• Steel and Composite Structures
• /
• v.51 no.4
• /
• pp.363-375
• /
• 2024

Within the optimization field, addressing the intricate posed by fluidic pressure loads on functionally graded structures with frequency-related designs is a kind of complex design challenges. This paper thus introduces an innovative density-based topology optimization strategy for frequency-constraint functionally graded structures incorporating Darcy's law and a drainage term. It ensures consistent treatment of design-dependent fluidic pressure loads to frequency-related structures that dynamically adjust their direction and location throughout the design evolution. The porosity of each finite element, coupled with its drainage term, is intricately linked to its density variable through a Heaviside function, ensuring a seamless transition between solid and void phases. A design-specific pressure field is established by employing Darcy's law, and the associated partial differential equation is solved using finite element analysis. Subsequently, this pressure field is utilized to ascertain consistent nodal loads, enabling an efficient evaluation of load sensitivities through the adjoint-variable method. Moreover, this novel approach incorporates load-dependent structures, frequency constraints, functionally graded material models, and polygonal meshes, expanding its applicability and flexibility to a broader range of engineering scenarios. The proposed methodology's effectiveness and robustness are demonstrated through numerical examples, including fluidic pressure-loaded frequency-constraint structures undergoing small deformations, where compliance is minimized for structures optimized within specified resource constraints.

• Geophysics and Geophysical Exploration
• /
• v.26 no.3
• /
• pp.114-125
• /
• 2023

The physics-informed neural network (PINN) has been proposed to overcome the limitations of various numerical methods used to solve partial differential equations (PDEs) and the drawbacks of purely data-driven machine learning. The PINN directly applies PDEs to the construction of the loss function, introducing physical constraints to machine learning training. This technique can also be applied to wave equation modeling. However, to solve the wave equation using the PINN, second-order differentiations with respect to input data must be performed during neural network training, and the resulting wavefields contain complex dynamical phenomena, requiring careful strategies. This tutorial elucidates the fundamental concepts of the PINN and discusses considerations for wave equation modeling using the PINN approach. These considerations include spatial coordinate normalization, the selection of activation functions, and strategies for incorporating physics loss. Our experimental results demonstrated that normalizing the spatial coordinates of the training data leads to a more accurate reflection of initial conditions in neural network training for wave equation modeling. Furthermore, the characteristics of various functions were compared to select an appropriate activation function for wavefield prediction using neural networks. These comparisons focused on their differentiation with respect to input data and their convergence properties. Finally, the results of two scenarios for incorporating physics loss into the loss function during neural network training were compared. Through numerical experiments, a curriculum-based learning strategy, applying physics loss after the initial training steps, was more effective than utilizing physics loss from the early training steps. In addition, the effectiveness of the PINN technique was confirmed by comparing these results with those of training without any use of physics loss.

• Journal of Advanced Marine Engineering and Technology
• /
• v.40 no.4
• /
• pp.264-269
• /
• 2016

This paper reports a fundamental study of temperature characteristics of a solar pond with an insulation layer. Further, these characteristics were compared with those of a solar pond without the insulation layer. The governing equation was discretized via finite difference method. The governing equations are two-dimensional unsteady-state second-order partial differential equations. The conclusions of the study are as follows: 1) If the depth of the solar pond was increased, the desired effect of increase in temperature was not produced because the amount of solar insolation received by the bottom of the solar pond decreased. 2) As the temperature of the soil during winter is higher than the temperature of the water in a solar pond, heat was transferred from the soil to the solar pond. 3) For the case of the solar pond with insulation layer, it was estimated that the dependence rate of solar energy was 83.3% and that of the boiler was 16.7%.

• Journal of the Korean Geotechnical Society
• /
• v.27 no.3
• /
• pp.75-83
• /
• 2011

Emerging issues related with fully coupled Thermo-Hydro-Mechanical (THM) behavior of unsaturated soil demand the development of a numerical tool in diverse geo-mechanical and geo-environmental areas. This paper presents general governing equations for coupled THM processes in unsaturated porous media. Coupled partial differential equations are derived from three mass balances equations (solid, water, and air), energy balance equation, and force equilibrium equation. With Galerkin formulation and time integration of these governing equations, finite element code is developed to find nonlinear solution of four main variables (displacement-u, gas pressure-$P_g$), liquid pressure-$P_1$), and temperature-T) using Newton's iterative scheme. Three cases of numerical simulations are conducted and discussed: one-dimensional drainage experiments (u-$P_g-P_1$), thermal consolidation (u-$P_1$-T), and effect of pile on surrounding soil due to surface temperature variation (u-$P_1$-T).

• Journal of the Korean Geotechnical Society
• /
• v.30 no.9
• /
• pp.67-75
• /
• 2014

Methane gas hydrate which is considered energy source for the next generation has an urgent need to develop reliable numerical simulator for coupled THM phenomena in the porous media, to minimize problems arising during the production and optimize production procedures. International collaborations to improve previous numerical codes are in progress, but they still have mismatch in the predicted value and unstable convergence. In this paper, FEM code for fully coupled THM phenomena is developed to analyze methane hydrate dissociation in the porous media. Coupled partial differential equations are derived from four mass balance equations (methane hydrate, soil, water, and hydrate gas), energy balance equation, and force equilibrium equation. Five main variables (displacement, gas saturation, fluid pressure, temperature, and hydrate saturation) are chosen to give higher numerical convergence through trial combinations of variables, and they can analyze the whole region of a phase change in hydrate bearing porous media. The kinetic model is used to predict dissociation of methane hydrate. Developed THM FEM code is applied to the comparative study on a Masuda's laboratory experiment for the hydrate production, and verified for the stability and convergence.

