Renato Pavanello

Bi-directional evolutionary structural optimization for design-dependent fluid pressure loading problems

This article presents an evolutionary topology optimization method for compliance minimization of structures under design-dependent pressure loads. In traditional density based topology optimization methods, intermediate values of densities for the solid elements arise along the iterations. Extra boundary parametrization schemes are demanded when these methods are applied to pressure loading problems. An alternative methodology is suggested in this article for handling this type of load. With an extended bi-directional evolutionary structural optimization method associated with a partially coupled fluid–structure formulation, pressure loads are modelled with hydrostatic fluid finite elements. Due to the discrete nature of the method, the problem is solved without any need of pressure load surfaces parametrization. Furthermore, the introduction of a separate fluid domain allows the algorithm to model non-constant pressure fields with Laplaces equation. Three benchmark examples are explored in order to show the achievements of the proposed method.

Synthesis of porous-acoustic absorbing systems by an evolutionary optimization method

Topology optimization is frequently used to design structures and acoustic systems in a large range of engineering applications. In this work, a method is proposed for maximizing the absorbing performance of acoustic panels by using a coupled finite element model and evolutionary strategies. The goal is to find the best distribution of porous material for sound absorbing panels. The absorbing performance of the porous material samples in a Kundt tube is simulated using a coupled porous–acoustic finite element model. The equivalent fluid model is used to represent the foam material. The porous material model is coupled to a wave guide using a modal superposition technique. A sensitivity number indicating the optimum locations for porous material to be removed is derived and used in a numerical hard kill scheme. The sensitivity number is used to form an evolutionary porous material optimization algorithm which is verified through examples.

Topology optimization for submerged buoyant structures

ABSTRACT This paper presents an evolutionary structural topology optimization method for the design of completely submerged buoyant modules with design-dependent fluid pressure loading. This type of structure is used to support offshore rig installation and pipeline transportation at all water depths. The proposed optimization method seeks to identify the buoy design that has the highest stiffness, allowing it to withstand deepwater pressure, uses the least material and has a minimum prescribed buoyancy. Laplaces equation is used to simulate underwater fluid pressure, and a polymer buoyancy module is considered to be linearly elastic. Both domains are solved with the finite element method. Using an extended bi-directional evolutionary structural optimization (BESO) method, the design-dependent pressure loads are modelled in a straightforward manner without any need for pressure surface parametrization. A new buoyancy inequality constraint sets a minimum required buoyancy effect, measured by the joint volume of the structure and its interior voids. Solid elements with low strain energy are iteratively removed from the initial design domain until a certain prescribed volume fraction. A test case is described to validate the optimization problem, and a buoy design problem is used to explore the features of the proposed method.

Numerical methods for the dynamics of unbounded domains

The present article discusses the relation between boundary conditions and the Sommerfeld radiation condition underlying the dynamics of unbounded domains. It is shown that the classical Dirichlet, Neumann and mixed boundary conditions do not fulfill the radiation condition. In the sequence, three strategies to incorporate the radiation condition in numerical methods are outlined. The inclusion of Infinite Elements in the realm of the Finite Element Method (FEM), the Dirichlet-to-Neumann (DtN) mapping and the Boundary Element Method (BEM) are described. Examples of solved dynamic problems in unbounded domains are given for the Helmholtz and the Navier operators. The advantages and limitations of the methodologies are discussed and pertinent literature is provided.

Dynamic Stationary Response of Reinforced Plates by the Boundary Element Method

A direct version of the boundary element method (BEM) is developed to model the stationary dynamic response of reinforced plate structures, such as reinforced panels in buildings, automobiles, and airplanes. The dynamic stationary fundamental solutions of thin plates and plane stress state are used to transform the governing partial differential equations into boundary integral equations (BIEs). Two sets of uncoupled BIEs are formulated, respectively, for the in-plane state (membrane) and for the out-of-plane state (bending). These uncoupled systems are joined to form a macro-element, in which membrane and bending effects are present. The association of these macro-elements is able to simulate thin-walled structures, including reinforced plate structures. In the present formulation, the BIE is discretized by continuous and/or discontinuous linear elements. Four displacement integral equations are written for every boundary node. Modal data, that is, natural frequencies and the corresponding mode shapes of reinforced plates, are obtained from information contained in the frequency response functions (FRFs). A specific example is presented to illustrate the versatility of the proposed methodology. Different configurations of the reinforcements are used to simulate simply supported and clamped boundary conditions for the plate structures. The procedure is validated by comparison with results determined by the finite element method (FEM).

