• Korean Chemical Engineering Research
• /
• v.61 no.4
• /
• pp.542-549
• /
• 2023

Since the growing interest in surrogate modeling, there has been continuous research aimed at simulating nonlinear chemical processes using data-driven machine learning. However, the opaque nature of machine learning models, which limits their interpretability, poses a challenge for their practical application in industry. Therefore, this study aims to analyze chemical processes using Explainable Artificial Intelligence (XAI), a concept that improves interpretability while ensuring model accuracy. While conventional sensitivity analysis of chemical processes has been limited to calculating and ranking the sensitivity indices of variables, we propose a methodology that utilizes XAI to not only perform global and local sensitivity analysis, but also examine the interactions among variables to gain physical insights from the data. For the ammonia synthesis process, which is the target process of the case study, we set the temperature of the preheater leading to the first reactor and the split ratio of the cold shot to the three reactors as process variables. By integrating Matlab and Aspen Plus, we obtained data on ammonia production and the maximum temperatures of the three reactors while systematically varying the process variables. We then trained tree-based models and performed sensitivity analysis using the SHAP technique, one of the XAI methods, on the most accurate model. The global sensitivity analysis showed that the preheater temperature had the greatest effect, and the local sensitivity analysis provided insights for defining the ranges of process variables to improve productivity and prevent overheating. By constructing alternative models for chemical processes and using XAI for sensitivity analysis, this work contributes to providing both quantitative and qualitative feedback for process optimization.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.29 no.5
• /
• pp.405-411
• /
• 2016

This paper presented the theory related to a two dimensional linear cross-sectional analysis, recovery relationship and a one-dimensional nonlinear beam analysis for composite beam with initial twist and high aspect ratio. Using VABS including related theory, preceding research data of the composite wing structure has been modeled and compared. Cross-sectional analysis was performed and 1-D beam was modeled at cutting point including all the details of real geometry and material. The 3-D strain distribution and margin of safety at recovery point was calculated based on the global behavior of the 1-D beam analysis and visualize numerical results.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.19 no.2
• /
• pp.128-135
• /
• 2007

In response to anthropogenic global warming due to a buildup greenhouse gas, the effect of the air temperature on water temperature has been noticed. Therefore, effects have been made to build an air/water temperature relationship at three study regions using the data collected by the Ministry of the Maritime Affairs and Fisheries (MOMAF). The air/water relationship varies with time-scale and weekly time-scale was chosen for the study. The data were fitted to the S-shaped non-linear relationship, and the parameters for the S-curve were derived using a single-criteria multi-parameter optimization scheme. Separate regression curves were fitted to consider seasonal hysteresis at the Masan site. The study results support the S-shaped non-linear relationship is the best fit for the air/water relationship at the Korean coastal zone. This study will contribute to determine the future policy regarding water quality and ecosystem for the decision-driving organization.

• Korean Journal of Remote Sensing
• /
• v.26 no.5
• /
• pp.537-547
• /
• 2010

The importance of vegetation in studies of global climate and biogeochemical cycles is well recognized. Especially. the FPAR (fraction of photosynthetically active radiation) is one of the important parameters in ecosystem productivity and carbon budget models. Therefore, accurate estimates of vegetation parameters are increasingly important in environmental impact assessment studies. In this study, optical FPAR using the Terra MODIS (MODerate resolution Imaging Spectroradiometer), SPOT VEGETATION and ECOCLIMAP data reproduced on the Korean peninsula. We applied the empirical method which is usually estimated as a linear or nonlinear function of vegetation indices. As results, we estimated the accurate expression which is 0.9039 of $R^2$ in cropland and 0.7901 of $R^2$ in forest. Finally, this study could be demonstrated to calibrate that produced FPAR while the overall pattern and random noise through the comparative analysis of FPAR on the reference data. Optimal use of input parameter on the Korean peninsula should be helping the accuracy of output as well as the improved quality of research.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.26 no.3
• /
• pp.173-182
• /
• 2013

The building design projects which are being proceeded nowadays pursue a complex and various shape of structures, escaping from the traditional and regular shape of buildings. In this new trend of the architecture, there rises a demand of the research in the structural engineering for the effective realization of such complex-shaped buildings which disassembles the orthogonality of frames. As a distinguished characteristics of the buildings in a complex-shape, there frequently are inclined columns included in the structural frame. The inclined column causes extra axial force and bending moment at the beam-column connection so it is necessary to assess those effects on the structural behavior of the frame and the connection by experiment or analysis. However, with comparing to the studies on the normal beam-column connections, the inclined column connections have not been studied sufficiently. Therefore, this study evaluated the beam-column connections having an inclined column using nonlinear and finite element analysis method. In this paper, steel moment frames having inclined columns were analyzed by the nonlinear pushover analysis to check the global behavior and beam-column connection models were analyzed by the finite element analysis to check the buckling behavior and the fracture potentials.

