Documents published in Scipedia

• Physics-informed neural network vs finite element method for modeling coupled water and solute flow in unsaturated soils

  H. Kamil, A. Soulaïmani, A. Beljadid

  WCCM2024.

  
  

  Abstract

  Accurate modeling of water infiltration and solute transport in unsaturated soils is critical for various applications. These include optimizing irrigation practices to conserve water and minimize environmental impact, as well as predicting the fate of contaminants in soil and groundwater. This study explores the application of the vanilla physics informed neural network (PINN) approach for modeling the coupled system of water flow and solute transport in unsaturated soils. We compare the performance of PINN with the Galerkin finite element method (FEM) to evaluate their effectiveness. Various techniques are implemented to improve the PINN solver, including adaptive activation functions. Numerical tests were carried out to evaluate the efficiency of the PINN solver in comparison to the FEM. The findings reveal that PINN can achieve accuracy comparable to FEM, albeit at a significantly higher computational cost during training, while maintaining fast inference times.

  Abstract
  Accurate modeling of water infiltration and solute transport in unsaturated soils is critical for various applications. These include optimizing irrigation practices to conserve [...]

  • 0
  •  read

• A predictor-corrector second-order time-stepping schemes for solving water flow and solute transport in unsaturated porous media

  N. Toutlini, H. Kamil, A. Soulaïmani, A. Beljadid

  WCCM2024.

  
  

  Abstract

  The objective of this study is to numerically solve the coupled system of water flow and solute transport in unsaturated porous media using a noniterative predictor-corrector temporal scheme for the Richards equation and a semi-implicit temporal scheme for the advection dispersion equation (ADE). The standard and non-standard Galerkin finite element methods are used for spatial discretization. Three different techniques are proposed to calculate the pressure head in the Levrett equation. These techniques are different in terms of the chosen shape functions in the finite element space. The proposed schemes offer distinct advantages due to the linear nature of the resulting system, facilitating easy implementation and avoiding the issues associated with the divergence of iterative schemes. We evaluated the robustness and efficacy of the suggested methods using a computational experiment to simulate soil salinity and water flow in loamy soil. We compared it with data found in the literature. The results provide compelling evidence confirming the proposed methods’ effectiveness and stability.

  Abstract
  The objective of this study is to numerically solve the coupled system of water flow and solute transport in unsaturated porous media using a noniterative predictor-corrector [...]

  • 0
  •  read

• Incompressible viscous fluid analysis around complex shapes using Isogeometric Analysis

  Y. Sakai, K. Kashiyama, H. Hasebe

  WCCM2024.

  
  

  Abstract

  This paper aims to apply the Isogeometric Analysis(IGA) to fluid-structure interaction problem in the civil engineering field. Recently, IGA has attracted much attention as an analysis method to structure with arbitrary surfaces. In this paper, IGA is applied to a twodimensional incompressible viscous flow problem as a basic study for the fluid-structure interaction analysis using IGA. The vortex induced vibration of a circular cylinder is investigated as a numerical example, and the effectiveness and validity of the coupled analysis using IGA are discussed.

  Abstract
  This paper aims to apply the Isogeometric Analysis(IGA) to fluid-structure interaction problem in the civil engineering field. Recently, IGA has attracted much attention as [...]

  • 0
  •  read

• Development of traffic noise evaluation system using finite element method

  H. Miyauchi, K. Kashiyama, H. Yoshikawa

  WCCM2024.

  
  

  Abstract

  This paper presents a traffic noise evaluation system based on acoustic theory. The finite element method is employed for unsteady wave equations, which is suitable for arbitrary shapes and has excellent applicability to non-uniform materials. The 3D wave equation is employed for the governing equation and the Perfectly Matched Layer (PML) method is utilized as a treatment method for boundary condition. In order to consider multiple moving sound sources such as a traffic noise, a time-variant convolution method is introduced. The auralization method based on VR technology is also introduced to understand the noise level intuitively

  Abstract
  This paper presents a traffic noise evaluation system based on acoustic theory. The finite element method is employed for unsteady wave equations, which is suitable for arbitrary [...]

