOLUTION METHODS The finite element formulations mentioned in the earlier sections fall into two categories: a space time formulation with moving meshes or a semi-discrete formulation with non-moving meshes. Full discretizations of these formulations lead to coupled, nonlinear equation systems that need to be solved in each time step of the simulation. Whether we are using a space time formulation or a semi-discrete formulation, we can represent the system of equations that needs to be solved as follows: N (d n+1 ) = F. (1) Here d n+1 is the vector of nodal unknowns. In a semi-discrete formulation, this vector contains the unknowns associated with marching from time level n to n + 1. In a space time formulation, it contains the unknowns associated with the finite element formulation written for the space time slab between time levels n and n+1 (see ). We solve Eq. (1) with the Newton Raphson method: N d ( ) ( ) d i n+1 = F N d i n+1, (2) d i n+1 where i is the step counter for the Newton Raphson sequence, and d i n+1 is the increment computed for d i n+1. The large sparse linear system in Eq. (2), which needs to be solved in each Newton Raphson iteration, may be written as Ax = b. (3) In the class of computations we typically carry out, this linear system is too large to be solved via a direct sparse solver. Therefore, we resort to solving it via a preconditioned iterative scheme. In each iteration, the most time-consuming steps are: (a) computing the residual of the linear system at hand, and (b) solving a linear system of the form r = b Ax, (4) Pz = r, (5) where P is the chosen preconditioner for the linear system in Eq. (3).

6 Finite Element Methods: 1970 s and beyond Computing the residual can be achieved in different ways. The computation can be based on: (i) a sparse-matrix storage of A, (ii) storing just element-level matrices (element-matrix-based), or (iii) storing just the element-level vectors (elementvector-based). This last strategy is also called the matrix-free technique. The preconditioner P is chosen such that it is much simpler to solve systems of the form in Eq. (5) than systems in Eq. (3). Moreover, it should be noted that the closer the inverse of P is to the inverse of A, the fewer the required iterations (to achieve an acceptable solution in each Newton Raphson iteration) are and the more effective the preconditioner is. Updating the solution vector x is also an important step in the iterative solver. Several update methods are available, and the GMRES 39 scheme is among the more popular ones. In constructing iterative flow solvers, we concentrated on choosing good preconditioners P and computing effectively the residual r. We adopted computational techniques that assure efficient implementation on parallel computing platforms. For example, the parallel-ready methods we use for the residual computations include those that are element-matrix-based, 40 elementvector-based, 40 and sparse-matrix-based. 32 The element-vector-based methods were successfully used also by other researchers in the context of parallel computations (see for example 6,41 ). The element-vector-based computations can be based on numerical or analytical definitions of the directional derivatives involved (see 17,18 ). Provided that they can be used without a major difficulty, we prefer the analytical definitions. A mixed analytical/numerical element-vector-based (AEVB/NEVB) computation technique was introduced in 17,18 for fluid structure interaction problems in which analytical element-vector-based computations are exceedingly difficult for the coupling matrix blocks. We have developed some advanced preconditioners such as the Clustered-Elementby-Element (CEBE) preconditioner 30 and the mixed CEBE and Cluster Companion (CC) preconditioner. 30 We have also successfully implemented the CEBE preconditioner in conjunction with an ILU approximation. 32 However, our typical parallel computations are based on diagonal and nodal-block-diagonal preconditioners. These are very simple preconditioners, but are also very simple to implement on parallel platforms. More on our parallel implementations can be found in SPECIAL PRECONDITIONING TECHNIQUES Numerical simulation of fluid particle interactions were extensively investigated by Tezduyar and his associates 14 16,31,33,42 45 with the methods described in earlier sections. In this section we only illustrate an advanced preconditioning method for the linear systems that might arise in such simulations. We describe this method in the context of an Arbitrary Lagrangian Eulerian (ALE) technique, instead of the DSD/SST formulation. A structurally symmetric matrix representation of the resulting Jacobians has been proposed in. 46 These nonsingular linear systems are of the form A B E 1 B T C E 2 F 1 F 2 G u p U = f 1 f 2 f 3, (6) where the first 2 2 block is nonsingular, in which A and C are nonsymmetric and nonsingular of order n and m < n, respectively, and B is of maximal column rank m. The matrix G is nonsingular of order 3q for 2D problems, where q is the number

7 T.E. Tezduyar and A. Sameh / PARALLEL COMPUTING particles. Here, u denotes the fluid velocity at the interior finite element nodes, p denotes the pressure, and U represents the particle velocities and rotations. If we do not use a stabilized formulation then the matrix C vanishes. In this case Eq. (6) becomes a bordered nonsymmetric saddle-point problem. Solving such saddle-point problems via iterative schemes is challenging. We outline in this section a very effective nested preconditioned iterative scheme, which is also quite suitable for implementation on parallel computers (see 26 ), for handling the saddle-point problem given in Eq. (6). From the discretization of the Navier Stokes equations, and the choice of the time-step size, it can be seen that the symmetric part of A, i.e. A s = (A + A T )/2, is diagonally dominant and positive definite, and that A is positive stable. The preconditioned nested scheme in 26 is given by the following outer inner iterations: ( ) ( ) ( ) A B x b outer: solve B T =, 0 y c ( ) (7) As B inner: solve Mz = r, where M = B T. 0 The system in the outer iterations is solved with GMRES(k), while the system in the inner iterations is solved with a preconditioned Richardson scheme. There are two keys to the effectiveness of the above scheme. First is that for most of the computation A s is more dominant than the skew-symmetric part of A, A ss = (A A T )/2. Second, in the preconditioned Richardson s iteration, we only need to solve two simple linear systems of the form Âv = a, (8) (B T  1 B)w = b. (9) Here  1 is a diagonal matrix that approximates A 1 s and chosen so as to minimize the Frobenius norm of (I A s  1 ). Such minimization can be performed with perfect parallelism, dealing simultaneously with all columns of A s. Moreover, we have shown in 26 that ρ(i A s  1 ) < 1, where ρ(s) is the spectral radius of a square matrix S. Using such Â, solving Eq. (8) becomes trivial, and Eq. (9) needs to be solved via the conjugate gradient algorithm without preconditioning and with only a relaxed convergence criterion (reducing the original residual by a factor of 10 3 ). This nested scheme is quite suitable for matrix-free formulation as one only needs an effective routine for matrix vector products. The algorithm has also proven to be more effective than other preconditioned schemes in which the above inner iterations are solved with GMRES. Our scheme is more effective than using GMRES with ( ) I B M = B T (10) 0 as suggested in 47 or M = ( A 0 0 B T A 1 B as suggested in. 48 Moreover, our scheme in 26 proved to be superior to all ILU preconditioning varieties of GMRES, including its most recent version, ARMS. 49 ) (11)

8 Finite Element Methods: 1970 s and beyond 5 PARALLEL COMPUTING PLATFORMS High performance computing is vital not only for solving important problems in science and engineering disciplines, but also for advancing these disciplines. Vector processors, introduced by Seymour Cray in the late 1970 s, were the supercomputers of the day for conducting full-scale simulations in academic, government, and industry research laboratories. Parallel computers of that decade were mainly experimental with the best known being the Illiac-IV, designed by the late Daniel Slotnick at the University of Illinois, and built by the Burroughs Corporation, and installed at NASA Ames. In the early 1980 s, however, the most powerful supercomputers were the Cray C 90 and T 90 parallel architectures introduced by Cray Research, in which each CPU was a powerful vector processor. However, with the increase in speed of uniprocessors, and with the slow pace of adopting more advanced simulation models, many organizations and research laboratories found it much more economical to use clusters of PC s with various commercially available interconnection networks. Such architectures, including PC-cluster-like architectures introduced by HP and IBM, for example, were also encouraged by the introduction of the Top 500 List. This list is based on theoretical peak (in Gflops) and computational rate achieved in solving a dense linear system using Gaussian elimination (the LINPACK Benchmark). While this benchmark gives an indication of the suitability of the parallel architecture for dense matrix computation, it provides the user with very little idea about the architecture s performance on real-life applications in science and engineering. In fact, experience in industry and government laboratories on such PC-cluster architectures have shown that most real applications achieve sustained performance that cannot exceed 10% of the performance achieved on the LINPACK benchmark. The introduction of the Japanese Earth Simulator, in which the architecture combines parallelism and vector processing, alerted the US to the fact that the direction of high performance computing in the country needed more attention. The Earth Simulator is not only ranked first in the Top 500 list, but also capable of a sustained performance at 30% to 50% of its peak performance on many science and engineering applications. Ironically, the Earth Simulator architecture and system software are similar to some of the high performance computing projects discontinued in the US in the early 1990 s. This is by no means a disapproval of PC-cluster architectures. These are economical and quite suitable for code development but they are certainly not the parallel architectures that can meet the high-end computing-power needs of research and development activities. These concerns have recently prompted US agencies to sponsor studies and workshops that address the future of supercomputing and the need for revitalization of high-end computing in the US. 6 NUMERICAL EXAMPLES 6.1 Parachute soft-landing dynamics Soft landing with the aid of a retraction device reduces the landing impact for payloads delivered with parachutes. In this example, for a T 10 parachute, a pneumatic muscle actuator (PMA) placed between the suspension lines and the payload serves at the retraction device by causing rapid contraction just before landing.