• Journal of Astronomy and Space Sciences
• /
• v.36 no.4
• /
• pp.249-264
• /
• 2019

In this study, the precise orbit determination (POD) software is developed for optical observation. To improve the performance of the estimation algorithm, a nonlinear batch filter, based on the unscented transform (UT) that overcomes the disadvantages of the least-squares (LS) batch filter, is utilized. The LS and UT batch filter algorithms are verified through numerical simulation analysis using artificial optical measurements. We use the real optical observation data of a low Earth orbit (LEO) satellite, Cryosat-2, observed from optical wide-field patrol network (OWL-Net), to verify the performance of the POD software developed. The effects of light travel time, annual aberration, and diurnal aberration are considered as error models to correct OWL-Net data. As a result of POD, measurement residual and estimated state vector of the LS batch filter converge to the local minimum when the initial orbit error is large or the initial covariance matrix is smaller than the initial error level. However, UT batch filter converges to the global minimum, irrespective of the initial orbit error and the initial covariance matrix.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.12
• /
• pp.1238-1245
• /
• 2014

This paper proposes an RBFN-based adaptive tracking controller for an electrically driven mobile robot with parametric uncertainties and external disturbances. A mobile robot model considered in this paper includes all models of the robot body and actuators with uncertain kinematic and dynamic parameters, and uncertain frictions and external disturbances. The proposed controller consists of an RBFN(Radial Basis Function Network) and a robust adaptive controller. The presented RBFN is used to approximate unknown nonlinear robot dynamic functions. The proposed controller is adjusted by the adaptation laws obtained through the Lyapunov stability analysis. The proposed control scheme does not a priori need the accurate knowledge of all parameters in the robot kinematics, robot dynamics and actuator dynamics. Also, nominal parameter values are not required in the controller. The global stability of the closed-loop robot control system is guaranteed using the Lyapunov stability theory. Simulation results show the validity and robustness of the proposed control scheme.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.20 no.3
• /
• pp.98-105
• /
• 2012

Recently, most of the countries started to regulate the emission of vehicle because of the global warming. The engine scooter is also one of the factor which cause the pollution. The hybrid system of a vehicle has many advantages such as fuel saving and emission reduction. The purpose of this study is to choose optimal size of engine, motor and battery for hybrid scooter system using Dynamic programming. The dynamic programming is an effective method to find an optimal solution for the complicated nonlinear system, which contains various constraints of control variables. The power source size of hybrid scooter was chosen through the backward simulator using dynamic programming. From the analysis, we choose the optimal size of each power source. To verify the optimal size of the power source, the Forward simulation was carried out. As a result, the fuel efficiency of hybrid scooter has significantly increased in comparison with that of engine scooter.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.1
• /
• pp.215-236
• /
• 2017

Obtaining accurate location information is important in practical applications of wireless sensor networks (WSNs). The distance vector hop (DV-Hop) is a frequently-used range-free localization algorithm in WSNs, but it has low localization accuracy. Moreover, despite various improvements to DV-Hop-based localization algorithms, maintaining a balance between high localization accuracy and good stability and convergence is still a challenge. To overcome these shortcomings, we proposed an improved DV-Hop localization algorithm based on the bat algorithm (IBDV-Hop) for WSNs. The IBDV-Hop algorithm incorporates optimization methods that enhance the accuracy of the average hop distance and fitness function. We also introduce a nonlinear dynamic inertial weight strategy to extend the global search scope and increase the local search accuracy. Moreover, we develop an updated solutions strategy that avoids premature convergence by the IBDV-Hop algorithm. Both theoretical analysis and simulation results show that the IBDV-Hop algorithm achieves higher localization accuracy than the original DV-Hop algorithm and other improved algorithms. The IBDV-Hop algorithm also exhibits good stability, search capability and convergence, and it requires little additional time complexity and energy consumption.

• Journal of Korea Water Resources Association
• /
• v.31 no.6
• /
• pp.855-862
• /
• 1998

This paper investigates the impact of the forecast error on performance of a reservoir system for hydropower production. Forecast error is measured as th Root Mean Square Error (RMSE) and parametrically varied within a Generalized Maintenance Of Variance Extension (GMOVE) procedure. A set of transition probabilities are calculated as a function of the RMSE of the GMOVE procedure and then incorporated into a Bayesian Stochastic Dynamic Programming model which derives monthly operating policies and assesses their performance. As a case study, the proposed methodology is applied to the Skagit Hydropower System (SHS) in Washington state. The results show that the system performance is a nonlinear function of RMSE and therefor suggested that continued improvements in the current forecast accuracy correspond to gradually greater increase in performance of the SHS.