• Journal of the Korea Society of Computer and Information
• /
• v.25 no.7
• /
• pp.39-46
• /
• 2020

Recently, for designing virtual characters in the battle game field effectively, some methods are very needed to predicate the win rates of the battle of them efficiently. In this paper, we propose a method to solve this problem by combining simulation and machine learning. Firstly, a simulation is used to analyze the win rates of the battle of virtual characters in the battle game. In addition, we apply a regression model based machine learning scheme to predict win rates of the battle of virtual characters according to their abilities. Our experimental results using suggested method show that it is almost no difference between the win rates of the simulation and the prediction results using the machine learning scheme. And also, we can obtain good performance in the experiment using only simple regression based machine learning model.

• Journal of Korean Society on Water Environment
• /
• v.23 no.6
• /
• pp.951-960
• /
• 2007

Window interface to Hydrological Simulation Program-FORTRAN (WinHSPF) developed by the United States Environmental Protection Agency (EPA) was applied to the upstream of Nam-Han river watershed to examine its applicability for loading estimates in watershed scale and to evaluate non-point source control scenarios using BMPRAC in WinHSPF. The WinHSPF model was calibrated and verified for water flow using Ministry of Construction and Transportation (MOCT, 3 stations, 2003~2005) and water qualities using Ministry of Environment (MOE, 5 station, 2000~2006). Water flow and water quality simulation results were also satisfactory over the total simulation period. But outliers were occurred in the time series data of TN and TP at some regions and periods. Therefore, it required more profit calibration process for more various parameters. As a result, all the study was performed within the expectation considering the complexity of the watershed, pollutant sources and land uses intermixed in the watershed. The estimated pollutant load for annual average about $BOD_5$, T-N and T-P respectively. Nonpoint source loading had a great portion of total pollutant loading, about 86.5~95.2%. In WinHSPF, BMPRAC was applied to evaluate non-point source control scenarios (constructed wetland, wet detention ponds and infiltration basins). All the scenarios showed efficiency of non-point source removal. Overall, the HSPF model is adequate for simulating watersheds characteristics, and its application is recommended for watershed management and evaluation of best management practices.

• Journal of the Korean Institute of Educational Facilities
• /
• v.22 no.6
• /
• pp.23-28
• /
• 2015

• The Korean Publising Journal, Monthly
• /
• s.368
• /
• pp.54-55
• /
• 2006

• Social Workers
• /
• no.5 s.37
• /
• pp.18-22
• /
• 2005

• Feed Journal
• /
• v.4 no.1 s.29
• /
• pp.92-95
• /
• 2006

• Proceedings of the Korea Water Resources Association Conference
• /
• 2001.05a
• /
• pp.557-562
• /
• 2001

• KOREAN POULTRY JOURNAL
• /
• v.41 no.4
• /
• pp.76-77
• /
• 2009

• Journal of Environmental Impact Assessment
• /
• v.21 no.1
• /
• pp.161-169
• /
• 2012

Recently various attempts have been made to apply HSPF model to calculate runoff and diffuse pollution loads of stream and reservoir watersheds. Because the role of standard flow is very important in the water quality modelling of Total Water Pollution Load Management, HSPF was used as a means of estimating standard flow. In this study, BASINS 4.0 and WinHSPF was applied to the Gomakwoncheon watershed, genetic algorithm(GA) and influence coefficient algorithm were used to calibrate the runoff parameters of the WinHSPF. The objective function is the sum of the squares of the normalized residuals of the observed and calculated flow and it is optimized using GA. Estimates of the optimum runoff parameters are made at each iteration of the influence coefficient algorithm. The calibration results showed a relatively good correspondence between the observed and the calculated values. The standard flow(Q275) of the Gomakwoncheon watershed was estimated using the ten years of weather data.

• Advances in nano research
• /
• v.1 no.2
• /
• pp.105-109
• /
• 2013

We report dry thermal development of negative resist polystyrene with low molecular weight. When developed on a hotplate at $350^{\circ}C$ for 30 min, polystyrene showed reasonable high contrast and resolution (30 nm half-pitch), but low sensitivity. Resist sensitivity was greatly improved at lower development temperatures, though at the cost of reduced contrast. In addition, we observed the thickness reduction due to thermal development was higher for larger remaining film thickness, implying the thermal development process is not just a surface process and the more volatile chains below the top surface may diffuse to the surface and get evaporated.