• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.4
• /
• pp.501-507
• /
• 2021

Neural networks with differentiable activation functions are differentiable with respect to input variables. We improve the approximation capability of neural networks by using the gradient and Hessian of neural networks to satisfy the differential equations of the problems of interest. We apply differential neural networks to the pricing of financial options, where stochastic differential equations and the Black-Scholes partial differential equation represent the differential relation of price of option and underlying assets, and the first and second derivatives of option price play an important role in financial engineering. The proposed neural network learns - (a) the sample paths of option prices generated by stochastic differential equations and (b) the Black-Scholes equation at each time and asset price. Experimental results show that the proposed method gives accurate option values and the first and second derivatives.

• Structural Engineering and Mechanics
• /
• v.62 no.1
• /
• pp.65-77
• /
• 2017

The free vibration analysis of fluid conveying Timoshenko pipeline with different boundary conditions using Differential Transform Method (DTM) and Adomian Decomposition Method (ADM) has not been investigated by any of the studies in open literature so far. Natural frequencies, modes and critical fluid velocity of the pipelines on different supports are analyzed based on Timoshenko model by using DTM and ADM in this study. At first, the governing differential equations of motion of fluid conveying Timoshenko pipeline in free vibration are derived. Parameter for the nondimensionalized multiplication factor for the fluid velocity is incorporated into the equations of motion in order to investigate its effects on the natural frequencies. For solution, the terms are found directly from the analytical solution of the differential equation that describes the deformations of the cross-section according to Timoshenko beam theory. After the analytical solution, the efficient and easy mathematical techniques called DTM and ADM are used to solve the governing differential equations of the motion, respectively. The calculated natural frequencies of fluid conveying Timoshenko pipelines with various combinations of boundary conditions using DTM and ADM are tabulated in several tables and figures and are compared with the results of Analytical Method (ANM) where a very good agreement is observed. Finally, the critical fluid velocities are calculated for different boundary conditions and the first five mode shapes are presented in graphs.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.448-448
• /
• 2000

This paper deals with dynamic behaviour analysis for pipeline inspection gauge (PIG) flow control in natural gas pipeline. The dynamic behaviour of the PIG is depending on the different Pressure between the rear and nose parts, which is generated by injected gas flow behind PIG's tail and expelled gas flow in front of its nose. To analyze the dynamic behaviour characteristics such as gas flow in pipeline, and the PIG's position and velocity, mathematical model is derived as two types of a nonlinear hyperbolic partial differential equation for unsteady flow analysis of the PIG driving and expelled gas, and nonhomogeneous differential equation for dynamic analysis of PIG. The nonlinear equation is solved by method of characteristics (MOC) with the regular rectangular grid under appropriate initial and boundary conditions. The Runge-Kuta method is used when we solve the steady flow equations to get initial flow values and the dynamic equation of PIG. The gas upstream and downstream of PIG are divided into a number of elements of equal length. The sampling time and distance are chosen under Courant-Friedrich-Lewy (CFL) restriction. The simulation is performed with a pipeline segment in the Korea Gas Corporation (KOGAS) low pressure system, Ueijungboo-Sangye line. The simulation results show us that the derived mathematical model and the proposed computational scheme are effective for estimating the position and velocity of PIG with different operational conditions of pipeline.

• Journal of applied mathematics & informatics
• /
• v.32 no.5_6
• /
• pp.807-816
• /
• 2014

The asymptotic behavior of solutions of Lyapunov type matrix Volterra integro differential equation, in which the coefficient matrices are not stable, is studied by the method of reduction.

• Journal of Korean Society of Steel Construction
• /
• v.22 no.3
• /
• pp.241-251
• /
• 2010

This paper presents the development of a new analytical solution to the governing differential equation for isotropic annular sector plates subjected to uniform loading in a three-dimensional polar coordinate system. The 4th order governing partial differential equation (PDE) was converted to an ordinary differential equation (ODE) by assuming the Levy-type series solution form and the subsequent mathematical operations. Finally, a series-type solution was assembled with homogeneous and nonhomogeneous solution parts after operating real values and complex conjugates derived from the characteristic equation. To demonstrate the convergence rate and the accuracy of the featured method, several examples with various sector angles were selected and solved. The deflections and internal moments in the example annular sector plates that were obtained from the proposed solution were compared with those obtained from other analytical studies and numerical analyses using the finite element analysis package program, ABAQUS. Very good agreement with the results of other analytical and numerical methodologies was shown.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1992.10a
• /
• pp.349-353
• /
• 1992

The purpose of this paper is to develop methods for the dynamic analysis of multibody system that consist of interconnected rigid and deformable component. The equations of motion are derived by using the Lagrange's equation and finite element theory for the elastic mechanism systems. The type of equation of motion is the differential algebraic equation included kinematic nonlinear algebraic equation. The generalized coordinate partitioning method is used for solving this equation. To show the validity of this analysis solver, couple of models were canalized and those results were compared with the commercial package(ADAMS).

• Journal of the Chungcheong Mathematical Society
• /
• v.21 no.3
• /
• pp.413-426
• /
• 2008

An asymptotic solution of a fourth order critically damped nonlinear differential system has been found by means of extended Krylov-Bogoliubov-Mitropolskii (KBM) method. The solutions obtained by this method agree with those obtained by numerical method. The method is illustrated by an example.

• Bulletin of the Korean Mathematical Society
• /
• v.59 no.6
• /
• pp.1495-1510
• /
• 2022

This paper aims to investigate the initial boundary value problem of the nonlinear viscoelastic Petrovsky type equation with nonlinear damping and logarithmic source term. We derive the blow-up results by the combination of the perturbation energy method, concavity method, and differential-integral inequality technique.

• Journal of Electrical Engineering and Technology
• /
• v.5 no.4
• /
• pp.538-544
• /
• 2010

This paper proposes an estimation algorithm for the separate primary and secondary leakage inductances of a three phase $Y-\Delta$ transformer using least squares method. The voltage equations from the primary and secondary windings are combined into a differential equation to estimate the separate primary and secondary leakage inductances in order to use the line current of the delta winding. Separate primary and secondary leakage inductances are obtained by applying least squares method to the differential equation. The performance of the proposed algorithm is validated under transient states, such as magnetic inrush and overexcitation, as well as in the steady state with various cut-off frequencies of low-pass filter. The proposed technique can accurately generate separate leakage inductances both in the steady and transient states.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.16 no.2
• /
• pp.21-27
• /
• 2008

An inverse method is introduced to construct the problem for the error analysis of the numerical solution of initial value problem. These problems constructed through this method have a known exact solution, even though analytical solutions are generally not obtainable. The process leading to the exact solution makes use of an initially available approximate numerical solution. A smooth interpolation of the approximate solution is forced to exactly satisfy the differential equation by analytically deriving a small forcing function to absorb all of the errors in the interpolated approximate solution. Using this special case exact solution, it is possible to investigate the relationship between global errors of a candidate numerical solution process and the associated tuning parameters for a given problem. Under the assumption that the original differential equation is well-posed with respect to the small perturbations, we thereby obtain valuable information about the optimal choice of the tuning parameters and the achievable accuracy of the numerical solution.