• Journal of the Korean Society of Systems Engineering
• /
• v.18 no.2
• /
• pp.75-93
• /
• 2022

Machine learning (ML) data-driven meta-model is proposed as a surrogate model to reduce the excessive computational cost of the physics-based model and facilitate the real-time prediction of a nuclear power plant's transient response. To forecast the transient response three machine learning (ML) meta-models based on recurrent neural networks (RNNs); specifically, Long Short Term Memory (LSTM), Gated Recurrent Unit (GRU), and a sequence combination of Convolutional Neural Network (CNN) and LSTM are developed. The chosen accident scenario is a control element assembly withdrawal at power concurrent with the Loss Of Offsite Power (LOOP). The transient response was obtained using the best estimate thermal hydraulics code, MARS-KS, and cross-validated against the Design and control document (DCD). DAKOTA software is loosely coupled with MARS-KS code via a python interface to perform the Best Estimate Plus Uncertainty Quantification (BEPU) analysis and generate a time series database of the system response to train, test and validate the ML meta-models. Key uncertain parameters identified as required by the CASU methodology were propagated using the non-parametric Monte-Carlo (MC) random propagation and Latin Hypercube Sampling technique until a statistically significant database (181 samples) as required by Wilk's fifth order is achieved with 95% probability and 95% confidence level. The three ML RNN models were built and optimized with the help of the Talos tool and demonstrated excellent performance in forecasting the most probable NPP transient response. This research was guided by the Systems Engineering (SE) approach for the systematic and efficient planning and execution of the research.

• Proceedings of the KIEE Conference
• /
• 2006.07a
• /
• pp.143-144
• /
• 2006

The recent power systems are very complex and to model them completely is impractical for analysis of electromagnetic transient Therefore areas outside the immediate area of interest must be represented by some form of Frequency Dependent Network Equivalent (FDNE). In this paper a method for developing Frequency Dependent Network Equivalent (FDNE) using S-domain rational Function Fitting is presented and demonstrated. The FDNE is generated by Linearized Least Squares Fitting(LSF) of the frequency response of a S-domain formulation. This Three-port FDNE have been applied to the test AC power system. The electromagnetic transient package PSCAD/EMTDC is used to assess the transient response of the Three-port FDNE developed under different condition. The study results have indicated the robustness and accuracy of Three-port FDNE for analisys of electromagnetic transient and harmonic assessment.

• Proceedings of the KIEE Conference
• /
• 2006.07a
• /
• pp.145-146
• /
• 2006

The recent power systems are very complex and to model them completely is impractical for analysis of electromagnetic transient. Therefore areas outside the immediate area of interest must be represented by some form of Frequency Dependent Network Equivalent (FDNE). In this paper a method for developing Frequency Dependent Network Equivalent (FDNE) using Z-domain rational Function Fitting is presented and demonstrated. The FDNE is generated by Linearized Least Squares Fitting(LSF) of the frequency response of a Z-domain formulation. This Three-port FDNE have been applied to the test AC power system. The electromagnetic transient package PSCAD/EMTDC is used to assess the transient response of the Three-port FDNE developed under different condition. The study results have indicated the robustness and accuracy of Three-port FDNE for analisys of electromagnetic transient and harmonic assessment.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.16 no.9 s.114
• /
• pp.926-936
• /
• 2006

In this study, structural vibration analyses of a smart unmanned aerial vehicle (UAV) have been conducted considering dynamic loads generated by rotating rotor and wakes. The present UAV (TR-S5-03) finite element model is constructed as a full three-dimensional configuration with different fuel conditions and tilting angles for helicopter, transition and airplane flight modes. Practical computational procedure for modal transient response analysis (MTRA) is established using general purpose finite element method (FEM) and computational fluid dynamics (CFD) technique. The dynamic loads generated by rotating blades in the transient and forward flight conditions are calculated by unsteady CFD technique with sliding mesh concept. As the results of present study, transient structural displacements and accelerations are presented in detail. In addition, vibration characteristics of structural parts and installed equipments are investigated for different fuel conditions and tilting angles.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.367-375
• /
• 2006

In this study, structural vibration analyses of a smart unmanned aerial vehicle (UAV) have been conducted considering dynamic loads generated by rotating rotor and wakes. The present UAV (TR-S5-03) finite element model is constructed as a full three-dimensional configuration with different fuel conditions and tilting angles for helicopter, transition and airplane flight modes. Practical computational procedure for modal transient response analysis (MTRA) is established. using general purpose finite element method (FEM) and computational fluid dynamics (CFD) technique. The dynamic loads generated by rotating blades in the transient and forward flight conditions are calculated by unsteady CFD technique with sliding mesh concept. As the results of present study, transient structural displacements and accelerations are presented in detail. In addition, vibration characteristics of structural parts and installed equipments are investigated for different fuel conditions and tilting angles.