9 T.E. Tezduyar and A. Sameh / PARALLEL COMPUTING The DSD/SST formulation is used with the automatic mesh moving method. The motion and deformation of the parachute is governed by the membrane and cable equations, which are solved together with the fluid mechanics equations. A blockiterative coupling method is used for solving the two systems simultaneously. The blocks of linear equation systems are solved iteratively with the GMRES update method. The computation is performed on a 41-node PC cluster in the Department of Mechanical Engineering and Materials Science at Rice University. This system has two 1.7GHz Pentium IV processors on each node sharing 2 GigaBytes of memory. Figure 1 shows the payload trajectory. For more on soft-landing simulations Figure 1: Parachute soft-landing dynamics. Payload trajectory for soft landing of a T 10 parachute during and immediately after retraction. The straight line is the trajectory that the payload would have had without the retraction. The parachutes displayed illustrate the deformations of the canopy and the cables. The length scale used in displaying the parachutes is not the same as it is for the trajectory graph. and the methods used, see A performance evaluation test for the Balance Scheme If the saddle-point problem discussed in Section 4, i.e., [ ] [ ] [ ] A B x b B T =, (12) 0 y c is reordered to minimize its bandwidth via the reverse Cuthill McKee scheme, for example, we obtain a narrow banded system that is sparse within the band. Naturally, this system is indefinite. It is also rather difficult to solve via a Krylov subspace method without an effective preconditioning strategy. As an alternative to the iterative strategy in Section 4, we describe here the Balance Scheme for iterative solution of these narrow banded systems. This scheme is also quite suitable for

10 Finite Element Methods: 1970 s and beyond parallel computers (see 25 and related work in 51 and 52 ). We show that it is superior to GMRES with ILUT preconditioning and also superior to classical direct methods for solving banded systems. Let the banded system, of order N and bandwidth ν, resulting after reordering Eq. (12) be given by Ku = h. To illustrate the main idea in this scheme (see 25 ) we partition the banded matrix of coefficients K as two coupled linear systems: [ A1 B C 2 A 2 ] u 1 µ u 2 = [ h1 h 2 ]. (13) Here A 1 and A 2 are each of order (N/2) ( ) N ν 2, and B1 and C 2 are each of order (N/2) ν. Thus, we have two underdetermined systems: [ ] [ ] E1 T u1 µ = h µ 1, E2 T = h 2, (14) where E1 T = (A 1, B 1 ) and E2 T = (C 2, A 2 ). The solution to these two systems are given as [ ] [ ] u1 µ = p µ 1 + (I P 1 )z 1, = p 2 + (I P 2 )z 2. (15) Here p i = E i (Ei T E i ) 1 h i, with i = 1, 2, are the particular solutions of system, and the orthogonal projectors P i are given as E i (Ei TE i) 1 Ei T. The particular solutions are obtained using the conjugate gradient algorithm preconditioned via the incomplete Cholesky factorization of (Ei T E i ). The vectors z i are arbitrary but need to be determined to assure that µ = µ. Imposing this condition yields the underdetermined reduced system Mz = q of order ν (N ν), where q is trivially obtained from the vectors p i, and M consists of rows of the projectors (I P i ). This reduced system can be cast in the form u 2 u 2 MM T w = q, (16) where MM T is the sum of sections of (I P i ). Note that M is not formed explicitly, rather, the system in Eq. (16) is solved via the conjugate gradient algorithm. Here one only needs to perform the multiplication of MM T by a vector. This amounts to performing multiplications of (I P i ) by a vector c, i.e., d = (I P i )c. This in turn reduces to computing the residual of the least-squares problem min d c E i d 2, again, using the conjugate gradient algorithm. Once the reduced system is solved for w, the solution vector is obtained in a straightforward way. This algorithm can be generalized to multiple partitions γ > 2, generating γ independent linear least-squares problems that can be solved in parallel for the particular solutions p i, i = 1, 2,..., γ. Naturally there is also the opportunity for parallelism in handling each least-squares problem, thus creating multi-level parallelism. For solving indefinite linear systems this algorithm proved to be superior to GMRES with ILUT preconditioning. This is illustrated via the system resulting from the finite element discretization of the partial differential equation u + α u + βu = f in Ω, u = 0 on Γ.

11 T.E. Tezduyar and A. Sameh / PARALLEL COMPUTING Using an unstructured mesh to discretize the problem domain Ω, we obtain a sparse linear system of equations that can be reordered into a narrow banded system that is sparse within the band. For certain values of α and β the resulting linear system becomes strongly indefinite. Table 1 shows that, even for a system of order as small as 3101, our algorithm (with γ = 8 partitions) proves to be more robust than GM- RES(20) with ILUT preconditioning (allowing a fill-in of up to 15 elements per row and a drop tolerance of 10 4 ). Moreover, since the reordered system results in a very Table 1: Balance GMRES α β iter r 2 iter r failure failure failure narrow band, we also compare the Balance Scheme with the classical banded system solver in the linear algebra package ScaLAPACK. 53 For this purpose, we chose a banded Toeplitz system that is sparse within the band, with the nonzero diagonals at the following distances from the main diagonal: { b l, 1, 1, b u }. The values assigned to these diagonals are: 1, 1, 1, 1. Table 2 shows, on an SGI-Origin 2000, the time consumed by the Balance Scheme (with γ = 16) and the resulting speed improvements over ScaLAPACK. We note that ScaLAPACK reaches its maximum speed when the number of processors is two. Table 2: speed improvement #of Time for over ScaLAPACK Problem processors Balance on 2 processors N = 16, seconds 1 bandwidth = seconds seconds 5 N = 32, seconds 2 bandwidth = seconds seconds 12 7 CONCLUDING REMARKS We provided a brief outline of the role of parallel computing in finite element computations. We focused on giving examples of how we address some of the computational challenges involved in parallel, 3D finite element computation of complex engineering problems. To that end, we provided our overview in the context of computational fluid mechanics, with emphasis on moving boundaries and interfaces. This is a complex but important class of problems that includes fluid object and

12 Finite Element Methods: 1970 s and beyond fluid structure interactions and free-surface and two-fluid flows. In developing the methods needed to address the computational challenges involved in this class of problems, we made sure that 3D computations with these methods could be carried out efficiently on parallel computers. For example, in designing the mesh update methods needed to accommodate the changes in the spatial domain occupied by the fluid, we made sure that parallel efficiency is not disturbed. Similarly, maintaining the parallel efficiency was one of the key principles we observed in developing the iterative methods needed to solve the large, coupled equation systems. To demonstrate how the methods developed can be applied to complex engineering problems, we presented a numerical simulation involving aerodynamics and fluid structure interactions. We also presented a performance evaluation test for a preconditioned iterative method for banded systems. ACKNOWLEDGMENTS The first author was supported by the US Army Natick Soldier Center (Contract No. DAAD16-03-C-0051), NSF (Grant No. EIA ), and NASA Johnson Space Center (Grant No. NAG9-1435). The second author was supported by NSF (Grant No. CCR and ERC ). REFERENCES [1] S.L. Johnsson and K.K. Mathur, Experience With the Conjugate Gradient Method for Stress Analysis on a Data Parallel Supercomputer, International Journal for Numerical Methods in Engineering, 27 (1989) [2] K.K. Mathur and S.L. Johnsson, The finite element method on a data parallel computing system, International Journal of High Speed Computing, 1 (1989) [3] S.L. Johnsson and K.K. Mathur, Data Structures and Algorithms for the Finite Element Method on a Data Parallel Supercomputer, International Journal for Numerical Methods in Engineering, 29 (1990) [4] Z. Johan, K.K. Mathur, S.L. Johnsson, and T.J.R. Hughes, An efficient communications strategy for finite element methods on the Connection Machine CM-5 system, Computer Methods in Applied Mechanics and Engineering, 113 (1994) [5] Z. Johan, K.K. Mathur, S.L. Johnsson, and T.J.R. Hughes, Scalability of finite element applications on distributed-memory parallel computers, Computer Methods in Applied Mechanics and Engineering, 119 (1994) [6] Z. Johan, K.K. Mathur, S.L. Johnsson, and T.J.R. Hughes, A case study in parallel computation: Viscous flow around an Onera M6 wing, International Journal for Numerical Methods in Fluids, 21 (1995) [7] T.E. Tezduyar, M. Behr, S. Mittal, and A.A. Johnson, Computation of unsteady incompressible flows with the finite element methods space time formulations, iterative strategies and massively parallel implementations, in New