• Fire Science and Engineering
• /
• v.26 no.3
• /
• pp.79-84
• /
• 2012

In this study thermal characteristics of heat-responsive element considering conduction, convection and rate of change of element using Response Time Index (RTI) applied to sensitivity test of sprinkler head at home and aborad are theoretically investigated. Analytic solution of temperature distributions with radial direction and time is obtained form energy transport equations, non-homogeneous 2th order partial differential equation, applying to constant wall temperature and symmetric condition in order to analyze thermal characteristics of heat-responsive element for circular cylindrical geometry. Base on the results, the analytic method of this study is fundamental data to practical use for sensitivity test of sprinkler head and design of heat-responsive element.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.21 no.5
• /
• pp.577-582
• /
• 2015

This paper is the fundamental study for the application of cold storage system to the transportation equipment by sea and land. This numerical study presents the solid-liquid phase change phenomenon of calcium chloride solution of 30wt %. The governing equations are 1-dimensional unsteady state heat transfer equations of $1^{st}$ order partial differential equations. This type of latent heat storage material is often usable in fishery vessel for controlling the temperature of container with constant condition. The governing equation was discretized with finite difference method and the program was composed with Mathcad program. The main parameters of this solution were the initial temperature of heat storage material, ambient temperature of cold air and the velocity of cold air. The data of boundary layer thickness becomes thin with the increasing of cold air flowing velocity and also the heat storage completion time become shorten.

• Smart Structures and Systems
• /
• v.18 no.2
• /
• pp.355-374
• /
• 2016

This paper presents and compares a one-dimensional (1D) bending theory for piezoelectric thin beam-type structures with resistive-inductive electrodes to ANSYS$^{(R)}$ three-dimensional (3D) finite element (FE) analysis. In particular, the lateral deflections and vibrations of slender piezoelectric beams are considered. The peculiarity of the piezoelectric beam model is the modeling of electrodes in such a manner that is does not fulfill the equipotential area condition. The case of ideal, perfectly conductive electrodes is a special case of our 1D model. Two-coupled partial differential equations are obtained for the lateral deflection and for the voltage distribution along the electrodes: the first one is an extended Bernoulli-Euler beam equation (second-order in time, forth order in space) and the second one the so-called Telegrapher's equation (second-order in time and space). Analytical results of our theory are validated by 3D electromechanically coupled FE simulations with ANSYS$^{(R)}$. A clamped-hinged beam is considered with various types of electrodes for the piezoelectric layers, which can be either resistive and/or inductive. A natural frequency analysis as well as quasi-static and dynamic simulations are performed. A good agreement between the extended beam theory and the FE results is found. Finally, the practical relevance of this type of electrodes is shown. It is found that the damping capability of properly tuned resistive or resistive-inductive electrodes exceeds the damping performance of beams, where the electrodes are simply linked to an optimized impedance.

• Water for future
• /
• v.24 no.4
• /
• pp.49-58
• /
• 1991

Flood routing in the Youngsan River was performed for the flood event of July, 1989 by two finite difference methods. The Saint Venant eq., a kind of hyperbolic partial differential equation is employed as governing equation and the explicit scheme (Leap Frog) and implicit scheme (Preissmann) are used to discretize the GE. As for the external boundary conditions, discharge and tidal elevation are upstream and downstream BC, respectively and estuary dam is included in internal BC. Lateral inflows and upstream discharges are the hourly results from storage function method, At Naju station, a Relatively upstream points in this river, the outputs are interpreted as good ones by comparing two numerical results of FDMs with the observed data and the calibrated results by storage function method. and two computational results are compared at the other sites, from middle stream and downstream points, and thus are considered reliable. Therefore, we can conclude from this research that these numerical models are adaptable in simulating and forecasting the flood in natural channels in Korea as well as existing hydrologic models. And the study about optimal gate control at the flood time is expected as further study using these models.

• Journal of Aerospace System Engineering
• /
• v.14 no.3
• /
• pp.17-23
• /
• 2020

In this paper, the efficient periodic unsteady flow analysis is developed by using a Chebyshev collocation operator applied to the time differential term of the governing equations. The partial implicit time integration method was also applied in the governing equation for a fluid, which means flux terms were implicitly processed for a time integration and the time derivative terms were applied explicitly in the form of the source term by applying the Chebyshev collocation operator. To verify this method, we applied the 1D unsteady Burgers equation and the 2D oscillating airfoil. The results were compared with the existing unsteady flow frequency analysis technique, the Harmonic Balance Method, and the experimental data. The Chebyshev collocation operator can manage time derivatives for periodic and non-periodic problems, so it can be applied to non-periodic problems later.