  • 0
  •  read

• Effect of long-term sea water exposure on dielectric materials

  E. Stanisauskis Weiss, E. Warner, V. Rossell-Olmos

  WCCM2024.

  
  

  Abstract

  Dielectric materials, which are commonly used in capacitors, could increase energy storage density on a per volume basis in film capacitors compared to current technologies accommodating ever-increasing power demands. Recent work in this area has brought about dramatic increases in the dielectric permittivity and moderate increases in dielectric loss, leading to increased material performance on a per volume basis. However, little is known about the aging and breakdown of these materials, which could decrease the performance of these films over time due to decaying dielectric loss and energy storage density. A basic study of the aging of two different state-of-the-art dielectric materials, 3M's Very High Bond (VHB) 4910, commonly used in actuator applications, and bi-axially oriented polypropylene (BOPP), commonly used in large wound film capacitators, is completed. Accelerated life tests using distilled water are conducted to simulate the aging of these materials in a marine environment. An acceleration factor is determined by diffusion studies of distilled water into the materials. Aminabhavi’s and Crank’s methods are used and compared to compute the diffusion coefficient. The two methods produce identical activation energies and, in turn, acceleration factors. The success of this work could actively exhibit the promise of these materials in microelectronic uses.

  Abstract
  Dielectric materials, which are commonly used in capacitors, could increase energy storage density on a per volume basis in film capacitors compared to current technologies [...]

  • 0
  •  read

• Coupling multiple resonances for enhancing sound transmission loss of acoustic metamaterials

  D. Roca, G. Sal-Anglada, D. Yago, J. Cante

  WCCM2024.

  
  

  Abstract

  Recent developments in acoustic metamaterials have been focused on broadening the attenuating bandwidth features towards lower frequency ranges, well below 1000 Hz, as well as tackling manufacturing issues. In this context, a multi-resonant layered acoustic metamaterial (MLAM) was proposed as a practical realization for addressing both challenges. The MLAM’s layered-based design makes it amenable to large-scale manufacturing and the periodic features of each layer enable the application of computational homogenization models to characterize the sound transmission loss (STL) response. Combining these models with optimization techniques allows to determine realistic MLAM designs that trigger multiple resonances in broad frequency ranges. By exploiting coupling mechanisms these resonances translate into multiple STL peaks that produce a broadband continuous frequency range of attenuation, i.e., without transmission peaks in-between. In this work, the proposed computational homogenization model is presented and applied to the design of different MLAM configurations. The goal is to assess the influence of the number of coupled resonating layers in the STL response of the whole MLAM panel, in terms of increasing the attenuation intensity and the effective frequency bandwidth. The results demonstrate the STL enhancements features obtained from exploiting coupling mechanisms, compared to other acoustic metamaterial configurations based on local resonance phenomena. In this context, the proposed MLAM technology exhibits a great potential to provide an efficient, easy-to-manufacture solution to the sound insulation problem at low frequency ranges

  Abstract
  Recent developments in acoustic metamaterials have been focused on broadening the attenuating bandwidth features towards lower frequency ranges, well below 1000 Hz, as well [...]

  • 0
  •  read

• Application of the Harmonic Balance Method to predict wave propagation in one-dimensional nonlinear metamaterial excited harmonically

  R. Moura Barbosa, A. Serpa

  WCCM2024.

  
  

  Abstract

  This work exposes a computational procedure designed to aid in modeling mechanical systems featuring stiffness nonlinearity. The basis of the procedure is the Harmonic Balance Method, which is combined with a numerical continuation technique. To present the efficacy of the approach, a one-dimensional nonlinear metamaterial is analyzed. The aim is to demonstrate the suitability of the procedure to extract information regarding higher harmonic generation and the influence of the amplitude of excitation on the system dynamic response.

  Abstract
  This work exposes a computational procedure designed to aid in modeling mechanical systems featuring stiffness nonlinearity. The basis of the procedure is the Harmonic Balance [...]

  • 0
  •  read

• Ritter-Križaić iteracion method of truss constructions

  V. Križaić, T. Rodiger, N. Križaić, J. Križaić

  WCCM2024.

