• Proceedings of the Korea Society for Simulation Conference
• /
• 1994.10a
• /
• pp.1-2
• /
• 1994

With the prevalence of computers in modern organizations, simulation is receiving more atention as an effectvie decision -making tool. Simualtion is a computer-based numerical technique which uses mathmatical and logical models to approximate the behaviror of a real-world system. However, iptimization of synamic stochastic systems often defy analytical and algorithmic soluions. Although a simulation approach is often free fo the liminting assumption s of mathematical modeling, cost and time consiceration s make simulation the henayst's last resort. Therefore, whenever possible, analytical and algorithmica solutions are favored over simulation. This paper discussed the issues and procedrues for using simulation as a tool for optimization of stochastic complex systems that are dmodeled by computer simulation . Its emphasis is mostly on issues that are speicific to simulation optimization instead of consentrating on the general optimizationand mathematical programming techniques . A simulation optimization problem is an optimization problem where the objective function. constraints, or both are response that can only be evauated by computer simulation. As such, these functions are only implicit functions of decision parameters of the system, and often stochastic in nature as well. Most of optimization techniqes can be classified as single or multiple-resoneses techniques . The optimization of single response functins has been researched extensively and consists of many techniques. In the single response category, these strategies are gradient based search techniques, stochastic approximate techniques, response surface techniques, and heuristic search techniques. In the multiple response categroy, there are basically five distinct strategies for treating the responses and finding the optimum solution. These strategies are graphica techniqes, direct search techniques, constrained optimization techniques, unconstrained optimization techniques, and goal programming techniques. The choice of theprocedreu to employ in simulation optimization depends on the analyst and the problem to be solved. For many practival and industrial optimization problems where some or all of the system components are stochastic, the objective functions cannot be represented analytically. Therefore, modeling by computersimulation is one of the most effective means of studying such complex systems. In this paper, after discussion of simulation optmization techniques, the applications of above techniques will be presented in the modeling process of many flexible manufacturing systems.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.1
• /
• pp.11-25
• /
• 2021

The objective of this study was to analyze the impact of activation functions on flood forecasting model based on Artificial neural networks (ANNs). The traditional activation functions, the sigmoid and tanh functions, were compared with the functions which have been recently recommended for deep neural networks; the ReLU, leaky ReLU, and ELU functions. The flood forecasting model based on ANNs was designed to predict real-time runoff for 1 to 6-h lead time using the rainfall and runoff data of the past nine hours. The statistical measures such as R2, Nash-Sutcliffe Efficiency (NSE), Root Mean Squared Error (RMSE), the error of peak time (ETp), and the error of peak discharge (EQp) were used to evaluate the model accuracy. The tanh and ELU functions were most accurate with R2=0.97 and RMSE=30.1 (㎥/s) for 1-h lead time and R2=0.56 and RMSE=124.6~124.8 (㎥/s) for 6-h lead time. We also evaluated the learning speed by using the number of epochs that minimizes errors. The sigmoid function had the slowest learning speed due to the 'vanishing gradient problem' and the limited direction of weight update. The learning speed of the ELU function was 1.2 times faster than the tanh function. As a result, the ELU function most effectively improved the accuracy and speed of the ANNs model, so it was determined to be the best activation function for ANNs-based flood forecasting.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.147-170
• /
• 2022

Although the number of patent which is one of the core outputs of technological innovation continues to increase, the number of low-value patents also hugely increased. Therefore, efficient evaluation of patents has become important. Estimation of patent lifespan which represents private value of a patent, has been studied for a long time, but in most cases it relied on a linear model. Even if machine learning methods were used, interpretation or explanation of the relationship between explanatory variables and patent lifespan was insufficient. In this study, patent lifespan (number of renewals) is predicted based on the idea that patent lifespan represents the value of the patent. For the research, 4,033,414 patents applied between 1996 and 2017 and finally granted were collected from USPTO (US Patent and Trademark Office). To predict the patent lifespan, we use variables that can reflect the characteristics of the patent, the patent owner's characteristics, and the inventor's characteristics. We build four different models (Ridge Regression, Random Forest, Feed Forward Neural Network, Gradient Boosting Models) and perform hyperparameter tuning through 5-fold Cross Validation. Then, the performance of the generated models are evaluated, and the relative importance of predictors is also presented. In addition, based on the Gradient Boosting Model which have excellent performance, Accumulated Local Effects Plot is presented to visualize the relationship between predictors and patent lifespan. Finally, we apply Kernal SHAP (SHapley Additive exPlanations) to present the evaluation reason of individual patents, and discuss applicability to the patent evaluation system. This study has academic significance in that it cumulatively contributes to the existing patent life estimation research and supplements the limitations of existing patent life estimation studies based on linearity. It is academically meaningful that this study contributes cumulatively to the existing studies which estimate patent lifespan, and that it supplements the limitations of linear models. Also, it is practically meaningful to suggest a method for deriving the evaluation basis for individual patent value and examine the applicability to patent evaluation systems.

