• Journal of the Korean Operations Research and Management Science Society
• /
• v.16 no.1
• /
• pp.135-146
• /
• 1991

This paper shows the infinite horizon model of Partially Observable Markov Decision Process with lagged information. The lagged information is uncertain delayed observation of the process under control. Even though the optimal policy of the model exists, finding the optimal policy is very time consuming. Thus, the aim of this study is to find an .eplison.-optimal stationary policy minimizing the expected discounted total cost of the model. .EPSILON.- optimal policy is found by using a modified version of the well known policy iteration algorithm. The modification focuses to the value determination routine of the algorithm. Some properties of the approximation functions for the expected discounted cost of a stationary policy are presented. The expected discounted cost of a stationary policy is approximated based on these properties. A numerical example is also shown.

• Proceedings of the KSR Conference
• /
• 2001.05a
• /
• pp.123-128
• /
• 2001

Numerical simulation is conducted to clarify the heat transfer and fluid flow characteristics of brake shoe for freight car. High order up-wind scheme for governing equations, k-epsilon turbulent model and SIMPLEC algorithm based on finite volume method are used to solve the physical shoe model. The governing equations are solved by TDMA(Tri-Diagonal Matrix Algorithm) with line-by-line method and block correction. From the results of simulation, the characteristics of cooling pattern is strongly affected by the velocity of train and the material of shoe. The face lift of shoe affects on the temperature distribution of rear surface of shoe as well as the front surface of that. Due to the grooves in shoe, it will be expected to cool the frictional heat and result in the reduction of maintenance efforts.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.7 no.2
• /
• pp.375-389
• /
• 2015

An improved boundary element method is presented for numerical analysis of hydrodynamic behavior of marine structures. A new algorithm for numerical solution of the finite depth free-surface Green function in three dimensions is developed based on multiple series representations. The whole range of the key parameter R/h is divided into four regions, within which different representation is used to achieve fast convergence. The well-known epsilon algorithm is also adopted to accelerate the convergence. The critical convergence criteria for each representation are investigated and provided. The proposed method is validated by several well-documented benchmark problems.

• Proceedings of the KSR Conference
• /
• 1999.11a
• /
• pp.93-98
• /
• 1999

Numerical simulation is conducted to clarify the heat transfer and fluid flow characteristics of brake disk-lining for rolling stock. Multiple rotational reference frame, k-epsilon turbulent model and SIMPLE algorithm based on finite volume method are used to solve the physical disk-lining model. The governing equations are solved by TDMA(TriDiagonal Matrix Algorithm) with line-by-line method and block correction, From the results of simulation, the characteristics of cooling pattern is strongly affected by the grooves in lining. The face lift of lining affects on the temperature distribution of rear surface of lining as well as the front surface of that. Due to the grooves in lining, it will be expected to extend the maintenance life circle of lining.

• Nuclear Engineering and Technology
• /
• v.55 no.11
• /
• pp.4102-4111
• /
• 2023

In order to better perform thermal hydraulic calculation and analysis of supercritical water reactor, based on the experimental data of supercritical water, the model training and predictive analysis of the heat transfer coefficient of supercritical water were carried out by using the support vector machine (SVM) algorithm. The changes in the prediction accuracy of the supercritical water heat transfer coefficient are analyzed by the changes of the regularization penalty parameter C, the slack variable epsilon and the Gaussian kernel function parameter gamma. The predicted value of the SVM model obtained after parameter optimization and the actual experimental test data are analyzed for data verification. The research results show that: the normalization of the data has a great influence on the prediction results. The slack variable has a relatively small influence on the accuracy change range of the predicted heat transfer coefficient. The change of gamma has the greatest impact on the accuracy of the heat transfer coefficient. Compared with the calculation results of traditional empirical formula methods, the trained algorithm model using SVM has smaller average error and standard deviations. Using the SVM trained algorithm model, the heat transfer coefficient of supercritical water can be effectively predicted and analyzed.

