• Korean Journal of Remote Sensing
• /
• v.4 no.1
• /
• pp.41-47
• /
• 1988

The operational NOAA satellite temperature soundings are utilized to determine the horizontal wind and vertical motion fields for a polar low case over the East Asian region by solving the nonlinear balance equation and the omega equation. Preliminary results demonstrate that the balanced wind and vertical motion fields derived from the satellite data give reasonable synoptic patterns associated with the polar low. This encourages the use of satellite information as inputs in the numerical weather prediction models.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.39 no.10
• /
• pp.985-993
• /
• 2015

Free-piston Stirling engines (FPSEs) have attracted much attention in the renewable energy field as a key device in the conversion from thermal to mechanical energy, and in the recycling of waste energy. Traditional Stirling engines consist of two pistons that are connected by a mechanical link, while FPSEs are formed as a vibration system by connecting each piston to a spring without a physical link. To ensure the correct design and control of operations, this requires elaborate dynamic-performance predictions. In this paper, we present the performance-prediction methodology using a linear and nonlinear dynamic analytical model considering the external load of FPSEs. We perform linear analyses to predict the operating point of the engine using the root locus technique. Using nonlinear analysis, we also predict the amplitude of pistons by performing numerical integration considering both the linear and nonlinear damping terms of the external load. We utilize the predicted dynamic behavior to predict the engine performance. In addition, we compare the experiment results and existing model predictions for RE-1000 to verify the reliability of the analytical model.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.6
• /
• pp.489-498
• /
• 2023

Advancements in hardware performance and computing technology have facilitated the progress of climate prediction models to address climate change. The Korea Meteorological Administration employs the GloSea6 model with supercomputer technology for operational use. Various universities and research institutions utilize the Low-GloSea6 model, a low-resolution coupled model, on small to medium-scale servers for weather research. This paper presents an analysis using Intel VTune Profiler on Low-GloSea6 to facilitate smooth weather research on small to medium-scale servers. The tri_sor_dp_dp function of the atmospheric model, taking 1125.987 seconds of CPU time, is identified as a hotspot. Nonlinear regression models, a machine learning technique, are applied and compared to existing functions conducting numerical operations. The K-Nearest Neighbors regression model exhibits superior performance with MAE of 1.3637e-08 and SMAPE of 123.2707%. Additionally, the Light Gradient Boosting Machine regression model demonstrates the best performance with an RMSE of 2.8453e-08. Therefore, it is confirmed that applying a nonlinear regression model to the tri_sor_dp_dp function during the execution of Low-GloSea6 could be a viable alternative.

• Smart Structures and Systems
• /
• v.14 no.5
• /
• pp.785-809
• /
• 2014

In this paper, an Adaptive nerou-based inference system (ANFIS) is being used for the prediction of shear strength of high strength concrete (HSC) beams without stirrups. The input parameters comprise of tensile reinforcement ratio, concrete compressive strength and shear span to depth ratio. Additionally, 122 experimental datasets were extracted from the literature review on the HSC beams with some comparable cross sectional dimensions and loading conditions. A comparative analysis has been carried out on the predicted shear strength of HSC beams without stirrups via the ANFIS method with those from the CEB-FIP Model Code (1990), AASHTO LRFD 1994 and CSA A23.3 - 94 codes of design. The shear strength prediction with ANFIS is discovered to be superior to CEB-FIP Model Code (1990), AASHTO LRFD 1994 and CSA A23.3 - 94. The predictions obtained from the ANFIS are harmonious with the test results not accounting for the shear span to depth ratio, tensile reinforcement ratio and concrete compressive strength; the data of the average, variance, correlation coefficient and coefficient of variation (CV) of the ratio between the shear strength predicted using the ANFIS method and the real shear strength are 0.995, 0.014, 0.969 and 11.97%, respectively. Taking a look at the CV index, the shear strength prediction shows better in nonlinear iterations such as the ANFIS for shear strength prediction of HSC beams without stirrups.

• Journal of Korean Society of Water and Wastewater
• /
• v.21 no.5
• /
• pp.601-607
• /
• 2007

In order to maintain constant residual chlorine in sedimentation basin, It is necessary to develop real time prediction model of residual chlorine considering water treatment plant data such as water qualities, weather, and plant operation conditions. Based on the operation data acquired from K water treatment plant, prediction models of residual chlorine in sediment basin were accomplished. The input parameters applied in the models were water temperature, turbidity, pH, conductivity, flow rate, alkalinity and pre-chlorination dosage. The multiple regression models were established with linear and non-linear model with 5,448 data set. The corelation coefficient (R) for the linear and non-linear model were 0.39 and 0.374, respectively. It shows low correlation coefficient, that is, these multiple regression models can not represent the residual chlorine with the input parameters which varies independently with time changes related to weather condition. Artificial neural network models are applied with three different conditions. Input parameters are consisted of water quality data observed in water treatment process based on the structure of auto-regressive model type, considering a time lag. The artificial neural network models have better ability to predict residual chlorine at sediment basin than conventional linear and nonlinear multi-regression models. The determination coefficients of each model in verification process were shown as 0.742, 0.754, and 0.869, respectively. Consequently, comparing the results of each model, neural network can simulate the residual chlorine in sedimentation basin better than mathematical regression models in terms of prediction performance. This results are expected to contribute into automation control of water treatment processes.