SCR-Seafloor Interaction Modeling With Winkler, Pasternak and Kerr Beam-on-Elastic-Foundation Theories

This paper discusses the use of three analytical models for the SCR-soil interaction analysis. Analytical derivations for Winkler, Pasternak and Kerr models are presented. The effects of the adopted model on displacements and stresses near the riser touchdown point are addressed. Comparisons with the results from a continuous soil model calculated with finite elements are also presented. In the regions with large displacement gradients, the Kerr model, which considers the influence of the elasticity from neighboring area, presents the best agreement with the 3D Finite Element model.© 2009 ASME

An Evolutionary Optimization Method applied to Absorbing Poroelastic Systems

Poroelastic materials can be used in engineering applications such as: noise control of automobiles, acoustical insulation systems for aircrafts, industrial, environmental and domestic sound quality control, etc. The insulating systems must be as light as possible and the acoustic absorption in the low frequency domain must be maximized for certain gaps. Topology Optimization is frequently used to design structures and acoustic systems in a large range of engineering applications. In this work, we propose one method to maximize the absorbing performance of insulation poroelastic systems using a coupled finite element model and Evolutionary strategies. The goal is find the best distribution of poroelastic material on insulating systems. The absorbing performance of the poroelastic material samples in a Kundt tube is simulated using a coupled poroelastic and acoustic finite element model. The Biot‐Allard Coupled Model is used to represent the foam material. The porous material model is coupled to a waveguid...

Dynamic behavior of frame structures by boundary integral procedures

Abstract The Boundary Element Method is applied to synthesize a set of Boundary Integral Equations representing the uncoupled axial and flexural dynamic behavior of rectilinear Bernoulli–Euler beam elements in the frequency domain. In the sequence, these structural elements are coupled by the sub-region technique to model two-dimensional frame structures, in which the axial and flexural behaviors are coupled. This methodology is used to accurately recover modal data, eigenfrequencies and eigenmodes, of two frame structures. The usual Boundary Element procedure is recast to deliver simultaneously the values of variables at the element boundaries and at an arbitrary number of internal nodes. The inclusion of internal nodes allow to recover the structure eigenmodes and makes feasible the coupling of the assembled systems with a surrounding environment, for instance, an acoustic field. The results obtained are compared with a standard Finite Element eigenvalue analysis. It is shown that for increasing response frequencies, the Boundary Element scheme delivers modal data within a degree of accuracy, which is only obtained by the conventional Finite Element Method with considerable finer meshes.

Structural shape optimization of bonded joints using the ESO method and a honeycomb-like mesh

This work presents an application of the evolutionary structural optimization (ESO) method to optimize two of the most common bonded joints – the single lap joint and the double lap joint. The ESO method was used to shape the adherends contour to reduce peak stresses at the overlap ends. The shape optimization was performed under a von Mises rejection criterion. The von Mises stress was evaluated using the finite element model for both types of joints. It is proposed here to utilize a honeycomb-like mesh using hexagonal elements to minimize stress concentration problem. Additionally, the use of a honeycomb mesh in the ESO method led to a smoother adherend contour. Numerical results show how the stress distributions for both shear and peel stresses are improved after the application of this optimization method.

Experimental Assessment of the Behaviour of a Pipe Vibration Damper Underwater

Submarine petroleum pipelines, risers and jumpers suffer static and dynamic loads due to sea currents and waves, due to the displacements of the floating production units and due to the internal flow, among other causes. Mitigating the oscillations caused by such excitations is critical to the reliability and fatigue of those underwater bodies. The Pounding Tuned Mass Damper (PTMD) is one device that may be employed to absorb and dissipate vibrations. These devices have long been used for mechanical systems operating in the atmosphere, but are new for underwater applications. This paper presents a study of the behaviour of a PTMD working underwater.A small scale laboratory apparatus was built to assess the effect of the absorber on the oscillation of a pipe submerged in a water tank. The PTMD was attached to a test pipe section mounted on an elastic suspension harness. The PTMD model is a lumped mass-spring attachment similar to a tuned mass dumper (TMD) suppressor, but with the addition of a pounding layer, which limits the motion of the PTMD mass, dissipating the energy of the oscillating pipe through the impact of the PTMD mass against that layer. Free and forced oscillation experiments were executed in air and in water, with and without the oscillation absorber, to determine the effectiveness of the PTMD. The tests were run on a range of excitation frequencies and the amplification factors were obtained for each case.The data show a remarkable influence of the surrounding media on the dynamics of the pipe-absorber system, therefore the interaction with the water must be taken into consideration in the design of the system. Although the results are only a preliminary step on the development of a device applicable to an actual petroleum submarine pipeline, it was observed that the PTMD does indeed suppress the vibrations, but it must be properly configured to achieve an optimum performance. The data gathered from this work will also be useful in the improvement of a numerical model of the pipe-PTMD system for use in a computer simulator.Copyright