• International Journal of Computer Science & Network Security
• /
• v.22 no.7
• /
• pp.301-307
• /
• 2022

Energy consumption has grown alongside dramatic population increases. Statistics show that buildings in particular utilize a significant amount of energy, worldwide. Because of this, building energy prediction is crucial to best optimize utilities' energy plans and also create a predictive model for consumers. To improve energy prediction performance, this paper proposes a ResNet-LSTM model that combines residual networks (ResNets) and long short-term memory (LSTM) for energy consumption prediction. ResNets are utilized to extract complex and rich features, while LSTM has the ability to learn temporal correlation; the dense layer is used as a regression to forecast energy consumption. To make our model more robust, we employed Huber loss during the optimization process. Huber loss obtains high efficiency by handling minor errors quadratically. It also takes the absolute error for large errors to increase robustness. This makes our model less sensitive to outlier data. Our proposed system was trained on historical data to forecast energy consumption for different time series. To evaluate our proposed model, we compared our model's performance with several popular machine learning and deep learning methods such as linear regression, neural networks, decision tree, and convolutional neural networks, etc. The results show that our proposed model predicted energy consumption most accurately.

• Journal of Internet Computing and Services
• /
• v.10 no.4
• /
• pp.105-113
• /
• 2009

The feature models are widely used to model the commonalities and variabilities among the products in the domain engineering phase of software product line developments. The findings and corrections of the errors or consistencies in the feature models are essential to the successful software product line engineering. The aids of the automated tools are needed to perform the validation of the feature models effectively. This paper describes the approach based on JESS rule-base system to validate the feature models and proposes the feature model validation tool using this approach. The tool of this paper validates the feature models in real-time when modeling the feature models. Then it provides the information on existence of errors and the explanations on causes of those errors, which allows the feature modeler to create the error-free feature models. This attempt to validate the feature model using the rule-based system is supposed to be the first time in this research field.

• Magazine of the Korean Society of Agricultural Engineers
• /
• v.34 no.2
• /
• pp.73-84
• /
• 1992

The TIDE, finite element model for the simulation of tidal flow in shallow sea was tested for its applicability at the Saemangeum area. Several pre and post processors were developed to facilitate handling of the complicated and large amount of input and output data for the model developed. Also an operation scheme to run the model and the processors were established. As a result of calibration test using the observed data collected at 9 points within the region, linearlized friction coefficients were adjusted to be ranged 0.0027~0.0072, and water depths below the mean sea level at every nodes were changed to be increased generally by 1 meter. Comparisons of tidal velocities between the observed and the simulated for the 5 stations were made and obtained the result that the average relative error between simulated and observed tidal velocities was 11% for the maximum velocities and 22% for the minimum, and the absolute errors were less than 0.2m/sec. Also it was found that the average R.M.S. error between the velocities of observed and simulated was 0.119 m/sec and the average correlation coefficient was 0.70 showing close agreement. Another comparison test was done to show the result that R.M.S. error between the simulated and the observed tidal elevations at the 4 stations was 0.476m in average and the correlation coefficients were ranged 0.96~0.99. Though the simulated tidal circulation pattern in the region was well agreed with the observed, the simulated tidal velocities and elevations for specific points showed some errors with the observed. It was thought that the errors mainly due to the characteristics of TIDE Model which was developed to solve only with the linearized scheme. Finally it was concluded that, to improve the simulation results by the model, a new attempt to develop a fully nonlinear model as well as further calibration and the more reasonable generation of finite element grid would be needed.

• The KIPS Transactions:PartD
• /
• v.11D no.6
• /
• pp.1269-1276
• /
• 2004

The Parameter Estimation for software existing reliability models, Goel-Okumoto, Yamada-Ohba-Osaki model was reviewed and Rayleigh model based on Rayleigh distribution was studied. In this paper, we discusses comparison of parameter estimation using maximum likelihood estimator and Bayesian estimation based on Gibbs sampling to analysis of the estimator' pattern. Model selection based on sum of the squared errors and Braun statistic, for the sake of efficient model, was employed. A numerical example was illustrated using real data. The current areas and models of Superposition, mixture for future development are also employed.

• Journal of the Korean association of regional geographers
• /
• v.19 no.2
• /
• pp.352-361
• /
• 2013

This paper aims to analyze the root mean square errors of MODIS LST data and inland water temperature measurement data in order to use MODIS LST data as an input of numerical weather prediction model. MODIS LST data from July 2011 to June 2012 were compared to water temperature measurement data in the automated water quality measurement network. MODIS data have two composites: day-time and night-time. Monthly errors of day-time and night-time LST range $2{\sim}8^{\circ}C$ and $3{\sim}12^{\circ}C$, respectively. Temporally, monthly errors of day-time LST are less in fall and those of night-time LST are less in summer. Spatially, on the four major rivers including the Han, Nakdong, Geum, and Yeongsan rivers, the errors of Yeongsan river were the smallest, which location is the south-most among them. In this study, the errors of MODIS LST as an input of numerical weather prediction model were analyzed and the results can be used as an error level of MODIS LST data for inaccessible areas such as North Korea.

