COLLECTIONS

Documents published in Scipedia

• Smoothed surface gradients for panel methods

  J. D'Elía, M. Storti, S. Idelsohn

  Advances in Engineering Software (2000). Vol. 31 (5), pp. 339-346

  
  

  Abstract

  A weak form to compute the dipolar and monopolar surface gradients, related to a low-order panel method, is shown. The flow problem is formulated by means of a three-dimensional potential model and the discretization is based on Morino's formulation for the perturbation velocity potential. On the body surface, this representation reduces to a boundary integral equation with the source (or monopolar) and the doublet (or dipolar) densities. The first of the two is found by application of the boundary flow condition, and the second one is the unknown over the body surface. A lower panel method is used for the analytic integrations of both the monopolar and dipolar influence coefficients. The surface velocity field is computed after solving the linear system, with a strong and a weak form of the Stokes theorem, which is oriented to fairly non-structured panel meshes. The proposed method is validated by comparing the numerical results with analytical ones for an isolated sphere and includes a prediction over a car-like configuration.

  Abstract
  A weak form to compute the dipolar and monopolar surface gradients, related to a low-order panel method, is shown. The flow problem is formulated by means of a three-dimensional [...]

  • 12
  •  read

• A closed form for low-order panel methods

  J. D'Elía, M. Storti, S. Idelsohn

  Advances in Engineering Software (2000). Vol. 31 (5), pp. 347-353

  
  

  Abstract

  A closed form for the computation of the dipolar and monopolar influence coefficients related to a low-order panel method is shown. The flow problem is formulated by means of a three-dimensional potential model; the method of discretization is based on the Morino formulation for the perturbation velocity potential. On the body surface this representation reduces to an integral equation with the source (or monopolar) and the doublet (or dipolar) densities. The former is found by application of the boundary condition, and the latter is the unknown over the surface of the body. The lower panel method is used for the analytical integrations of the monopolar and dipolar influence coefficients, with special attention to avoid a logarithmic singularity in the monopolar matrix when flat fairly structured meshes that are common in ship-wave calculations are used.

  Abstract
  A closed form for the computation of the dipolar and monopolar influence coefficients related to a low-order panel method is shown. The flow problem is formulated by means [...]

  • 0
  •  read

• Evaluation of Finite Element Calculations in a Cracked Cylinder Under Internal Pressure by Speckle Photography

  G. Kaufmann, A. Lopergolo, S. Idelsohn

  J. Appl. Mech (1983). Vol. 50 (4a), pp. 896-897

  
  

  Abstract

  The usefulness of using the speckle photography technique in fracture mechanics to check numerical calculations is demonstrated for an internally pressurized cylinder with a surface flaw. A pointwise technique was utilized to measure the opening displacements along the crack line and the Mode I stress-intensity factor was determined by extrapolating these results to the crack tip. Finite element calculations were performed to be compared with experimental data and good agreement was obtained.

  Abstract
  The usefulness of using the speckle photography technique in fracture mechanics to check numerical calculations is demonstrated for an internally pressurized cylinder with [...]

  • 35
  •  read

• Transverse vibrations of an isotropic, simply supported rectangular plate with an orthotropic inclusion

  L. Ercoli, V. Sonzogni, S. Idelsohn, P. Laura

  Journal of Sound and Vibration (1992). Vol. 135 (2), pp. 217-221

  
  

  Abstract

  An approximate analytical solution is obtained for the title problem by using a Navier-type solution. It is shown that the fundamental frequency coefficient obtained by means of this procedure is in reasonably good agreement with the result obtained by using a finite element algorithmic methodology. Obviously the type of material disturbance considered introduces considerable change in the dynamic behavior of the structural element and this fact is clearly shown in the variation of the fundamental frequency value.

  Abstract
  An approximate analytical solution is obtained for the title problem by using a Navier-type solution. It is shown that the fundamental frequency coefficient obtained by means [...]

  • 13
  •  read

• Free vibrations of rectangular plates of exponentially varying thickness and with a free edge

  V. Sonzogni, S. Idelsohn, P. Laura, V. Cortínez

  Journal of Sound and Vibration (1990). Vol. 140 (3), pp. 513-522

  
  

  Abstract

  Transverse vibrations of the structural system described in the title are analyzed by using classical plate theory and employing two different methodologies: the recently developed “optimized Kantorovich approach”; and a finite element algorithmic procedure. It is shown that the fundamental frequency coefficients obtained by using both techniques for a rather wide variation of the governing geometric parameters are in very good agreement. Lower frequency coefficients of the plate system determined by means of the finite element method are also presented.

  Abstract
  Transverse vibrations of the structural system described in the title are analyzed by using classical plate theory and employing two different methodologies: the recently [...]