• Journal of Korean Port Research
• /
• v.13 no.1
• /
• pp.167-174
• /
• 1999

Numerical analysis of two-fluid flows for both water and air is carried out. Free-Surface flows with an arbitrary deformation have been simulated around two dimensional submerged hydrofoil. The computation is performed using a finite volume method with unstructured meshes and an interface capturing scheme to determine the shape of the free surface. The method uses control volumes with an arbitrary number of faces and allows cell-wise local mesh refinement. the integration in space is of second order based on midpoint rule integration and linear interpolation. The method is fully implicit and uses quadratic interpolation in time through three time levels The linear equation systems are solved by conjugate gradient type solvers and the non-linearity of equations is accounted for through picard iterations. The solution method is of pressure-correction type and solves sequentially the linearized momentum equations the continuity equation the conservation equation of one species and the equations or two turbulence quantities.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2004.05a
• /
• pp.216-221
• /
• 2004

In two dimensional incompressible flaws, a preconditioning technique called Hierarchical Iterative Procedure(HIP) has been implemented on a SUPG finite element formulation. By using the SUPG formulation, one can escape from the LBB constraint and hence achieve an equal order formulation. In this paper, we increased the order of interpolation up to cubic. The conjugate gradient squared(CGS) method is used for the outer iteration, and the HIP for the preconditioning for the incompressible Navier-Stokes equation. The hierarchical elements has been used to achieve a higher order accuracy in fluid flaw analyses, but a proper efficient iterative procedure for higher order finite element formulation has not been available so far. The numerical results by the present HIP for the lid driven cavity flaw showed the present procedure to be stable, very efficient and useful in flaw analyses in conjunction with hierarchical elements.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10a
• /
• pp.189-192
• /
• 1996

This paper presents a systematic approach to identify a linguistic fuzzy model for a multi-input and single-output complex system. Such a model is composed of fuzzy rules, and its output is inferred by the simplified reasoning. The structure and membership function parameters for a fuzzy model are automatically and simultaneously identified by GA (Genetic Algorithm). After GA search, optimal parameters for the fuzzy model are finely tuned by a gradient method. A numerical example is provided to evaluate the feasibility of the proposed approach. Comparison shows that the suggested approach can produce the linguistic fuzzy model with higher accuracy and a smaller number of rules than the ones achieved previously in other methods.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10a
• /
• pp.137-140
• /
• 1996

Finding line segments in an intensity image has been one of the most fundamental issues in computer vision. In complex scenes, it is hard to detect the locations of point features. Line features are more robust in providing greater positional accuracy. In this paper we present a robust "line features extraction" algorithm which extracts line feature in a single pass without using any assumptions and constraints. Our algorithm consists of five steps: (1) edge scanning, (2) edge normalization, (3) line-blob extraction, (4) line-feature computation, and (5) line linking. By using edge scanning, the computational complexity due to too many edge pixels is drastically reduced. Edge normalization improves the local quantization error induced from the gradient space partitioning and minimizes perturbations on edge orientation. We also analyze the effects of edge processing, and the least squares-based method and the principal axis-based method on the computation of line orientation. We show its efficiency with some real images.al images.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.803-806
• /
• 1997

In this paper, a force estimation method is proposed for the sensorless force control. For this, a disturbance observer is applied to each joint of an n degrees of freedom manipulator to obtain a simple equivalent robot dynamics(SERD) being represented as an n independent double integrator system. To estimate the output of disturbance observer in the absence of external force, the observer estimator is designed, where the uncertain parameters of the robot manipulator are adjusted by gradient method to minimize the output between the disturbance observer and the observer estimator. When the external force is exerted, the external force is estimated using the difference between the output of disturbance observer which include the external torque signal and that of observer estimator. And then, a force controller is designed for force feedback control employing the estimated force signal. To verify the effectiveness of the proposed force estimation method, several numerical examples are illustrated for the 2-axis planar type robot manipulator.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.2498-2501
• /
• 2003

In this paper, we present a wavelet neural network (WNN) approach to the solution of the path tracking problem for mobile robots that possess complexity, nonlinearity and noise. First, we discuss a WNN based control system where the control signals are directly obtained by minimizing the difference between the reference track and the pose of a mobile robot. This compact network structure is helpful to determine the number of hidden nodes and the initial value of weights. Then, the data with various noises provided by odometric and external sensors are here fused together by means of an Extended Kalman Filter (EKF) approach for the pose estimation problem of mobile robots. This control process is a dynamic on-line process that uses the wavelet neural network trained via the gradient-descent method with estimates from EKF. Finally, we verify the effectiveness and feasibility of the proposed control system through simulations.

• Proceedings of the Computational Structural Engineering Institute Conference
• /
• 2004.04a
• /
• pp.119-126
• /
• 2004

A continuum-based design sensitivity analysis and topology optimization methods are developed for power flow analysis. Efficient adjoint sensitivity analysis method is employed and further extended to topology optimization problems. Young's moduli of all the finite elements are selected as design variables and parameterized using a bulk material density function. The objective function and constraint are an energy compliance of the system and an allowable volume fraction, respectively. A gradient-based optimization, the modified method of feasible direction, is used to obtain the optimal material layout. Through several numerical examples, we notice that the developed design sensitivity analysis method is very accurate and efficient compared with the finite difference sensitivity. Also, the topology optimization method provides physically meaningful results. The developed is design sensitivity analysis method is very useful to systematically predict the impact on the design variations. Furthermore, the topology optimization method can be utilized in the layout design of structural systems.