13 T.E. Tezduyar and A. Sameh / PARALLEL COMPUTING Methods in Transient Analysis, PVP-Vol.246/AMD-Vol.143, ASME, New York, (1992) [8] T. Tezduyar, S. Aliabadi, M. Behr, A. Johnson, and S. Mittal, Parallel finiteelement computation of 3D flows, Computer, 26 (1993) [9] T.E. Tezduyar, S.K. Aliabadi, M. Behr, and S. Mittal, Massively parallel finite element simulation of compressible and incompressible flows, Computer Methods in Applied Mechanics and Engineering, 119 (1994) [10] C. Farhat, L. Fezoui, and S. Lanteri, Two-dimensional viscous flow computations on the CM-2: Unstructured meshes, upwind schemes and massively parallel computations, Computer Methods in Applied Mechanics and Engineering, 102 (1993) [11] C. Farhat and S. Lanteri, Simulation of compressible viscous flows on a variety of MPPs: Computational algorithms for unstructured dynamic meshes and performance results, Computer Methods in Applied Mechanics and Engineering, 119 (1994) [12] C. Farhat, M. Lesoinne, and N. Maman, Mixed explicit/implicit time integration of coupled aeroelastic problems: Three-field formulation, geometric conservation and distributed solution, International Journal for Numerical Methods in Fluids, 21 (1995) [13] C. Özturan, H.L. decougny, M.S. Shephard, and J.E. Flaherty, Parallel adaptive mesh refinement and redistribution on distributed memory computers, Computer Methods in Applied Mechanics and Engineering, 119 (1994) [14] T.E. Tezduyar, Stabilized finite element formulations for incompressible flow computations, Advances in Applied Mechanics, 28 (1992) [15] T.E. Tezduyar, M. Behr, and J. Liou, A new strategy for finite element computations involving moving boundaries and interfaces the deformingspatial-domain/space time procedure: I. The concept and the preliminary numerical tests, Computer Methods in Applied Mechanics and Engineering, 94 (1992) [16] T.E. Tezduyar, M. Behr, S. Mittal, and J. Liou, A new strategy for finite element computations involving moving boundaries and interfaces the deformingspatial-domain/space time procedure: II. Computation of free-surface flows, two-liquid flows, and flows with drifting cylinders, Computer Methods in Applied Mechanics and Engineering, 94 (1992) [17] T.E. Tezduyar, Finite element methods for flow problems with moving boundaries and interfaces, Archives of Computational Methods in Engineering, 8 (2001)

14 Finite Element Methods: 1970 s and beyond [18] T.E. Tezduyar, Finite element methods for fluid dynamics with moving boundaries and interfaces, in E. Stein, R. De Borst, and T.J.R. Hughes, editors, Encyclopedia of Computational Mechanics, Volume 3: Fluids, Chapter 17, John Wiley & Sons, [19] T.J.R. Hughes and A.N. Brooks, A multi-dimensional upwind scheme with no crosswind diffusion, in T.J.R. Hughes, editor, Finite Element Methods for Convection Dominated Flows, AMD-Vol.34, 19 35, ASME, New York, [20] A.N. Brooks and T.J.R. Hughes, Streamline upwind/petrov-galerkin formulations for convection dominated flows with particular emphasis on the incompressible Navier-Stokes equations, Computer Methods in Applied Mechanics and Engineering, 32 (1982) [21] T.E. Tezduyar and T.J.R. Hughes, Finite element formulations for convection dominated flows with particular emphasis on the compressible Euler equations, in Proceedings of AIAA 21st Aerospace Sciences Meeting, AIAA Paper , Reno, Nevada, (1983). [22] T.J.R Hughes and T.E. Tezduyar, Finite element methods for firstorder hyperbolic systems with particular emphasis on the compressible Euler equations, Computer Methods in Applied Mechanics and Engineering, 45 (1984) [23] T.E. Tezduyar, S. Mittal, S.E. Ray, and R. Shih, Incompressible flow computations with stabilized bilinear and linear equal-order-interpolation velocitypressure elements, Computer Methods in Applied Mechanics and Engineering, 95 (1992) [24] T.J.R. Hughes, L.P. Franca, and M. Balestra, A new finite element formulation for computational fluid dynamics: V. Circumventing the Babuška Brezzi condition: A stable Petrov Galerkin formulation of the Stokes problem accommodating equal-order interpolations, Computer Methods in Applied Mechanics and Engineering, 59 (1986) [25] G. Golub, A. Sameh, and V. Sarin, A parallel balance scheme for banded linear systems, Num. Lin. Alg. Appl., 8 (2001) [26] A. Baggag and A. Sameh, A nested iterative scheme for indefinite linear systems in particulate flows, to appear in Computer Methods in Applied Mechanics and Engineering. [27] T.J.R. Hughes and G.M. Hulbert, Space time finite element methods for elastodynamics: formulations and error estimates, Computer Methods in Applied Mechanics and Engineering, 66 (1988) [28] T. Tezduyar, S. Aliabadi, and M. Behr, Enhanced-Discretization Interface- Capturing Technique (EDICT) for computation of unsteady flows with interfaces, Computer Methods in Applied Mechanics and Engineering, 155 (1998)

15 T.E. Tezduyar and A. Sameh / PARALLEL COMPUTING [29] T.E. Tezduyar and J. Liou, Grouped element-by-element iteration schemes for incompressible flow computations, Computer Physcis Communications, 53 (1989) [30] T.E. Tezduyar, M. Behr, S.K. Aliabadi, S. Mittal, and S.E. Ray, A new mixed preconditioning method for finite element computations, Computer Methods in Applied Mechanics and Engineering, 99 (1992) [31] T. Tezduyar, S. Aliabadi, M. Behr, A. Johnson, V. Kalro, and M. Litke, Flow simulation and high performance computing, Computational Mechanics, 18 (1996) [32] V. Kalro and T. Tezduyar, Parallel iterative computational methods for 3D finite element flow simulations, Computer Assisted Mechanics and Engineering Sciences, 5 (1998) [33] T.E. Tezduyar, CFD methods for three-dimensional computation of complex flow problems, Journal of Wind Engineering and Industrial Aerodynamics, 81 (1999) [34] T. Tezduyar and Y. Osawa, Methods for parallel computation of complex flow problems, Parallel Computing, 25 (1999) [35] D.R. Lynch, Wakes in liquid-liquid systems, Journal of Computational Physics, 47 (1982) [36] A. Masud and T.J.R. Hughes, A space time Galerkin/least-squares finite element formulation of the Navier-Stokes equations for moving domain problems, Computer Methods in Applied Mechanics and Engineering, 146 (1997) [37] K. Stein, T. Tezduyar, and R. Benney, Mesh moving techniques for fluid structure interactions with large displacements, Journal of Applied Mechanics, 70 (2003) [38] T. Tezduyar, Finite element interface-tracking and interface-capturing techniques for flows with moving boundaries and interfaces, in Proceedings of the ASME Symposium on Fluid-Physics and Heat Transfer for Macro- and Micro-Scale Gas-Liquid and Phase-Change Flows (CD-ROM), ASME Paper IMECE2001/HTD-24206, ASME, New York, New York, (2001). [39] Y. Saad and M. Schultz, GMRES: A generalized minimal residual algorithm for solving nonsymmetric linear systems, SIAM Journal of Scientific and Statistical Computing, 7 (1986) [40] T. Tezduyar, S. Aliabadi, M. Behr, A. Johnson, V. Kalro, and M. Litke, High performance computing techniques for flow simulations, in M. Papadrakakis, editor, Solving Large-Scale Problems in Mechanics: Parallel Solution Methods in Computational Mechanics, Chapter 10, , John Wiley & Sons, 1997.