• Smart Structures and Systems
• /
• v.18 no.6
• /
• pp.1169-1188
• /
• 2016

Real-time substructuring techniques are currently an advanced experimental method for testing large size specimens in the laboratory. In dynamic substructuring, the whole tested system is split into two linked parts, the part of particular interest or nonlinearity, which is tested physically, and the remanding part which is tested numerically. To achieve near-perfect synchronization of the interface response between the physical specimen and the numerical model, a good controller is needed to compensate for transfer system dynamics, nonlinearities, uncertainties and time-varying parameters within the physical substructures. This paper presents the substructuring approach and control performance of the linear and the adaptive controllers for testing the dynamic characteristics of soil-structure-interaction system (SSI). This is difficult to emulate as an entire system in the laboratory because of the size and power supply limitations of the experimental facilities. A modified linear substructuring controller (MLSC) is proposed to replace the linear substructuring controller (LSC).The MLSC doesn't require the accurate mathematical model of the physical structure that is required by the LSC. The effects of parameter identification errors of physical structure and the shaking table on the control performance of the MLSC are analysed. An adaptive controller was designed to compensate for the errors from the simplification of the physical model in the MLSC, and from parameter identification errors. Comparative simulation and experimental tests were then performed to evaluate the performance of the MLSC and the adaptive controller.

• Korean Journal of Remote Sensing
• /
• v.33 no.1
• /
• pp.25-35
• /
• 2017

Spatial downscaling with fine resolution auxiliary variables has been widely applied to predict precipitation at fine resolution from coarse resolution satellite-based precipitation products. The spatial downscaling framework is usually based on the decomposition of precipitation values into trend and residual components. The fine resolution auxiliary variables contribute to the estimation of the trend components. The main focus of this study is on quantitative analysis of impacts of trend component estimates on predictive performance in spatial downscaling. Two regression models were considered to estimate the trend components: multiple linear regression (MLR) and geographically weighted regression (GWR). After estimating the trend components using the two models,residual components were predicted at fine resolution grids using area-to-point kriging. Finally, the sum of the trend and residual components were considered as downscaling results. From the downscaling experiments with time-series Tropical Rainfall Measuring Mission (TRMM) 3B43 precipitation data, MLR-based downscaling showed the similar or even better predictive performance, compared with GWR-based downscaling with very high explanatory power. Despite very high explanatory power of GWR, the relationships quantified from TRMM precipitation data with errors and the auxiliary variables at coarse resolution may exaggerate the errors in the trend components at fine resolution. As a result, the errors attached to the trend estimates greatly affected the predictive performance. These results indicate that any regression model with high explanatory power does not always improve predictive performance due to intrinsic errors of the input coarse resolution data. Thus, it is suggested that the explanatory power of trend estimation models alone cannot be always used for the selection of an optimal model in spatial downscaling with fine resolution auxiliary variables.

• Journal of the Korean Society of Safety
• /
• v.31 no.1
• /
• pp.126-131
• /
• 2016

Although many studies have been conducted to find solutions to deal with human errors effectively, violations have been rarely studied in depth. The violation is a type of human error when an employee takes an action with intention but does not intend harmful outcomes. Violations have characteristics similar to other types of human errors but it is difficult to understand the intention of an employee from accident reports. The objective of this study is to develop a conceptual model of violation errors for preventing accidents/failures in a nuclear power plant from violation errors. Based on the previous studies, the characteristics of violations were collected in 9 categories and 136 items. They were classified into three-kinds of characteristics (human-related, task-related, organization-related characteristics) to construct conceptual models of routine/situational violations. The representative cases of accidents/failures in a nuclear power plant were analyzed to derive the specific types of routine/situational violation occurrence. Three types of conceptual model for each violation were derived according to whether the basic, human-related, and task-related characteristics are included or not. The conceptual models can be utilized to develop guidelines to support employees preventing routine/situational violations and to develop supportive system in nuclear power plant.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.55 no.6
• /
• pp.101-112
• /
• 2013

Shuttle Radar Topography Mission Digital Elevation Model (SRTM DEM) offers opportunities to make advances in many research areas including hydrology by providing near-global scale elevation measurements at a uniform resolution. Its wide coverage and complimentary online access especially benefits researchers requiring topographic information of hard-to-access areas. However, SRTM DEM also contains inherent errors, which are subject to propagation with its manipulation into analysis outputs. Sensitivity of hydrologic analysis to the errors has not been fully understood yet. This study investigated their impact on estimation of hydrologic derivatives such as slope, stream network, and watershed boundary using Monte Carlo simulation and spatial moving average techniques. Different amount of the errors and their spatial auto-correlation structure were considered in the study. Two sub-watersheds of Geum and Deadong River areas located in South and North Korea, respectively, were selected as the study areas. The results demonstrated that the spatial presentations of stream networks and watershed boundaries and their length and area estimations could be greatly affected by the SRTM DEM errors, in particular relatively flat areas. In the Deadong River area, artifacts of the SRTM DEM created sinks even after the filling process and then closed drainage basin and short stream lines, which are not the case in the reality. These findings provided an evidence that SRTM DEM alone may not enough to accurately figure out the hydrologic feature of a watershed, suggesting need of local knowledge and complementary data.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.1
• /
• pp.122-131
• /
• 2011

A widely used single image-based 3D face reconstruction method, 3D morphable shape model, reconstructs an accurate 3D facial shape when 2D facial feature points are correctly extracted from an input face image. However, in the case that a user's cooperation is not available such as a real-time 3D face reconstruction system, this method can be vulnerable to the errors of automatic facial feature point extraction. In order to solve this problem, we automatically classify extracted facial feature points into two groups, erroneous and correct ones, and then reconstruct a 3D facial shape by using only the correctly extracted facial feature points. The experimental results showed that the 3D reconstruction performance of the proposed method was remarkably improved compared to that of the previous method which does not consider the errors of automatic facial feature point extraction.