• Journal of Korea Water Resources Association
• /
• v.54 no.5
• /
• pp.301-309
• /
• 2021

In this study, we tried to improve the performance of the existing U-net-based deep learning rainfall prediction model, which can weaken the meaning of time series order. For this, ConvLSTM2D U-Net structure model considering temporal consistency of data was applied, and we evaluated accuracy of the ConvLSTM2D U-Net model using a RainNet model and an extrapolation-based advection model. In addition, we tried to improve the uncertainty in the model training process by performing learning not only with a single model but also with 10 ensemble models. The trained neural network rainfall prediction model was optimized to generate 10-minute advance prediction data using four consecutive data of the past 30 minutes from the present. The results of deep learning rainfall prediction models are difficult to identify schematically distinct differences, but with ConvLSTM2D U-Net, the magnitude of the prediction error is the smallest and the location of rainfall is relatively accurate. In particular, the ensemble ConvLSTM2D U-Net showed high CSI, low MAE, and a narrow error range, and predicted rainfall more accurately and stable prediction performance than other models. However, the prediction performance for a specific point was very low compared to the prediction performance for the entire area, and the deep learning rainfall prediction model also had limitations. Through this study, it was confirmed that the ConvLSTM2D U-Net neural network structure to account for the change of time could increase the prediction accuracy, but there is still a limitation of the convolution deep neural network model due to spatial smoothing in the strong rainfall region or detailed rainfall prediction.

• Atmosphere
• /
• v.21 no.2
• /
• pp.197-208
• /
• 2011

The objective of this study is to describe the short-range forecast system of the Korea Air Force (KAF) and to verificate its performace in 2009 summer. The KAF weather prediction model system, based on the Weather Research and Forecasting (WRF) model (i.e., the KAF-WRF), is configured with a parent domain overs East Asia and two nested domains with the finest horizontal grid size of 2 km. Each domain covers the Korean peninsula and South Korea, respectively. The model is integrated for 84 hour 4 times a day with the initial and boundary conditions from National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) data. A quantitative verification system is constructed for the East Asia and Korean peninsula domains. Verification variables for the East Asia domain are 500 hPa temperature, wind and geopotential height fields, and the skill score is calculated using the difference between the analysis data from the NCEP GFS model and the forecast data of the KAF-WRF model results. Accuracy of precipitation for the Korean penisula domain is examined using the contingency table that is made of the KAF-WRF model results and the KMA (Korea Meteorological Administraion) AWS (Automatic Weather Station) data. Using the verification system, the operational model and parallel model with updated version of the WRF model and improved physics process are quantitatively evaluated for the 2009 summer. Over the East Aisa region, the parallel experimental model shows the better performance than the operation model. Errors of the experimental model in 500 hPa geopotential height near the Tibetan plateau are smaller than errors in the operational model. Over the Korean peninsula, verification of precipitation prediction skills shows that the performance of the operational model is better than that of the experimental one in simulating light precipitation. However, performance of experimental one is generally better than that of operational one, in prediction.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1D
• /
• pp.67-72
• /
• 2006

Temperature prediction in asphalt pavements is the one of most important factors for estimating the pavement response and predicting the pavement performance in the mechanistic-empirical pavement design. A study on temperature prediction procedure with variation of time and depth in asphalt pavements was conducted using field monitoring data. After selecting the temperature monitoring sections, the temperature sensors have been installed in different depths and the temperature data have been collected in every one hour. The developed pavement temperature prediction model was calibrated using field monitoring temperature data. The predicted temperatures were compared with measured temperatures at different seasons in selected sections. The results showed that the solar absorptivity and emissivity values in the fall is different from the values in other seasons. The predicted temperatures agree well with the measured temperatures at a wide range of temperatures. The temperature differences between each other fall in the range of ${\pm}3^{\circ}C$. It is also found that the regional characteristics did not affect the temperature prediction procedure.

• Earthquakes and Structures
• /
• v.26 no.2
• /
• pp.97-115
• /
• 2024

In this paper, a database consisting of the dynamic shear modulus ratio and damping ratio test data of clay obtained from 406 groups of triaxial tests is constructed with the starting area of Xiong'an New Area as the research background. The aim is to study the nonlinear dynamic properties of clay in this area under cyclic loading. The study found that the effective confining pressure and plasticity index have certain influences on the dynamic shear modulus ratio and damping ratio of clay in this area. Through data analysis, it was found that there was a certain correlation between effective confining pressure and plasticity index and dynamic shear modulus ratio and damping ratio, with fitting degree values greater than 0.1263 for both. However, other physical indices such as the void ratio, natural density, water content and specific gravity have only a small effect on the dynamic shear modulus ratio and the damping ratio, with fitting degree values of less than 0.1 for all of them. This indicates that it is important to consider the influence of effective confining pressure and plasticity index when studying the nonlinear dynamic properties of clays in this area. Based on the above, prediction models for the dynamic shear modulus ratio and damping ratio in this area were constructed separately. The results showed that the model that considered the combined effect of effective confining pressure and plasticity index performed best. The predicted dynamic shear modulus ratio and damping ratio closely matched the actual curves, with approximately 88% of the data falling within ±1.3 times the measured dynamic shear modulus ratio and approximately 85.1% of the data falling within ±1.3 times the measured damping ratio. In contrast, the prediction models that considered only a single influence deviated from the actual values, particularly the model that considered only the plasticity index, which predicted the dynamic shear modulus ratio and the damping ratio within a small distribution range close to the average of the test values. When compared with existing prediction models, it was found that the predicted dynamic shear modulus ratio in this paper was slightly higher, which was due to the overall hardness of the clay in this area, leading to a slightly higher determination of the dynamic shear modulus ratio by the prediction model. Finally, for the dynamic shear modulus ratio and damping ratio of the engineering site in the starting area of Xiong'an New Area, we confirm that the prediction formulas established in this paper have high reliability and provide the applicable range of the prediction model.