16 Finite Element Methods: 1970 s and beyond [41] Z. Johan, T.J.R. Hughes, and F. Shakib, A globally convergent matrix-free algorithm for implicit time-marching schemes arising in finite element analysis in fluids, Computer Methods in Applied Mechanics and Engineering, 87 (1991) [42] A.A. Johnson and T.E. Tezduyar, Simulation of multiple spheres falling in a liquid-filled tube, Computer Methods in Applied Mechanics and Engineering, 134 (1996) [43] A.A. Johnson and T.E. Tezduyar, 3D simulation of fuid-particle interactions with the number of particles reaching 100, Computer Methods in Applied Mechanics and Engineering, 145 (1997) [44] A.A. Johnson and T.E. Tezduyar, Advanced mesh generation and update methods for 3D flow simulations, Computational Mechanics, 23 (1999) [45] A. Johnson and T. Tezduyar, Methods for 3D computation of fluid-object interactions in spatially-periodic flows, Computer Methods in Applied Mechanics and Engineering, 190 (2001) [46] M. Knepley, V. Sarin, and A. Sameh, Parallel simulation of particulate flows, in Lecture Notes in Computer Science, number 1457, , Springer, [47] I. Perugia and V. Simoncini, Block-diagonal and indefinite symmetric preconditioners for mixed finite element formulations, Num. Lin. Alg. Appl., 7 (2000) [48] M. Murphy, G. Golub, and A. Wathen, A note on preconditioning for indefinite linear systems, SIAM J. Sci. Comput., 21 (2000) [49] Y. SaaD and B. Suchomel, Arms: an algebraic recursive multi-level solver for general sparse linear systems, Num. Lin. Alg. Appl., 9 (2000) [50] K. Stein, T. Tezduyar, S. Sathe, M. Senga, C. Ozcan, T. Soltys, V. Kumar, R. Benney, and R. Charles, Simulation of parachute dynamics during control line input operations, in Proceedings of the 17th AIAA Aerodynamic Decelerator Systems Technology Conference, AIAA Paper , Monterey, California, (2003). [51] A. Sameh and V. Sarin, Parallel algorithms for indefinite linear systems, Parallel Computing, 28 (2002) [52] A. Sameh and V. Sarin, Hybrid parallel linear system solvers, Int. J. of Comp. Fluid Dynamics, 12 (1999) [53] L. Blackford et al., ScaLAPACK User s Guide. SIAM, Philadelphia, 1997.

Advanced Mesh Update Techniques for Problems Involving Large Displacements

Advanced Mesh Update Techniques for Problems Involving Large Displacements WCCM V Fifth World Congress on Computational Mechanics July 7,, Vienna, Austria Eds.: H.A. Mang, F.G. Rammerstorfer, J. Eberhardsteiner Advanced Mesh Update Techniques for Problems Involving Large Displacements

More information

A NEW MIXED PRECONDITIONING METHOD BASED ON THE CLUSTERED ELEMENT -BY -ELEMENT PRECONDITIONERS

A NEW MIXED PRECONDITIONING METHOD BASED ON THE CLUSTERED ELEMENT -BY -ELEMENT PRECONDITIONERS Contemporary Mathematics Volume 157, 1994 A NEW MIXED PRECONDITIONING METHOD BASED ON THE CLUSTERED ELEMENT -BY -ELEMENT PRECONDITIONERS T.E. Tezduyar, M. Behr, S.K. Aliabadi, S. Mittal and S.E. Ray ABSTRACT.

More information

COMPUTATIONAL METHODS FOR ENVIRONMENTAL FLUID MECHANICS

COMPUTATIONAL METHODS FOR ENVIRONMENTAL FLUID MECHANICS COMPUTATIONAL METHODS FOR ENVIRONMENTAL FLUID MECHANICS Tayfun Tezduyar tezduyar@rice.edu Team for Advanced Flow Simulation and Modeling (T*AFSM) Mechanical Engineering and Materials Science Rice University

More information

Corrected/Updated References

Corrected/Updated References K. Kashiyama, H. Ito, M. Behr and T. Tezduyar, "Massively Parallel Finite Element Strategies for Large-Scale Computation of Shallow Water Flows and Contaminant Transport", Extended Abstracts of the Second

More information

COMPUTATIONAL FLUIDAND SOLID MECHANICS

COMPUTATIONAL FLUIDAND SOLID MECHANICS COMPUTATIONAL FLUIDAND SOLID MECHANICS K.J. Bathe, editor Proceedings First MIT Conference on ComDutational Fluid and Solid Mechanics June t2.t5,2oot ElSeVief 968 Aerodynamic interaction between multiple

More information

We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors

We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists. International authors and editors We are IntechOpen, the world s leading publisher of Open Access books Built by scientists, for scientists 3,800 116,000 120M Open access books available International authors and editors Downloads Our

More information

A High-Order Accurate Unstructured GMRES Solver for Poisson s Equation

A High-Order Accurate Unstructured GMRES Solver for Poisson s Equation A High-Order Accurate Unstructured GMRES Solver for Poisson s Equation Amir Nejat * and Carl Ollivier-Gooch Department of Mechanical Engineering, The University of British Columbia, BC V6T 1Z4, Canada

More information

MESHLESS SOLUTION OF INCOMPRESSIBLE FLOW OVER BACKWARD-FACING STEP

MESHLESS SOLUTION OF INCOMPRESSIBLE FLOW OVER BACKWARD-FACING STEP Vol. 12, Issue 1/2016, 63-68 DOI: 10.1515/cee-2016-0009 MESHLESS SOLUTION OF INCOMPRESSIBLE FLOW OVER BACKWARD-FACING STEP Juraj MUŽÍK 1,* 1 Department of Geotechnics, Faculty of Civil Engineering, University

More information

SELECTIVE ALGEBRAIC MULTIGRID IN FOAM-EXTEND

SELECTIVE ALGEBRAIC MULTIGRID IN FOAM-EXTEND Student Submission for the 5 th OpenFOAM User Conference 2017, Wiesbaden - Germany: SELECTIVE ALGEBRAIC MULTIGRID IN FOAM-EXTEND TESSA UROIĆ Faculty of Mechanical Engineering and Naval Architecture, Ivana

More information

1.2 Numerical Solutions of Flow Problems

1.2 Numerical Solutions of Flow Problems 1.2 Numerical Solutions of Flow Problems DIFFERENTIAL EQUATIONS OF MOTION FOR A SIMPLIFIED FLOW PROBLEM Continuity equation for incompressible flow: 0 Momentum (Navier-Stokes) equations for a Newtonian

More information

AMS526: Numerical Analysis I (Numerical Linear Algebra)

AMS526: Numerical Analysis I (Numerical Linear Algebra) AMS526: Numerical Analysis I (Numerical Linear Algebra) Lecture 20: Sparse Linear Systems; Direct Methods vs. Iterative Methods Xiangmin Jiao SUNY Stony Brook Xiangmin Jiao Numerical Analysis I 1 / 26

More information

cuibm A GPU Accelerated Immersed Boundary Method

cuibm A GPU Accelerated Immersed Boundary Method cuibm A GPU Accelerated Immersed Boundary Method S. K. Layton, A. Krishnan and L. A. Barba Corresponding author: labarba@bu.edu Department of Mechanical Engineering, Boston University, Boston, MA, 225,

More information

Mesh Decomposition and Communication Procedures for Finite Element Applications on the Connection Machine CM-5 System

Mesh Decomposition and Communication Procedures for Finite Element Applications on the Connection Machine CM-5 System Mesh Decomposition and Communication Procedures for Finite Element Applications on the Connection Machine CM-5 System The Harvard community has made this article openly available. Please share how this

More information

STABILIZED FINITE ELEMENT METHODS FOR INCOMPRESSIBLE FLOWS WITH EMPHASIS ON MOVING BOUNDARIES AND INTERFACES

STABILIZED FINITE ELEMENT METHODS FOR INCOMPRESSIBLE FLOWS WITH EMPHASIS ON MOVING BOUNDARIES AND INTERFACES STABILIZED FINITE ELEMENT METHODS FOR INCOMPRESSIBLE FLOWS WITH EMPHASIS ON MOVING BOUNDARIES AND INTERFACES A THESIS SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY Marek

More information

NIA CFD Seminar, October 4, 2011 Hyperbolic Seminar, NASA Langley, October 17, 2011

NIA CFD Seminar, October 4, 2011 Hyperbolic Seminar, NASA Langley, October 17, 2011 NIA CFD Seminar, October 4, 2011 Hyperbolic Seminar, NASA Langley, October 17, 2011 First-Order Hyperbolic System Method If you have a CFD book for hyperbolic problems, you have a CFD book for all problems.