• Computers and Concrete
• /
• v.19 no.4
• /
• pp.419-427
• /
• 2017

With the rising global environmental awareness on energy saving and carbon reduction, as well as the environmental transition and natural disasters resulted from the greenhouse effect, waste resources should be efficiently used to save environmental space and achieve environmental protection principle of "sustainable development and recycling". This study used recycled cement mortar and adopted the volumetric method for experimental design, which replaced cement (0%, 10%, 20%, 30%) with recycled materials (fly ash, slag, glass powder) to test compressive strength and ultrasonic pulse velocity (UPV). The hyperbolic function for nonlinear multivariate regression analysis was used to build prediction models, in order to study the effect of different recycled material addition levels (the function of $R_m$(F, S, G) was used and be a representative of the content of recycled materials, such as fly ash, slag and glass) on the compressive strength and UPV of cement mortar. The calculated results are in accordance with laboratory-measured data, which are the mortar compressive strength and UPV of various mix proportions. From the comparison between the prediction analysis values and test results, the coefficient of determination $R^2$ and MAPE (mean absolute percentage error) value of compressive strength are 0.970-0.988 and 5.57-8.84%, respectively. Furthermore, the $R^2$ and MAPE values for UPV are 0.960-0.987 and 1.52-1.74%, respectively. All of the $R^2$ and MAPE values are closely to 1.0 and less than 10%, respectively. Thus, the prediction models established in this study have excellent predictive ability of compressive strength and UPV for recycled materials applied in cement mortar.

• Asian Pacific Journal of Cancer Prevention
• /
• v.13 no.3
• /
• pp.927-930
• /
• 2012

Background and Objectives: Artificial neural networks (ANNs) are flexible and nonlinear models which can be used by clinical oncologists in medical research as decision making tools. This study aimed to predict distant metastasis (DM) of colorectal cancer (CRC) patients using an ANN model. Methods: The data of this study were gathered from 1219 registered CRC patients at the Research Center for Gastroenterology and Liver Disease of Shahid Beheshti University of Medical Sciences, Tehran, Iran (January 2002 and October 2007). For prediction of DM in CRC patients, neural network (NN) and logistic regression (LR) models were used. Then, the concordance index (C index) and the area under receiver operating characteristic curve (AUROC) were used for comparison of neural network and logistic regression models. Data analysis was performed with R 2.14.1 software. Results: The C indices of ANN and LR models for colon cancer data were calculated to be 0.812 and 0.779, respectively. Based on testing dataset, the AUROC for ANN and LR models were 0.82 and 0.77, respectively. This means that the accuracy of ANN prediction was better than for LR prediction. Conclusion: The ANN model is a suitable method for predicting DM and in that case is suggested as a good classifier that usefulness to treatment goals.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.24 no.2
• /
• pp.193-200
• /
• 2011

A BOR(Boil-Off Rate) prediction model for the NO96 membrane-type LNG insulation containment filled with superlite powders during laden voyage is presented in this paper. Finite element model for the unsteady-state heat transfer analysis is constructed by considering the air and water conditions and by employing the homogenization method to simplify the complex insulation material composition. BOR is evaluated in terms of the total amount of heat invaded into LNGCC and its variation to the major variables is investigated by the parametric heat transfer analysis. Based upon the parametric results, a BOR prediction model which is in function of the LNG tank size, the insulation layer thickness and the powder thermal conductivity is derived. Through the verification experiment, the accuracy of the derived prediction model is justified such that the maximum relative difference is less than 1% when compared with the direct numerical estimation using the FEM analysis.

• Journal of the Institute of Convergence Signal Processing
• /
• v.5 no.3
• /
• pp.190-197
• /
• 2004

This paper presents automated prediction of transient ischemic episodes using neural networks(NN) based pattern matching method. The learning algorithm used to train the multilayer networks is a modified backpropagation algorithm. The algorithm updates parameters of nonlinear function in a neuron as well as connecting weights between neurons to improve learning speed. The performance of the method was evaluated using ECG signals of the MIT/BIH long-term database. Experimental results for 15 records(237 ischemic episodes) show that the average sensitivity and specificity of ischemic episode prediction are 85.71% and 71.11%, respectively. It is also found that the proposed method predicts an average of 45.53［sec］ ahead real ischemia. These results indicate that the NN approach as the pattern matching classifier can be a useful tool for the prediction of transient ischemic episodes.

• The Korean Journal of Applied Statistics
• /
• v.34 no.5
• /
• pp.697-710
• /
• 2021

Relative error prediction is preferred over ordinary prediction methods when relative/percentile errors are regarded as important, especially in econometrics, software engineering and government official statistics. The relative error prediction techniques have been developed in linear/nonlinear regression, nonparametric regression using kernel regression smoother, and stationary time series models. However, random effect models have not been used in relative error prediction. The purpose of this article is to extend relative error prediction to some of generalized linear mixed model (GLMM) with panel data, which is the random effect models based on gamma, lognormal, or inverse gaussian distribution. For better understanding, the real auto insurance data is used to predict the claim size, and the best predictor and the best relative error predictor are comparatively illustrated.