• Fire Science and Engineering
• /
• v.13 no.4
• /
• pp.20-29
• /
• 1999

This paper describes the smoke filling process of a fire field model based on a self-developed SMEP(Smoke Movement Estimating Program) code to the simulation of fire induced flows in the two types of atrium space containing a ceiling heat flux. The SMEP using PISO algorithm solves conservation equations for mass, momentum, energy and species, together with those for the modified k- epsilon turbulence model with buoyancy term. Compressibility is assumed and the perfect gas law is used. The results of the calculated upper-layer average temperature and smoke layer interface height has shown reasonable agreement compared with the zone models. The zone models used are the CFAST developed at the Building and Fire Research Laboratory NIST U.S.A. and the NBTC one-room of FIRECALC developed at CSIRO, Australia. The smoke layer interface heights that are important in fire safety were not as sensitive as the smoke layer temperature to the nature of ceiling heat flux condition.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.13 no.2
• /
• pp.287-298
• /
• 1989

A numerical work was performed to study the flow behaviors of the open channel type flow with its geometric boundary conditions being similar to that of the Multi-Stage-Flash evaporator with and without a baffle. For the analysis, two-dimensional steady turbulent flow was assumed and the widely known k-.epsilon. turbulence model was usded. SIMPLE algorithm and the power difference scheme were used for the numerical approach. Numerical results generally agree with the previous experimental results though there are some uncertainties at far downstream and near the free surface due to the three dimensionality of the flow and surface waves. Without a baffle, the flow has basically the shape of the submerged plane wall jet with its upper boundary at downstream being sharply curved toward the free surface. For the flow with a baffle, recirculation flow patterns are observed at the upper inlet portion and at the backside of the baffle. For the case without a baffle, it was also confirmed that the ratio between the liquid level and the gate opening height is the most important parameter to determine the flow behavior.

• New & Renewable Energy
• /
• v.16 no.1
• /
• pp.15-24
• /
• 2020

For mechanical load testing of wind turbines, capture matrix is constructed for various range of wind speeds according to the international standard IEC 61400-13. The conventional method wastes considerable amount of data by its invalid data policy -segment data into 10 minutes then remove invalid ones. Previously, we have suggested an alternative way to save the total amount of data to build a capture matrix, but the efficient selection of data has been still under question. The paper introduces optimization algorithms to construct capture matrix with less data. Heuristic algorithm (simple stacking and lowest frequency first), population method (particle swarm optimization) and Q-Learning accompanied with epsilon-greedy exploration are compared. All algorithms show better performance than the conventional way, where the distribution of enhancement was quite diverse. Among the algorithms, the best performance was achieved by heuristic method (lowest frequency first), and similarly by particle swarm optimization: Approximately 28% of data reduction in average and more than 40% in maximum. On the other hand, unexpectedly, the worst performance was achieved by Q-Learning, which was a promising candidate at the beginning. This study is helpful for not only wind turbine evaluation particularly the viewpoint of cost, but also understanding nature of wind speed data.

• Journal of the military operations research society of Korea
• /
• v.17 no.1
• /
• pp.61-83
• /
• 1991

There has been a great deal of research dealing with the optimal replacement of stochastically deteriorating equipment and research in queueing systems with a finite calling population. However. there has been a lack of research which combines these two areas to deal with optimal replacement for a fixed number of machines and a limited number of repairmen. In this research, an optimal control policy for replacement involving two machines and one repairman is developed by investigating a class of age replacement policies in the context of controlling a G/M/1 queueing system with a finite calling population. The control policy to be imposed on this problem is an age-dependent control policy, described by the control limit $t^{\ast}$. The control limit is operational only when the repairman is idle; that is. if both machines are working, as soon as a machine reaches the age $t^{\ast}$ it is taken out of service for replacememt. We obtain the ${\epsilon}$-optimal control age which will minimize the long-run average system cost. An algorithm is developed that is applicable to general failure time distributions and cost functions. The algorithm does not require the condition of unimodality for implementation.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.22 no.3
• /
• pp.336-345
• /
• 1998

The substantial loss behind axial flow rotor was generated by wake, various vortices in the hub region and the leakage vortex in the tip region. Particularly, the leakage vortex formed near blade tip was one of the main causes of the reduction of performance, the generation of noise and the aerodynamic vibration in rotor downstream. In this study, the three-dimensional flowfields in an axial flow rotor for various tip clearances were calculated, and the numerical results were compared with the experimental ones. The numerical technique was based on SIMPLE algorithm using standard k-.epsilon. model (WFM). Through calculations, the effects of the tip clearance on the overall performance of rotor and the loss distributions, and the increase in the displacement, momentum, and blade-force-deficit thickness of the casing wall boundary layer were investigated. The mass-averaged flow variables behind rotor agreed well with the experimental results. The presence of the tip leakage vortex behind rotor was described well. Although the loci of leakage vortex by calculation showed some differences compared with the experimental results, its behavior for various tip clearances was clarified by examining the loci of vortex center.