More information

Solution of 2D Euler Equations and Application to Airfoil Design

Solution of 2D Euler Equations and Application to Airfoil Design WDS'6 Proceedings of Contributed Papers, Part I, 47 52, 26. ISBN 8-86732-84-3 MATFYZPRESS Solution of 2D Euler Equations and Application to Airfoil Design J. Šimák Charles University, Faculty of Mathematics

More information

Development of an Integrated Computational Simulation Method for Fluid Driven Structure Movement and Acoustics

Development of an Integrated Computational Simulation Method for Fluid Driven Structure Movement and Acoustics Development of an Integrated Computational Simulation Method for Fluid Driven Structure Movement and Acoustics I. Pantle Fachgebiet Strömungsmaschinen Karlsruher Institut für Technologie KIT Motivation

More information

EDICT for 3D computation of two- uid interfaces q

EDICT for 3D computation of two- uid interfaces q Comput. Methods Appl. Mech. Engrg. 190 (2000) 403±410 www.elsevier.com/locate/cma EDICT for 3D computation of two- uid interfaces q Tayfun E. Tezduyar a, *, Shahrouz Aliabadi b a Mechanical Engineering

More information

Case C3.1: Turbulent Flow over a Multi-Element MDA Airfoil

Case C3.1: Turbulent Flow over a Multi-Element MDA Airfoil Case C3.1: Turbulent Flow over a Multi-Element MDA Airfoil Masayuki Yano and David L. Darmofal Aerospace Computational Design Laboratory, Massachusetts Institute of Technology I. Code Description ProjectX

More information

Isogeometric Analysis of Fluid-Structure Interaction

Isogeometric Analysis of Fluid-Structure Interaction Isogeometric Analysis of Fluid-Structure Interaction Y. Bazilevs, V.M. Calo, T.J.R. Hughes Institute for Computational Engineering and Sciences, The University of Texas at Austin, USA e-mail: {bazily,victor,hughes}@ices.utexas.edu

More information

Computational Fluid Dynamics - Incompressible Flows

Computational Fluid Dynamics - Incompressible Flows Computational Fluid Dynamics - Incompressible Flows March 25, 2008 Incompressible Flows Basis Functions Discrete Equations CFD - Incompressible Flows CFD is a Huge field Numerical Techniques for solving

More information

ALE METHODS FOR DETERMINING STATIONARY SOLUTIONS OF METAL FORMING PROCESSES

ALE METHODS FOR DETERMINING STATIONARY SOLUTIONS OF METAL FORMING PROCESSES European Congress on Computational Methods in Applied Sciences and Engineering ECCOMAS 2000 Barcelona, 11-14 September 2000 c ECCOMAS ALE METHODS FOR DETERMINING STATIONARY SOLUTIONS OF METAL FORMING PROCESSES

More information

Numerical Simulations of Fluid-Structure Interaction Problems using MpCCI

Numerical Simulations of Fluid-Structure Interaction Problems using MpCCI Numerical Simulations of Fluid-Structure Interaction Problems using MpCCI François Thirifay and Philippe Geuzaine CENAERO, Avenue Jean Mermoz 30, B-6041 Gosselies, Belgium Abstract. This paper reports

More information

The Development of a Navier-Stokes Flow Solver with Preconditioning Method on Unstructured Grids

The Development of a Navier-Stokes Flow Solver with Preconditioning Method on Unstructured Grids Proceedings of the International MultiConference of Engineers and Computer Scientists 213 Vol II, IMECS 213, March 13-15, 213, Hong Kong The Development of a Navier-Stokes Flow Solver with Preconditioning

More information

FINITE ELEMENT SOLUTION OF NAVIER-STOKES EQUATIONS USING KRYLOV SUBSPACE METHODS

FINITE ELEMENT SOLUTION OF NAVIER-STOKES EQUATIONS USING KRYLOV SUBSPACE METHODS HEFAT2014 10 th International Conference on Heat Transfer, Fluid Mechanics and Thermodynamics 14 16 July 2014 Orlando, Florida FINITE ELEMENT SOLUTION OF NAVIER-STOKES EQUATIONS USING KRYLOV SUBSPACE METHODS

More information

High-Order Numerical Algorithms for Steady and Unsteady Simulation of Viscous Compressible Flow with Shocks (Grant FA )

High-Order Numerical Algorithms for Steady and Unsteady Simulation of Viscous Compressible Flow with Shocks (Grant FA ) High-Order Numerical Algorithms for Steady and Unsteady Simulation of Viscous Compressible Flow with Shocks (Grant FA9550-07-0195) Sachin Premasuthan, Kui Ou, Patrice Castonguay, Lala Li, Yves Allaneau,

More information

Hierarchical Divergence-Free Bases and Their Application to Particulate Flows

Hierarchical Divergence-Free Bases and Their Application to Particulate Flows V. Sarin 1 Department of Computer Science, Texas A&M University, College Station, TX 77843 e-mail: sarin@cs.tamu.edu A. H. Sameh Department of Computer Science, Purdue University, West Lafayette, IN 47907

More information

Fluid structure interaction modeling of complex parachute designs with the space time finite element techniques

Fluid structure interaction modeling of complex parachute designs with the space time finite element techniques Computers & Fluids xxx (25) xxx xxx www.elsevier.com/locate/compfluid Fluid structure interaction modeling of complex parachute designs with the space time finite element techniques S. Sathe a, *, R. Benney

More information

The 3D DSC in Fluid Simulation

The 3D DSC in Fluid Simulation The 3D DSC in Fluid Simulation Marek K. Misztal Informatics and Mathematical Modelling, Technical University of Denmark mkm@imm.dtu.dk DSC 2011 Workshop Kgs. Lyngby, 26th August 2011 Governing Equations

More information

Numerical and theoretical analysis of shock waves interaction and reflection

Numerical and theoretical analysis of shock waves interaction and reflection Fluid Structure Interaction and Moving Boundary Problems IV 299 Numerical and theoretical analysis of shock waves interaction and reflection K. Alhussan Space Research Institute, King Abdulaziz City for

More information

Keywords: Block ILU preconditioner, Krylov subspace methods, Additive Schwarz, Domain decomposition

Keywords: Block ILU preconditioner, Krylov subspace methods, Additive Schwarz, Domain decomposition BLOCK ILU PRECONDITIONERS FOR PARALLEL AMR/C SIMULATIONS Jose J. Camata Alvaro L. G. A. Coutinho Federal University of Rio de Janeiro, NACAD, COPPE Department of Civil Engineering, Rio de Janeiro, Brazil

More information

Incompressible Viscous Flow Simulations Using the Petrov-Galerkin Finite Element Method

Incompressible Viscous Flow Simulations Using the Petrov-Galerkin Finite Element Method Copyright c 2007 ICCES ICCES, vol.4, no.1, pp.11-18, 2007 Incompressible Viscous Flow Simulations Using the Petrov-Galerkin Finite Element Method Kazuhiko Kakuda 1, Tomohiro Aiso 1 and Shinichiro Miura

More information

Chapter 6. Petrov-Galerkin Formulations for Advection Diffusion Equation

Chapter 6. Petrov-Galerkin Formulations for Advection Diffusion Equation Chapter 6 Petrov-Galerkin Formulations for Advection Diffusion Equation In this chapter we ll demonstrate the difficulties that arise when GFEM is used for advection (convection) dominated problems. Several

More information

FAST ALGORITHMS FOR CALCULATIONS OF VISCOUS INCOMPRESSIBLE FLOWS USING THE ARTIFICIAL COMPRESSIBILITY METHOD

FAST ALGORITHMS FOR CALCULATIONS OF VISCOUS INCOMPRESSIBLE FLOWS USING THE ARTIFICIAL COMPRESSIBILITY METHOD TASK QUARTERLY 12 No 3, 273 287 FAST ALGORITHMS FOR CALCULATIONS OF VISCOUS INCOMPRESSIBLE FLOWS USING THE ARTIFICIAL COMPRESSIBILITY METHOD ZBIGNIEW KOSMA Institute of Applied Mechanics, Technical University

More information

Available online at ScienceDirect. Procedia Engineering 90 (2014 )

Available online at  ScienceDirect. Procedia Engineering 90 (2014 ) Available online at www.sciencedirect.com ScienceDirect Procedia Engineering 90 (2014 ) 288 293 10th International Conference on Mechanical Engineering, ICME 2013 Parallelization of enriched free mesh

More information

LS-DYNA 980 : Recent Developments, Application Areas and Validation Process of the Incompressible fluid solver (ICFD) in LS-DYNA.