• Steel and Composite Structures
• /
• v.49 no.5
• /
• pp.587-602
• /
• 2023

This article presents the numerical modelling of transient heat transfer in highly heterogeneous composite materials where the thermal conductivity, specific heat and density are assumed to be directional-dependent. This article uses a coupled finite element-finite difference scheme to perform the transient heat transfer analysis of unidirectional (1D) and multidirectional (2D/3D) functionally graded composite panels. Here, 1D/2D/3D functionally graded structures are subjected to nonuniform heat source and inhomogeneous boundary conditions. Here, the multidirectional functionally graded materials are modelled by varying material properties in individual or in-combination of spatial directions. Here, fully spatial-dependent material properties are evaluated using Voigt's micromechanics scheme via multivariable power-law functions. The weak form is obtained through the Galerkin method and solved further via the element-space and time-step discretisation through the 2D-isoparametric finite element and the implicit backward finite difference schemes, respectively. The present model is verified by comparing it with the previously reported results and the commercially available finite element tool. The numerous illustrations confirm the significance of boundary conditions and material heterogeneity on the transient temperature responses of 1D/2D/3D functionally graded panels.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.34 no.1
• /
• pp.87-99
• /
• 1992

A mathematical model, UNSATR which predicts the seepage flow through the body of dike especially under the tidal fluctuation has been developed. This model has been revised from UNSAT2 model which was developed on the basis of the saturated-unsaturated theory by Neuman. UNSATR has been verified and applied to the hydraulic model in order to estimated the seepage quantity, the formation of free water surface etc. The results lead to the following conclusions : 1. Seepage rates between the mathematical model and hydraulic model experiment are very similar to each other both in constant and transient water level conditions. 2. The lapsed time to be steady state of the free water surface becomes late as the tidal levels are relatively low mainly due to the seepage flow from the unsaturated zone of the body of dike. 3. Under the transient state of water levels, owing to the flow from the unsaturated domain, streamlines crossing to the free water surface are found and time lag during a falling tide may allow the free water surface inside the body of dike to stand at a high level than the outside water level. 4. The utility and validity of UNSATR model are convinced when the analyses on seepage problems through the porous embankment of the soil structures on the conditions of the steady and unsteady states are carried out.

• Nuclear Engineering and Technology
• /
• v.28 no.3
• /
• pp.299-310
• /
• 1996

A two-step approach has been used to obtain a new criterion for the onset of slug formation : (1) In the first step, a more general expression than the existing models for the onset of slug flow criterion has been derived from the analysis of singular points and neutral stability conditions of the transient one-dimensional two-phase flow equations of two-fluid model. (2) In the second step, introducing simplifications and incorporating a parameter into the general expression obtained in the first step to satisfy a number of physical conditions a priori specified, a new simple criterion for the onset of slug flow has been derived. Comparisons of the present model with existing models and experimental data show that the present model agrees very closely with Taitel & Dukler's model and experimental data in horizontal pipes. In an inclined pipe ($\theta$ =50$^{\circ}$), however, the difference between the predictions of the present model and those of existing models is appreciably large and the present model gives the best agreement with Ohnuki et al.'s data.

• Advances in environmental research
• /
• v.6 no.1
• /
• pp.35-51
• /
• 2017

The use of Granular Activated Carbon (GAC) and naturally occurring silica (Sand) as filtration media in water and waste water treatment systems is very common. While GAC offers the additional functionality of being an "adsorptive" filter for dissolved organics it is also more expensive. In this paper we present an experimental evaluation of the performance of a bed of GAC for colloid removal and compare the same with that from an equivalent bed of Sand. The experiments are performed in an "intermittent" manner over extended time, to "simulate" performance over the life of the filter bed. The experiments were continued till a significant drop in water flow rate through the bed was observed. A novel "deposition" and "detachment" rate based transient mathematical model is developed. It is observed that the data from the experiments can be explained by the above model, for different aqueous phase electrolyte concentrations. The model "parameters", namely the "deposition" and "detachment" rates are evaluated for the 2 filter media studied. The model suggests that the significantly better performance of GAC in colloid filtration is probably due to significantly lower detachment of colloids from the same. While the "deposition" rates are higher for GAC, the "detachment" rates are significantly lower, which makes GAC more effective than sand for colloid removal by over an order of magnitude.

• 한국전산유체공학회:학술대회논문집
• /
• 2000.10a
• /
• pp.15-24
• /
• 2000

The transient response of an airfoil to a rapidly deploying spoiler is numerically investigated using the turbulent compressible Navier-Stokes equations in two dimensions. Algebraic Baldwin-Lomax model, Wilcox $\kappa-\omega$ model, and SST $\kappa-\omega$ turbulence model are used to calculate the unsteady separated flow due to the rapid spoiler deployment. The spoiler motion relative to a stationary airfoil is treated by an overset grid hounded by a Dynamic Domain-Dividing Line which has been devised by the authors. The adverse effects of the spoiler influenced by the spoiler location and the hinge gap are expounded. The numerical results are in reasonably good agreement with the existing experimental data.