• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.50 no.9
• /
• pp.399-399
• /
• 2001

This paper proposes TAR(Threshold Autoregressive) model for short-term load forecasting including temperature variable. In the scatter diagram of daily peak load versus daily high or low temperature, we can find out that the load-temperature relationship has a negative slope in the lower regime and a positive slope in the upper regime due to the heating and cooling load, respectively. TAR model is adequate for analyzing these phenomena since TAR model is a piecewise linear autoregressive model. In this paper, we estimated and forecasted one day-ahead daily peak load by applying TAR model using this load-temperature characteristic in these regimes. The results are compared with those of linear and quadratic regression models.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.50 no.9
• /
• pp.309-405
• /
• 2001

This paper proposes TAR(Threshold Autoregressive) model for short-term load forecasting including temperature variable. In the scatter diagram of daily peak load versus daily high or low temperature, we can find out that the load-temperature relationship has a negative slope in the lower regime and a positive slope in the upper regime due to the heating and cooling load, respectively. TAR model is adequate for analyzing these phenomena since TAR model is a piecewise linear autoregressive model. In this paper, we estimated and forecasted one day-ahead daily peak load by applying TAR model using this load-temperature characteristic in these regimes. The results are compared with those of linear and quadratic regression models.

• Journal of Ocean Engineering and Technology
• /
• v.32 no.1
• /
• pp.1-8
• /
• 2018

In order to analyze the response of the offshore structure numerically, the linear potential theory is generally applied for simplicity, and only the radiation damping is considered among various damping forces. Therefore, the results of a numerical simulation can be different from the motion of the structure in a real environment. To reduce the differences between the simulation results and experimental results, the viscous damping, which affects the motion of the structure, is also taken into account. The appropriate damping model is essential for the numerical simulation in order to obtain precise responses of the offshore structure. In this study, various damping models such as linear or quadratic damping and the nonlinear drag force from numerous slender bodies were used to simulate the free decay motion of the platform, and its characteristics were confirmed. The optimized damping model was found by comparing the simulation results to the experimental results. The hydrodynamic forces and wave exciting forces of the structure were obtained using WAMIT, and the free decay test was simulated using OrcaFlex. A free decay test of the scale model was performed by KRISO.

• Progress in Medical Physics
• /
• v.28 no.1
• /
• pp.11-15
• /
• 2017

Relative biological effectiveness (RBE) of particle beam needs to be evaluated at particle beam therapy centers before the clinical application of the particle beam. However, since RBE analysis is implemented manually, it is useful to have a tool that can easily and effectively handle the data of experiments to generate cell survival curve and to analyze RBE simultaneously. In this work, the development of a program for RBE analysis of particle beam therapy was presented. This RBE analysis program was developed to include two parts; fitting the cell survival curves to linear-quadratic model and calculating the RBE values at a certain endpoint using fitting results. This program was also developed to simultaneously compare and analyze the template results that stored experiment data with photon and particle beam irradiations. The results of the cell survival curve obtained by each irradiation can be analyzed by the user on a desired data after reading the template stored in the easy-to-use excel file. The analysis results include the cell survival curves with error range, which are appeared in the screen and the ${\alpha}$ and ${\beta}$ parameters of linear-quadratic model with 95% confidence intervals, RBE values, and $R^2$ values to evaluate goodness-of-fit of survival curves to model, which are stored in a text cvs file. This software can generate cell survival curve, fit to model, and calculate RBE all at once with raw experiment data, so it helps users to save time for data handling and to reduce the possibility of making error on analysis. As a coming plan, we will create a user-friendly graphical user interface to present the results more intuitively.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.9
• /
• pp.45-52
• /
• 2006

In this study, a nonlinear control by feedback linearization method, one of nonlinear control schemes based on the nonlinear model, is proposed to suppress the flutter of a supersonic composite panel using piezoelectric materials. Most of the previous panel flutter controllers are the LQR(Linear Quadratic Regulator) which is based on the linear model. A nonlinear feedback linearizing controller proposed in this study considers the nonlinear characteristics of the system model. We use the actuator implemented by piezoceramic PZT. Using the principle of virtual displacements and a finite element discretization with the conforming four-node rectangular element, we first derive the discretized dynamic equations of motion, which are transformed into a nonlinear coupled-modal equations of motion of state space form. The effectiveness of the proposed method is also compared with the LQR based on the linear model through numerical simulations in the time domain using the Newmark method.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.3
• /
• pp.145-156
• /
• 2000

In this paper, we propose a new methodology which includes the optimal design procedure of Polynomial Neural Networks(PNN) structure for model identification of complex and nonlinear system. The proposed PNN algorithm is based on GMDA(Group Method of Data handling) method and its structure is similar to Neural Networks. But the structure of PNN is not fixed like in conventional Neural Networks and can be generated. The each node of PNN structure uses several types of high-order polynomial such as linear, quadratic and cubic, and is connected as various kinds of multi-variable inputs. In other words, the PNN uses high-order polynomial as extended type besides quadratic polynomial used in GMDH, and the number of input of its node in each layer depends on that of variables used in the polynomial. The design procedure to obtain an optimal model structure utilizing PNN algorithm is shown in each stage. The study is illustrated with the aid of pH neutralization process data besides representative time series data for gas furnace process used widely for performance comparison, and shows that the proposed PNN algorithm can produce the model with higher accuracy than previous other works. And performance index related to approximation and prediction capabilities of model is evaluated and also discussed.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.45 no.6
• /
• pp.978-985
• /
• 2021

This study aims to propose a prediction model for the drape coefficient using artificial neural networks and to analyze the nonlinear relationship between the drape properties and physical properties of fabrics. The study validates the significance of each factor affecting the fabric drape through multiple linear regression analysis with a sample size of 573. The analysis constructs a model with an adjusted R² of 77.6%. Seven main factors affect the drape coefficient: Grammage, extruded length values for warp and weft (m_warp, m_weft), coefficients of quadratic terms in the tensile-force quadratic graph in the warp, weft, and bias directions (c_warp, c_weft, c_bias), and force required for 1% tension in the warp direction (f_warp). Finally, an artificial neural network was created using seven selected factors. The performance was examined by increasing the number of hidden neurons, and the most suitable number of hidden neurons was found to be 8. The mean squared error was .052, and the correlation coefficient was .863, confirming a satisfactory model. The developed artificial neural network model can be used for engineering and high-quality clothing design. It is expected to provide essential data for clothing appearance, such as the fabric drape.