LS-DYNA 980 : Recent Developments, Application Areas and Validation Process of the Incompressible fluid solver (ICFD) in LS-DYNA. 12 th International LS-DYNA Users Conference FSI/ALE(1) LS-DYNA 980 : Recent Developments, Application Areas and Validation Process of the Incompressible fluid solver (ICFD) in LS-DYNA Part 1 Facundo Del

More information

Comparison of parallel preconditioners for a Newton-Krylov flow solver

Comparison of parallel preconditioners for a Newton-Krylov flow solver Comparison of parallel preconditioners for a Newton-Krylov flow solver Jason E. Hicken, Michal Osusky, and David W. Zingg 1Introduction Analysis of the results from the AIAA Drag Prediction workshops (Mavriplis

More information

Parallel High-Order Geometric Multigrid Methods on Adaptive Meshes for Highly Heterogeneous Nonlinear Stokes Flow Simulations of Earth s Mantle

Parallel High-Order Geometric Multigrid Methods on Adaptive Meshes for Highly Heterogeneous Nonlinear Stokes Flow Simulations of Earth s Mantle ICES Student Forum The University of Texas at Austin, USA November 4, 204 Parallel High-Order Geometric Multigrid Methods on Adaptive Meshes for Highly Heterogeneous Nonlinear Stokes Flow Simulations of

More information

Recent developments in the solution of indefinite systems Location: De Zwarte Doos (TU/e campus)

Recent developments in the solution of indefinite systems Location: De Zwarte Doos (TU/e campus) 1-day workshop, TU Eindhoven, April 17, 2012 Recent developments in the solution of indefinite systems Location: De Zwarte Doos (TU/e campus) :10.25-10.30: Opening and word of welcome 10.30-11.15: Michele

More information

A MULTI-DOMAIN ALE ALGORITHM FOR SIMULATING FLOWS INSIDE FREE-PISTON DRIVEN HYPERSONIC TEST FACILITIES

A MULTI-DOMAIN ALE ALGORITHM FOR SIMULATING FLOWS INSIDE FREE-PISTON DRIVEN HYPERSONIC TEST FACILITIES A MULTI-DOMAIN ALE ALGORITHM FOR SIMULATING FLOWS INSIDE FREE-PISTON DRIVEN HYPERSONIC TEST FACILITIES Khalil Bensassi, and Herman Deconinck Von Karman Institute for Fluid Dynamics Aeronautics & Aerospace

More information

Parallel edge-based implementation of the finite element method for shallow water equations

Parallel edge-based implementation of the finite element method for shallow water equations Parallel edge-based implementation of the finite element method for shallow water equations I. Slobodcicov, F.L.B. Ribeiro, & A.L.G.A. Coutinho Programa de Engenharia Civil, COPPE / Universidade Federal

More information

Numerical Methods in Aerodynamics. Fluid Structure Interaction. Lecture 4: Fluid Structure Interaction

Numerical Methods in Aerodynamics. Fluid Structure Interaction. Lecture 4: Fluid Structure Interaction Fluid Structure Interaction Niels N. Sørensen Professor MSO, Ph.D. Department of Civil Engineering, Alborg University & Wind Energy Department, Risø National Laboratory Technical University of Denmark

More information

Driven Cavity Example

Driven Cavity Example BMAppendixI.qxd 11/14/12 6:55 PM Page I-1 I CFD Driven Cavity Example I.1 Problem One of the classic benchmarks in CFD is the driven cavity problem. Consider steady, incompressible, viscous flow in a square

More information

FOURTH ORDER COMPACT FORMULATION OF STEADY NAVIER-STOKES EQUATIONS ON NON-UNIFORM GRIDS

FOURTH ORDER COMPACT FORMULATION OF STEADY NAVIER-STOKES EQUATIONS ON NON-UNIFORM GRIDS International Journal of Mechanical Engineering and Technology (IJMET Volume 9 Issue 10 October 2018 pp. 179 189 Article ID: IJMET_09_10_11 Available online at http://www.iaeme.com/ijmet/issues.asp?jtypeijmet&vtype9&itype10

More information

TAU mesh deformation. Thomas Gerhold

TAU mesh deformation. Thomas Gerhold TAU mesh deformation Thomas Gerhold The parallel mesh deformation of the DLR TAU-Code Introduction Mesh deformation method & Parallelization Results & Applications Conclusion & Outlook Introduction CFD

More information

Faculty of Mechanical and Manufacturing Engineering, University Tun Hussein Onn Malaysia (UTHM), Parit Raja, Batu Pahat, Johor, Malaysia

Faculty of Mechanical and Manufacturing Engineering, University Tun Hussein Onn Malaysia (UTHM), Parit Raja, Batu Pahat, Johor, Malaysia Applied Mechanics and Materials Vol. 393 (2013) pp 305-310 (2013) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/amm.393.305 The Implementation of Cell-Centred Finite Volume Method

More information

Application of Finite Volume Method for Structural Analysis

Application of Finite Volume Method for Structural Analysis Application of Finite Volume Method for Structural Analysis Saeed-Reza Sabbagh-Yazdi and Milad Bayatlou Associate Professor, Civil Engineering Department of KNToosi University of Technology, PostGraduate

More information

Mid-Year Report. Discontinuous Galerkin Euler Equation Solver. Friday, December 14, Andrey Andreyev. Advisor: Dr.

Mid-Year Report. Discontinuous Galerkin Euler Equation Solver. Friday, December 14, Andrey Andreyev. Advisor: Dr. Mid-Year Report Discontinuous Galerkin Euler Equation Solver Friday, December 14, 2012 Andrey Andreyev Advisor: Dr. James Baeder Abstract: The focus of this effort is to produce a two dimensional inviscid,

More information

A higher-order finite volume method with collocated grid arrangement for incompressible flows

A higher-order finite volume method with collocated grid arrangement for incompressible flows Computational Methods and Experimental Measurements XVII 109 A higher-order finite volume method with collocated grid arrangement for incompressible flows L. Ramirez 1, X. Nogueira 1, S. Khelladi 2, J.

More information

Investigation of cross flow over a circular cylinder at low Re using the Immersed Boundary Method (IBM)

Investigation of cross flow over a circular cylinder at low Re using the Immersed Boundary Method (IBM) Computational Methods and Experimental Measurements XVII 235 Investigation of cross flow over a circular cylinder at low Re using the Immersed Boundary Method (IBM) K. Rehman Department of Mechanical Engineering,

More information

Partial Differential Equations

Partial Differential Equations Simulation in Computer Graphics Partial Differential Equations Matthias Teschner Computer Science Department University of Freiburg Motivation various dynamic effects and physical processes are described

More information

Case C2.2: Turbulent, Transonic Flow over an RAE 2822 Airfoil

Case C2.2: Turbulent, Transonic Flow over an RAE 2822 Airfoil Case C2.2: Turbulent, Transonic Flow over an RAE 2822 Airfoil Masayuki Yano and David L. Darmofal Aerospace Computational Design Laboratory, Massachusetts Institute of Technology I. Code Description ProjectX

More information

The Immersed Interface Method

The Immersed Interface Method The Immersed Interface Method Numerical Solutions of PDEs Involving Interfaces and Irregular Domains Zhiiin Li Kazufumi Ito North Carolina State University Raleigh, North Carolina Society for Industrial

More information

Scalability of Finite Element Applications on Distributed-Memory Parallel Computers

Scalability of Finite Element Applications on Distributed-Memory Parallel Computers Scalability of Finite Element Applications on Distributed-Memory Parallel Computers The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters.

More information

Numerical Analysis of Shock Tube Problem by using TVD and ACM Schemes

Numerical Analysis of Shock Tube Problem by using TVD and ACM Schemes Numerical Analysis of Shock Tube Problem by using TVD and Schemes Dr. Mukkarum Husain, Dr. M. Nauman Qureshi, Syed Zaid Hasany IST Karachi, Email: mrmukkarum@yahoo.com Abstract Computational Fluid Dynamics

More information

High-Order Navier-Stokes Simulations using a Sparse Line-Based Discontinuous Galerkin Method

High-Order Navier-Stokes Simulations using a Sparse Line-Based Discontinuous Galerkin Method High-Order Navier-Stokes Simulations using a Sparse Line-Based Discontinuous Galerkin Method Per-Olof Persson University of California, Berkeley, Berkeley, CA 9472-384, U.S.A. We study some of the properties

More information

Modeling External Compressible Flow

Modeling External Compressible Flow Tutorial 3. Modeling External Compressible Flow Introduction The purpose of this tutorial is to compute the turbulent flow past a transonic airfoil at a nonzero angle of attack. You will use the Spalart-Allmaras

More information

STABILIZED FINITE ELEMENT METHOD WITH AN ALE STRATEGY TO SOLVE MOVING BOUNDARIES PROBLEMS

STABILIZED FINITE ELEMENT METHOD WITH AN ALE STRATEGY TO SOLVE MOVING BOUNDARIES PROBLEMS STABILIZED FINITE ELEMENT METHOD WITH AN ALE STRATEGY TO SOLVE MOVING BOUNDARIES PROBLEMS M. Viale and N. Nigro Universidad Nacional de Rosario Pelegrini 5, () Rosario, Argentina nnigro@intec.unl.edu.ar

More information

Module 1: Introduction to Finite Difference Method and Fundamentals of CFD Lecture 13: The Lecture deals with:

Module 1: Introduction to Finite Difference Method and Fundamentals of CFD Lecture 13: The Lecture deals with: The Lecture deals with: Some more Suggestions for Improvement of Discretization Schemes Some Non-Trivial Problems with Discretized Equations file:///d /chitra/nptel_phase2/mechanical/cfd/lecture13/13_1.htm[6/20/2012

More information

Development of a Consistent Discrete Adjoint Solver for the SU 2 Framework

Development of a Consistent Discrete Adjoint Solver for the SU 2 Framework Development of a Consistent Discrete Adjoint Solver for the SU 2 Framework Tim Albring, Max Sagebaum, Nicolas Gauger Chair for Scientific Computing TU Kaiserslautern 16th Euro-AD Workshop, Jena December

More information

Solving Partial Differential Equations on Overlapping Grids

Solving Partial Differential Equations on Overlapping Grids **FULL TITLE** ASP Conference Series, Vol. **VOLUME**, **YEAR OF PUBLICATION** **NAMES OF EDITORS** Solving Partial Differential Equations on Overlapping Grids William D. Henshaw Centre for Applied Scientific

More information

Introduction to ANSYS CFX

Introduction to ANSYS CFX Workshop 03 Fluid flow around the NACA0012 Airfoil 16.0 Release Introduction to ANSYS CFX 2015 ANSYS, Inc. March 13, 2015 1 Release 16.0 Workshop Description: The flow simulated is an external aerodynamics

More information

CFD MODELING FOR PNEUMATIC CONVEYING

CFD MODELING FOR PNEUMATIC CONVEYING CFD MODELING FOR PNEUMATIC CONVEYING Arvind Kumar 1, D.R. Kaushal 2, Navneet Kumar 3 1 Associate Professor YMCAUST, Faridabad 2 Associate Professor, IIT, Delhi 3 Research Scholar IIT, Delhi e-mail: arvindeem@yahoo.co.in

More information

MATHEMATICAL ANALYSIS, MODELING AND OPTIMIZATION OF COMPLEX HEAT TRANSFER PROCESSES

MATHEMATICAL ANALYSIS, MODELING AND OPTIMIZATION OF COMPLEX HEAT TRANSFER PROCESSES MATHEMATICAL ANALYSIS, MODELING AND OPTIMIZATION OF COMPLEX HEAT TRANSFER PROCESSES Goals of research Dr. Uldis Raitums, Dr. Kārlis Birģelis To develop and investigate mathematical properties of algorithms

More information

Impact of Far-Field Interactions on Performance of Multipole-Based Preconditioners for Sparse Linear Systems

Impact of Far-Field Interactions on Performance of Multipole-Based Preconditioners for Sparse Linear Systems Impact of Far-Field Interactions on Performance of Multipole-Based Preconditioners for Sparse Linear Systems Ananth Y. Grama Department of Computer Science Purdue University West Lafayette, IN 47907 ayg@cs.purdue.edu

More information

Stability Analysis of the Muscl Method on General Unstructured Grids for Applications to Compressible Fluid Flow

Stability Analysis of the Muscl Method on General Unstructured Grids for Applications to Compressible Fluid Flow Stability Analysis of the Muscl Method on General Unstructured Grids for Applications to Compressible Fluid Flow F. Haider 1, B. Courbet 1, J.P. Croisille 2 1 Département de Simulation Numérique des Ecoulements

More information

Computational Fluid Dynamics for Engineers

Computational Fluid Dynamics for Engineers Tuncer Cebeci Jian P. Shao Fassi Kafyeke Eric Laurendeau Computational Fluid Dynamics for Engineers From Panel to Navier-Stokes Methods with Computer Programs With 152 Figures, 19 Tables, 84 Problems and

More information

Sparse Matrices Reordering using Evolutionary Algorithms: A Seeded Approach

Sparse Matrices Reordering using Evolutionary Algorithms: A Seeded Approach 1 Sparse Matrices Reordering using Evolutionary Algorithms: A Seeded Approach David Greiner, Gustavo Montero, Gabriel Winter Institute of Intelligent Systems and Numerical Applications in Engineering (IUSIANI)

More information

Numerical Simulation of Coupled Fluid-Solid Systems by Fictitious Boundary and Grid Deformation Methods

Numerical Simulation of Coupled Fluid-Solid Systems by Fictitious Boundary and Grid Deformation Methods Numerical Simulation of Coupled Fluid-Solid Systems by Fictitious Boundary and Grid Deformation Methods Decheng Wan 1 and Stefan Turek 2 Institute of Applied Mathematics LS III, University of Dortmund,

More information

Nonsymmetric Problems. Abstract. The eect of a threshold variant TPABLO of the permutation

Nonsymmetric Problems. Abstract. The eect of a threshold variant TPABLO of the permutation Threshold Ordering for Preconditioning Nonsymmetric Problems Michele Benzi 1, Hwajeong Choi 2, Daniel B. Szyld 2? 1 CERFACS, 42 Ave. G. Coriolis, 31057 Toulouse Cedex, France (benzi@cerfacs.fr) 2 Department

More information

Case C1.3: Flow Over the NACA 0012 Airfoil: Subsonic Inviscid, Transonic Inviscid, and Subsonic Laminar Flows

Case C1.3: Flow Over the NACA 0012 Airfoil: Subsonic Inviscid, Transonic Inviscid, and Subsonic Laminar Flows Case C1.3: Flow Over the NACA 0012 Airfoil: Subsonic Inviscid, Transonic Inviscid, and Subsonic Laminar Flows Masayuki Yano and David L. Darmofal Aerospace Computational Design Laboratory, Massachusetts

More information

Gradient Free Design of Microfluidic Structures on a GPU Cluster

Gradient Free Design of Microfluidic Structures on a GPU Cluster Gradient Free Design of Microfluidic Structures on a GPU Cluster Austen Duffy - Florida State University SIAM Conference on Computational Science and Engineering March 2, 2011 Acknowledgements This work

More information

Debojyoti Ghosh. Adviser: Dr. James Baeder Alfred Gessow Rotorcraft Center Department of Aerospace Engineering

Debojyoti Ghosh. Adviser: Dr. James Baeder Alfred Gessow Rotorcraft Center Department of Aerospace Engineering Debojyoti Ghosh Adviser: Dr. James Baeder Alfred Gessow Rotorcraft Center Department of Aerospace Engineering To study the Dynamic Stalling of rotor blade cross-sections Unsteady Aerodynamics: Time varying

More information

ITU/FAA Faculty of Aeronautics and Astronautics

ITU/FAA Faculty of Aeronautics and Astronautics S. Banu YILMAZ, Mehmet SAHIN, M. Fevzi UNAL, Istanbul Technical University, 34469, Maslak/Istanbul, TURKEY 65th Annual Meeting of the APS Division of Fluid Dynamics November 18-20, 2012, San Diego, CA

More information

Microwell Mixing with Surface Tension

Microwell Mixing with Surface Tension Microwell Mixing with Surface Tension Nick Cox Supervised by Professor Bruce Finlayson University of Washington Department of Chemical Engineering June 6, 2007 Abstract For many applications in the pharmaceutical

More information

1 Past Research and Achievements

1 Past Research and Achievements Parallel Mesh Generation and Adaptation using MAdLib T. K. Sheel MEMA, Universite Catholique de Louvain Batiment Euler, Louvain-La-Neuve, BELGIUM Email: tarun.sheel@uclouvain.be 1 Past Research and Achievements

More information

Multigrid Solvers in CFD. David Emerson. Scientific Computing Department STFC Daresbury Laboratory Daresbury, Warrington, WA4 4AD, UK

Multigrid Solvers in CFD. David Emerson. Scientific Computing Department STFC Daresbury Laboratory Daresbury, Warrington, WA4 4AD, UK Multigrid Solvers in CFD David Emerson Scientific Computing Department STFC Daresbury Laboratory Daresbury, Warrington, WA4 4AD, UK david.emerson@stfc.ac.uk 1 Outline Multigrid: general comments Incompressible

More information

Introduction to Aerodynamic Shape Optimization

Introduction to Aerodynamic Shape Optimization Introduction to Aerodynamic Shape Optimization 1. Aircraft Process 2. Aircraft Methods a. Inverse Surface Methods b. Inverse Field Methods c. Numerical Optimization Methods Aircraft Process Conceptual

More information

Eulerian Techniques for Fluid-Structure Interactions - Part II: Applications

Eulerian Techniques for Fluid-Structure Interactions - Part II: Applications Published in Lecture Notes in Computational Science and Engineering Vol. 103, Proceedings of ENUMATH 2013, pp. 755-762, Springer, 2014 Eulerian Techniques for Fluid-Structure Interactions - Part II: Applications