  • 11
  •  read

• On the self-stressing modes in free vibration analysis

  S. Idelsohn, M. Geradin

  Journal of Sound and Vibration (1982). Vol. 83 (2), pp. 143-155

  
  

  Abstract

  The type of convergence to the eigenspectrum of a structure calculated from a finite element analysis is examined in light of the variational properties of the Rayleigh quotient and of Courant's maximum-minimum principle. The influence of the self-stressing modes on the solution is firt demonstrated theoretically and then shown practically on the results obtained via a family of equilibrium elements in which the number of self-stressing modes relative to the number of vibration modes may be varied as a parameter.

  Abstract
  The type of convergence to the eigenspectrum of a structure calculated from a finite element analysis is examined in light of the variational properties of the Rayleigh quotient [...]

  • 17
  •  read

• Fast fluid–structure interaction simulations using a displacement-based finite element model equipped with an explicit streamline integration prediction

  P. Ryzhakov, J. Miquel, S. Idelsohn, E. Oñate

  Comput. Methods Appl. Mech. Engrg., (2017). Vol. 315, pp. 1080-1097

  
  

  Abstract

  We propose here a displacement-based updated Lagrangian fluid model developed to facilitate a monolithic coupling with a wide range of structural elements described in terms of displacements. The novelty of the model consists in the use of the explicit streamline integration for predicting the end-of-step configuration of the fluid domain. It is shown that this prediction considerably alleviates the time step size restrictions faced by the former Lagrangian models due to the possibility of an element inversion within one time step. The method is validated and compared with conventional approaches using three numerical examples. Time step size and corresponding Courant numbers leading to optimal behavior in terms of computational efficiency are identified.

  Abstract
  We propose here a displacement-based updated Lagrangian fluid model developed to facilitate a monolithic coupling with a wide range of structural elements [...]

  • 24
  •  read

• Metal forming analysis by fourier series expansion and further uses of pseudo-concentrations

  H. Antúnez, S. Idelsohn, E. Dvorkin

  Computers and Structures (1992). Vol. 44 (1-2), pp. 435-451

  
  

  Abstract

  An analysis of metal forming processes with axial geometry is performed by an expansion in Fourier series on the circumferential direction. For that purpose, an incremental form is presented on the basis of the flow formulation. For every increment of the non-null imposed boundary conditions an algebraic system is formed in which the non-linearity due to the visco-plastic properties of the material is placed as part of the load term. In this way, the global stiffness matrix is block diagonal, the coupling due to material non-linearity being included on the right-hand side. Free surfaces are treated as an interface inside the elements, as an extension of the pseudo-concentration method.

  Abstract
  An analysis of metal forming processes with axial geometry is performed by an expansion in Fourier series on the circumferential direction. For that purpose, an incremental [...]

  • 19
  •  read

• An effective automatic incremental/iterative method for static nonlinear structural analysis

  J. Jeusette, G. Laschet, S. Idelsohn

  Computers and Structures (1989). Vol. 32 (1), pp. 125-135

  
  

  Abstract

  In the prebuckling range of the complete nonlinear response of a structure submitted to destabilizing loads, a linearized stability analysis is generally an interesting tool. Used in conjunction with an arc-length algorithm, incremental bifurcation analyses are able to give optimal definition for the increment predictor size. When the associated eigenvalue problem is solved by minimizing the finite element discretized form of the Rayleigh quotient via a preconditioned conjugate gradient technique, it can be very easily inserted in an automatic nonlinear equations solver. Such a procedure is presented in this paper after a review of existing arc-length schemes and step-size control methods. Moreover, for tracing the postbuckling range of the nonlinear response automatically, a new recurrence formula is proposed to control the step-size. The well behaviour of this algorithm is illustrated in two classical structural applications. 

  Abstract
  In the prebuckling range of the complete nonlinear response of a structure submitted to destabilizing loads, a linearized stability analysis is generally an interesting tool. [...]

  • 18
  •  read

• A load-dependent basis for reduced nonlinear structural dynamics

  S. Idelsohn, A. Cardona

  Computers and Structures (1985). Vol. 20 (1-3), pp. 203-210

  
  

  Abstract

  A computational algorithm for predicting the dynamical response of a nonlinear structure by means of a reduction scheme is described. In it, the nonlinear system of ordinary differential equations obtained from the finite element discretization is reduced, by employing a Rayleigh-Ritz technique. A new criterion for the computation of the basis vectors is proposed. It is based on a sequence of basis vectors proposed by Wilson et al. for linear dynamics problems, which was augmented by adding derivatives of these vectors with respect to the generalized displacement amplitudes. The later vectors allow the treatment of nonlinear problems.

  Abstract
  A computational algorithm for predicting the dynamical response of a nonlinear structure by means of a reduction scheme is described. In it, the nonlinear system of ordinary [...]

  • 21
  •  read