• Proceedings of The Korean Society of Agricultural and Forest Meteorology Conference
• /
• 2019.08a
• /
• pp.253-254
• /
• 2019

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.24 no.11
• /
• pp.833-839
• /
• 2014

The shooting noise caused by shooting training, which has strength and impacts, is becoming a serious damage to the residents around the shooting range and, consequently, the number of civil appeals against the shooting noise is on the constant increase. For this reason, the military examines the effects of the shooting noise at the stage of design in constructing a shooting range and tries to build a soundproof shooting range to minimize civil appeals. However, the lack of research and data concerning propagation and attenuation, both of which characterize the shooting noise from within a soundproof shooting range, even makes it so difficult to design a soundproof shooting range in constructing it. So this study used an acoustic simulation in a soundproof shooting range to identify acoustic and propagation characteristics within the shooting range and, on this basis, predicted the noise level at an exit of the soundproof shooting range. As a result, if the form and specifications of a soundproof shooting range were decided on at the stage of design, it was possible to use a simulation to design a soundproof shooting range with optimized acoustic performance and, on this basis, to predict a sound pressure level at an exit of the soundproof shooting range. On the basis of these data, it is probably possible to determine the degree of the effects of the shooting noise on the villages around a shooting range and the extent of damage to it and to minimize civil appeals against the shooting noise and resolve the issues of compensation and agreement with ease. This study is expected to provide useful data for designing and constructing a similar soundproof shooting range.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.22 no.5
• /
• pp.685-697
• /
• 1998

Series of recent k-.epsilon. model modification have been carried out with the aid of DNS data to include the effect of near wall. Though these methods opened new way of turbulence modelings, newly developed turbulence models of its kind had yet shortcomings in prediction for the turbulent flows with various Reynolds numbers and various geometric conditions. As a remedy for these shortcomings, a new k-.epsilon. model proposed here by improving the dissipation rate equation and the damping function for eddy viscosity model. The new dissipation rate equation was modeled based on the energy spectrum and magnitude analysis. The damping function for eddy viscosity was also formulated on the ground of distribution of dissipation rate length scales near a wall and the DNS data. The new k-.epsilon. model was applied to the fully developed turbulent flows in a channel and a pipe with a wide range of Reynolds numbers. Prediction results showed that the present model represents properly the turbulence properties in all turbulent regions over a wide range of Reynolds numbers.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.14 no.3
• /
• pp.131-140
• /
• 2015

The purpose of this study is to predict the fatigue life of a planet carrier of a slewing reducer for a tower crane. To predict the fatigue life of the carrier, the inertia endurance test was carried out, and then the input torque profile for the reducer was obtained. The load profile acting on the planet pins that assembled the carrier was calculated from the measured input torque profile using commercial gearbox analysis software. The stress profiles of the carrier weak points were analyzed from the calculated load profile and boundary conditions using commercial FE software, and the stress cycles were determined using the rainflow counting method. Finally, the fatigue life of the carrier was predicted using the equivalent stress range by considering the effect of mean stress, and an S-N curve was drawn up using the GL guideline and the cumulative damage law.

• The Journal of Korean Association of Computer Education
• /
• v.14 no.4
• /
• pp.63-70
• /
• 2011

Recommender systems, which predict and recommend items that may possibly draw users' interests, have been applied in various fields as e-commerce systems are widespread. Collaborative filtering, one of the major methodologies of recommender systems, recommends either items similar to those preferred by the user, or items preferred by the other similar user. Therefore, two problems determine its performance; one is correct estimation of similarity and the other is predicting the real rating of the recommended item. This study addresses the latter problem. Previous studies predict the real rating based on the mean of the ratings, but this study proposes a prediction based on the range of the ratings and investigates its performance through experiments. As a result, it is demonstrated that the proposed method improves the mean absolute error significantly, compared to the previous method.

• Journal of Broadcast Engineering
• /
• v.22 no.2
• /
• pp.240-252
• /
• 2017

Multi-view video codec based on the High Efficiency Video Coding (MV-HEVC) has high encoding complexity because it exploits an additional reference picture for disparity compensation prediction (DCP) when the picture of dependent view is encoded. In this paper, we propose an efficient method to reduce the complexity of disparity motion vector search for the MV-HEVC. The proposed method includes the initial search point decision method using affine transform and the adaptive search range decision method. The simulation results show that the proposed method reduces the complexity of disparity motion vector search up to 90.78% with negligible coding efficiency degradation. Also the results show that the proposed method outperforms other conventional techniques reducing complexity.