  
  

  Abstract

  The optimization of the dimensioning of constructive designs is constantly evolving. FEM, evolutionary, and other various methods are being developed, which are implemented with algorithms in computer simulations of building models. The problem with these methods is solving large differential equations, which is inconceivable without computers and large memories. The Ritter-Križaić (RK) iteration method works for both straight and oblique networks with one side, and it can even be used instead of trigonometric and FEM equations. It does this by adding the geometric properties of the networks and outside actions to the directional equations. By creating straightforward monograms of RK-FEM technology with straightforward differential or subspace equations that are simple to calculate by hand or draw with Mathcad tools, the RK-FEM loop enables COD to define various types of trusses and even other supports. RK-FEM COD is therefore used to create simulation games that explain many logical phenomena in the design of external and internal actions of beam supports, which can be compared to a spider thread or an ice plate structure as an RK string and even to the moon

  Abstract
  The optimization of the dimensioning of constructive designs is constantly evolving. FEM, evolutionary, and other various methods are being developed, which are implemented [...]

  • 0
  •  read

• Influence evaluation of flow diverter stent parent vessel coverage on cerebral aneurysm through the CFD-DEM coupling simulation

  Y. Ohkura, D. Watanabe, R. Taniguchi, S. Yamani

  WCCM2024.

  
  

  Abstract

  Cerebral aneurysms are a type of cerebrovascular disorder where a balloon-like bulge forms in part of an artery in the brain. One of the developed treatments for large cerebral aneurysms is the Flow-diverter Stent (FDS) placement technique. Effective treatment outcomes in cerebral aneurysm treatment using FDS require proper placement of the device. Improper placement can lead to increased blood flow velocity and Wall Shear Stress (WSS) within the aneurysm, as well as increased pressure, which suggests a potential risk of rupture in large aneurysms. Considering these circumstances, this study evaluates the impact of FDS positioning on cerebral aneurysms by creating multiple FDS placement models with the device positioned proximally and distally to the aneurysm. Subsequently, we conducted fluid-structure interaction simulation analyses using the Particle Finite Element Method-Second Generation (PFEM-2) for the non-Newtonian fluid model of blood and the Discrete Element Method (DEM) for the FDS. This study reports the results of comparing blood behavior, WSS, and pressure inside the cerebral aneurysm based on the FDS placement position.

  Abstract
  Cerebral aneurysms are a type of cerebrovascular disorder where a balloon-like bulge forms in part of an artery in the brain. One of the developed treatments for large cerebral [...]

  • 0
  •  read

• Delta-P1 model implementation for numerical simulation of photothermal cancer therapy in three-dimensional heterogeneous tissues

  R. Gomez Araque, C. Bustamante, R. Valencia, W. Florez

  WCCM2024.

  
  

  Abstract

  Photothermal therapy (PTT) stands as a promising avenue for cancer treatment. Metallic nanoparticles (NPs) absorb near-infrared light, inducing localized heating for tumor cell apoptosis. Predicting spatial temperature information in preclinical models is crucial due to cell death sensitivity to temperature changes. Heat transfer models, rely on the radiative transport equation (RTE), where its approximation is essential for this purpose. Existing models for the radiative transport equation, such as the Beer-Lambert law, the diffusion approximation, the discrete ordinates method, and Monte Carlo (MC) simulations, are widely used in the context of PTT. However, each of them has limitations. This study focuses on the δP1 model, wich is an extension of the diffusion approximation. Unlike standard diffusion approximation (SDA), the δP1 model treats forward and scattered light independently, preserving accuracy over a wider range of optical properties, including media with plasmonic NPs. The δP1 model equations are discretized and solved by the Finite Element Method (FEM) . Its numerical results for fluence rate in a heterogeneous geometry with nanoshells is compared to MC simulations and the standard diffusion approximation. This study validates and applies the model to the simulation of light transport in photothermal therapy in general two-dimensional geometries. Results demonstrate the δP1 shows a significant improvement over the SDA in heat transfer simulations in heterogeneous tissues geometries. This underscores its potential as a valuable tool for optimizing photothermal therapy preclinical models.

  Abstract
  Photothermal therapy (PTT) stands as a promising avenue for cancer treatment. Metallic nanoparticles (NPs) absorb near-infrared light, inducing localized heating for tumor [...]

  • 0
  •  read