More information

ALE Seamless Immersed Boundary Method with Overset Grid System for Multiple Moving Objects

ALE Seamless Immersed Boundary Method with Overset Grid System for Multiple Moving Objects Tenth International Conference on Computational Fluid Dynamics (ICCFD10), Barcelona,Spain, July 9-13, 2018 ICCFD10-047 ALE Seamless Immersed Boundary Method with Overset Grid System for Multiple Moving

More information

Accurate and Efficient Turbomachinery Simulation. Chad Custer, PhD Turbomachinery Technical Specialist

Accurate and Efficient Turbomachinery Simulation. Chad Custer, PhD Turbomachinery Technical Specialist Accurate and Efficient Turbomachinery Simulation Chad Custer, PhD Turbomachinery Technical Specialist Outline Turbomachinery simulation advantages Axial fan optimization Description of design objectives

More information

Contents. I The Basic Framework for Stationary Problems 1

Contents. I The Basic Framework for Stationary Problems 1 page v Preface xiii I The Basic Framework for Stationary Problems 1 1 Some model PDEs 3 1.1 Laplace s equation; elliptic BVPs... 3 1.1.1 Physical experiments modeled by Laplace s equation... 5 1.2 Other

More information

The WENO Method in the Context of Earlier Methods To approximate, in a physically correct way, [3] the solution to a conservation law of the form u t

The WENO Method in the Context of Earlier Methods To approximate, in a physically correct way, [3] the solution to a conservation law of the form u t An implicit WENO scheme for steady-state computation of scalar hyperbolic equations Sigal Gottlieb Mathematics Department University of Massachusetts at Dartmouth 85 Old Westport Road North Dartmouth,

More information

ENERGY-224 Reservoir Simulation Project Report. Ala Alzayer

ENERGY-224 Reservoir Simulation Project Report. Ala Alzayer ENERGY-224 Reservoir Simulation Project Report Ala Alzayer Autumn Quarter December 3, 2014 Contents 1 Objective 2 2 Governing Equations 2 3 Methodolgy 3 3.1 BlockMesh.........................................

More information

FOR P3: A monolithic multigrid FEM solver for fluid structure interaction

FOR P3: A monolithic multigrid FEM solver for fluid structure interaction FOR 493 - P3: A monolithic multigrid FEM solver for fluid structure interaction Stefan Turek 1 Jaroslav Hron 1,2 Hilmar Wobker 1 Mudassar Razzaq 1 1 Institute of Applied Mathematics, TU Dortmund, Germany

More information

A THREE-DIMENSIONAL ADAPTIVE WAVELET METHOD FOR FLUID STRUCTURE INTERACTION

A THREE-DIMENSIONAL ADAPTIVE WAVELET METHOD FOR FLUID STRUCTURE INTERACTION A THREE-DIMENSIONAL ADAPTIVE WAVELET METHOD FOR FLUID STRUCTURE INTERACTION N.K.-R. Kevlahan 1, O.V. Vasilyev 2, D. Goldstein 2, and A. Jay 1,3 kevlahan@mcmaster.ca 1 Department of Mathematics & Statistics,

More information

Example 24 Spring-back

Example 24 Spring-back Example 24 Spring-back Summary The spring-back simulation of sheet metal bent into a hat-shape is studied. The problem is one of the famous tests from the Numisheet 93. As spring-back is generally a quasi-static

More information

This is an author-deposited version published in: Eprints ID: 4362

This is an author-deposited version published in:   Eprints ID: 4362 This is an author-deposited version published in: http://oatao.univ-toulouse.fr/ Eprints ID: 4362 To cite this document: CHIKHAOUI Oussama, GRESSIER Jérémie, GRONDIN Gilles. Assessment of the Spectral

More information

course outline basic principles of numerical analysis, intro FEM

course outline basic principles of numerical analysis, intro FEM idealization, equilibrium, solutions, interpretation of results types of numerical engineering problems continuous vs discrete systems direct stiffness approach differential & variational formulation introduction

More information

2.7 Cloth Animation. Jacobs University Visualization and Computer Graphics Lab : Advanced Graphics - Chapter 2 123

2.7 Cloth Animation. Jacobs University Visualization and Computer Graphics Lab : Advanced Graphics - Chapter 2 123 2.7 Cloth Animation 320491: Advanced Graphics - Chapter 2 123 Example: Cloth draping Image Michael Kass 320491: Advanced Graphics - Chapter 2 124 Cloth using mass-spring model Network of masses and springs

More information

The Level Set Method applied to Structural Topology Optimization

The Level Set Method applied to Structural Topology Optimization The Level Set Method applied to Structural Topology Optimization Dr Peter Dunning 22-Jan-2013 Structural Optimization Sizing Optimization Shape Optimization Increasing: No. design variables Opportunity

More information

Solving Sparse Linear Systems. Forward and backward substitution for solving lower or upper triangular systems

Solving Sparse Linear Systems. Forward and backward substitution for solving lower or upper triangular systems AMSC 6 /CMSC 76 Advanced Linear Numerical Analysis Fall 7 Direct Solution of Sparse Linear Systems and Eigenproblems Dianne P. O Leary c 7 Solving Sparse Linear Systems Assumed background: Gauss elimination

More information

Final drive lubrication modeling

Final drive lubrication modeling Final drive lubrication modeling E. Avdeev a,b 1, V. Ovchinnikov b a Samara University, b Laduga Automotive Engineering Abstract. In this paper we describe the method, which is the composition of finite

More information

1 Exercise: Heat equation in 2-D with FE

1 Exercise: Heat equation in 2-D with FE 1 Exercise: Heat equation in 2-D with FE Reading Hughes (2000, sec. 2.3-2.6 Dabrowski et al. (2008, sec. 1-3, 4.1.1, 4.1.3, 4.2.1 This FE exercise and most of the following ones are based on the MILAMIN

More information

INTERIOR POINT METHOD BASED CONTACT ALGORITHM FOR STRUCTURAL ANALYSIS OF ELECTRONIC DEVICE MODELS

INTERIOR POINT METHOD BASED CONTACT ALGORITHM FOR STRUCTURAL ANALYSIS OF ELECTRONIC DEVICE MODELS 11th World Congress on Computational Mechanics (WCCM XI) 5th European Conference on Computational Mechanics (ECCM V) 6th European Conference on Computational Fluid Dynamics (ECFD VI) E. Oñate, J. Oliver

More information

Curved Mesh Generation and Mesh Refinement using Lagrangian Solid Mechanics

Curved Mesh Generation and Mesh Refinement using Lagrangian Solid Mechanics Curved Mesh Generation and Mesh Refinement using Lagrangian Solid Mechanics Per-Olof Persson University of California, Berkeley, Berkeley, CA 9472-384, U.S.A. Jaime Peraire Massachusetts Institute of Technology,

More information

Finite Volume Discretization on Irregular Voronoi Grids

Finite Volume Discretization on Irregular Voronoi Grids Finite Volume Discretization on Irregular Voronoi Grids C.Huettig 1, W. Moore 1 1 Hampton University / National Institute of Aerospace Folie 1 The earth and its terrestrial neighbors NASA Colin Rose, Dorling

More information

Parallel Interpolation in FSI Problems Using Radial Basis Functions and Problem Size Reduction

Parallel Interpolation in FSI Problems Using Radial Basis Functions and Problem Size Reduction Parallel Interpolation in FSI Problems Using Radial Basis Functions and Problem Size Reduction Sergey Kopysov, Igor Kuzmin, Alexander Novikov, Nikita Nedozhogin, and Leonid Tonkov Institute of Mechanics,

More information

A COUPLED FINITE VOLUME SOLVER FOR THE SOLUTION OF LAMINAR TURBULENT INCOMPRESSIBLE AND COMPRESSIBLE FLOWS

A COUPLED FINITE VOLUME SOLVER FOR THE SOLUTION OF LAMINAR TURBULENT INCOMPRESSIBLE AND COMPRESSIBLE FLOWS A COUPLED FINITE VOLUME SOLVER FOR THE SOLUTION OF LAMINAR TURBULENT INCOMPRESSIBLE AND COMPRESSIBLE FLOWS L. Mangani Maschinentechnik CC Fluidmechanik und Hydromaschinen Hochschule Luzern Technik& Architektur

More information

Rotating Moving Boundary Analysis Using ANSYS 5.7

Rotating Moving Boundary Analysis Using ANSYS 5.7 Abstract Rotating Moving Boundary Analysis Using ANSYS 5.7 Qin Yin Fan CYBERNET SYSTEMS CO., LTD. Rich Lange ANSYS Inc. As subroutines in commercial software, APDL (ANSYS Parametric Design Language) provides

More information