• The Journal of Fisheries Business Administration
• /
• v.54 no.1
• /
• pp.37-49
• /
• 2023

Climate change is a major global problem. Oysters, one of the most representative farmed fish in Korea, are attracting attention as candidates for blue carbon, an alternative to carbon neutrality. This study is analyzed by the SSP scenarios to determine the impact of oyster aquaculture production according to climate change. Based on the analysis, future productions of oysters are predicted by the SSP scenario. Significant differences by the SSP scenario are confirmed through predictive power tests among scenarios. Regression analysis was conducted from January 2001 to December 2014. As a result of the analysis, water temperature, water temperature quadratic term, salinity, salinity quadratic term, and month × water temperature cross term were estimated as significant variables. Oyster production which is predicted by the SSP scenario based on the significant variables from 2015 to 2022 was compared with actual production. The model with the highest predictive power was selected by RMSE and MAPE criteria. The predictive power was compared with the MDM test to determine which model was superior. As a result, based on RMSE and MAPE, the SSP1-2.6 scenario was selected as the best model and the SSP1-2.6, SSP2-4.5, and SSP3-7.0 scenarios all showed the same predictive power based on the MDM test. In conculusion, this study predicted oyster aquaculture production by 2030, not the distant future, due to the short duration of the analytical model. This study was found that oyster aquaculture production increased in all scenarios and there was no significant difference in predictive power by the SSP scenario.

• Journal of Electrochemical Science and Technology
• /
• v.15 no.3
• /
• pp.388-404
• /
• 2024

In the present work, the operating parameters were optimized using Box Behnken Design (BBD) in response surface methodology (RSM) to maximize the hydrogen production rate (R₁) and hydrogen production rate per unit watt consumed (R₂) of a proton exchange membrane electrolysis cell (PEMEC), a third response (R₃) which was the sum of the scaled values of R₁ and R₂ were selected to be maximized so that both hydrogen production rate and hydrogen production rate per unit watt consumed could be maximized. The major parameters which were influencing the experiment for enhancing the output responses were oxygen electrode/anode electrocatalyst loading (A), current supplied (B) and water inlet temperature (C). The commercial proton exchange membrane Nafion^® was used as the electrolyte. The acetylene black carbon (C_AB) supported IrO₂ was used as the electrocatalyst for preparing oxygen electrode/anode whereas commercial Pt (40 wt%)/C_HSA was used as the H₂ electrode/cathode electrocatalyst. The quadratic model was developed to predict the output/ responses and their proximity to the experimental output values. The developed model was found to be significant as the P values for both the responses were < 0.0001 and F values were greater than 1. The optimum condition for both the responses were O₂ electrode/anode electrocatalyst loading of 1.78 mg/cm², supplied current of 0.33 A and water inlet temperature of 54℃. The predicted values for hydrogen production rate (R₁) and hydrogen production rate per unit watt consumed (R₂) were 2.921 mL/min and 2.562 mL/(min·W), respectively obtained from the quadratic model. The error % between the predicted response values and experimental values were 1.47% and 3.08% for R₁ and R₂, respectively. This model predicted the optimum conditions reasonably in good agreement with the experimental conditions for the enhancement of the output responses of the developed PEM based electrolyser.

• Korean Journal of Poultry Science
• /
• v.44 no.4
• /
• pp.245-251
• /
• 2017

A dose-response experiment was conducted to determine the lysine requirement for broiler breeder hens during pre-peak production. Totally, one hundred and twenty-six flock 23-week-old Ross 308 broiler breeder hens with similar body weight were selected ($2,188{\pm}32g$) for a 6-week experiment. Hens were fed with a basal diet of corn-wheat-soybean meal formulated to achieve the Ross 308 breeder nutrient specifications (2016), except for lysine. The 7 graded, daily lysine intake levels used in this experiment were 732, 785, 838, 891, 944, 997, and 1,050 mg, and hens were restricted to 133 g of feed throughout this experiment. Pen based egg production were recorded once a day and all eggs were weighed daily. Age at sexual maturity was determined when the hens attained age at 25% production. Body weight at 23~29 weeks of age was not affected (P>0.05) by lysine levels. By fitting a linear-plateau model, the daily lysine requirements for feed conversion ratio, total produced egg weight, and age at sexual maturity at 23~29 weeks of age were estimated as 865, 907, and 891 mg, respectively. Using a quadratic-plateau model, the daily lysine requirement at 23~29 weeks of age were estimated as 974, 964, and 950 mg for feed conversion ratio, total produced egg weight, and age at sexual maturity, respectively. These results suggested that the daily lysine requirement for modern broiler breeder hens according to the National Research Council (1994) are insufficient for higher total produced egg weight, sexual maturity, and feed efficiency, and 120% of the NRC recommendation level would improve hen productivity when data are fitted under linear- and quadratic-plateau models.