• 상하수도학회지
• /
• 제36권4호
• /
• pp.239-248
• /
• 2022

The prediction of algal bloom is an important field of study in algal bloom management, and chlorophyll-a concentration(Chl-a) is commonly used to represent the status of algal bloom. In, recent years advanced machine learning algorithms are increasingly used for the prediction of algal bloom. In this study, XGBoost(XGB), an ensemble machine learning algorithm, was used to develop a model to predict Chl-a in a reservoir. The daily observation of water quality data and climate data was used for the training and testing of the model. In the first step of the study, the input variables were clustered into two groups(low and high value groups) based on the observed value of water temperature(TEMP), total organic carbon concentration(TOC), total nitrogen concentration(TN) and total phosphorus concentration(TP). For each of the four water quality items, two XGB models were developed using only the data in each clustered group(Model 1). The results were compared to the prediction of an XGB model developed by using the entire data before clustering(Model 2). The model performance was evaluated using three indices including root mean squared error-observation standard deviation ratio(RSR). The model performance was improved using Model 1 for TEMP, TN, TP as the RSR of each model was 0.503, 0.477 and 0.493, respectively, while the RSR of Model 2 was 0.521. On the other hand, Model 2 shows better performance than Model 1 for TOC, where the RSR was 0.532. Explainable artificial intelligence(XAI) is an ongoing field of research in machine learning study. Shapley value analysis, a novel XAI algorithm, was also used for the quantitative interpretation of the XGB model performance developed in this study.

• 한국해양공학회지
• /
• 제6권2호
• /
• pp.133-142
• /
• 1992

Beach evolution is of the most important problem is the coastal engineering. Especially, the structure construction through reclamation in the shallow water region nesar the beach will cause many severe problems around the structure. Beach evolution due to the construction of an artificial island in this study was studied using wave transform model and associated of an artificial island in this study was studied using wave transform model and associated sediment transport model. Numerical simulation of the model was applied to the Kwangan beach using the data of waves and shoreline of the area. The combined wave transform model and beach evolution model showed good results. The results show a breakwater will be needed to prevent severe erosion near the eastward Kwangan beach when construction an artificial island in the Suyong Bay. Good results of the study also suggest that the present model can be more widely applied to the prediction of beach evolution.

• 한국실내디자인학회:학술대회논문집
• /
• 한국실내디자인학회 1999년도 춘계학술발표대회 논문집
• /
• pp.133-136
• /
• 1999

This study intends to develop an evaluation model which could be used as an effective tool for criticizing the model house of multi-family housing. The evaluation model has been developed based on the theoretical framework suggested by previous results of research related on hosing design, which comprises the following four components: 1) the paradigm of future home and life-style, 2) marketing effects of housing developer, 3) goals and principles of housing design, and 4) results of user-preference study. In this study, an evaluation model was suggested as a preliminary form which would be modified in detail after series of tests in the field. The evaluation model will be used provide standardized criteria for the quality of model houses, and eventually help to improve the quality of multi-family housing design by balanced information from theorist, user and developer.

• 한국환경과학회지
• /
• 제3권2호
• /
• pp.129-140
• /
• 1994

An intermediate atmosphere-ocean coupled model appropnate for the study of El Nino has been developed. The model is not only economic to use but also contains several most important physical processes. The geometrical effects which were not confided in the previous intermediate model study of Ahn (1990), are included in the model for more realistic simulation of the event. The results show that the individual models respond appropriately to the given boundary conditions. At the same time, in the coupled model experiment, ENSO-like oceanic and atmospheric anomalies are also well simulated under an external triggering similar to the initiation forcing of ENSO. It is expected that this type of model can be effectively used for the. study and simulation of El Nido. More improvement of modeling may be Possible after inclusion of subsequent processes such as inclusion of ocean mixed layer dynamics.

• 한국정보시스템학회지:정보시스템연구
• /
• 제28권1호
• /
• pp.133-153
• /
• 2019

Purpose The purpose of this study is to design a regional electricity planning model rather than the existing single region ones and verify its usefulness. The regional electricity planning model is to determine both electricity distribution among regions and power plant planning at the same time satisfying regional demands and distribution networks. Design/methodology/approach This study made a regional electricity planning model by integrating power plant planning and electricity distribution among regions. The regional electricity planning model is formulated into a linear programming problem, and coded and run using the OSeMOSYS, one of open source energy systems. Findings According to the empirical analysis result, this study confirmed that the regional electricity planning model proposed in this study deducts the unfairness among regions in view of electricity and green house gas. In addition, the model is expected to be used in evaluating and developing the national policies concerning fine dust and/or green house gas.

• Computers and Concrete
• /
• 제9권1호
• /
• pp.1-19
• /
• 2012

The size-effect study of various fracture parameters obtained from two parameter fracture model, effective crack model, double-K fracture model and double-G fracture model is presented in the paper. Fictitious crack model (FCM) for three-point bend test geometry for cracked concrete beam of laboratory size range 100-400 mm is developed and the different fracture parameters from size effect model, effective crack model, double-K fracture model and double-G fracture model are evaluated using the input data obtained from FCM. In addition, the fracture parameters of two parameter fracture model are obtained using the mathematical coefficients available in literature. From the study it is concluded that the fracture parameters obtained from various nonlinear fracture models including the double-K and double-G fracture models are influenced by the specimen size. These fracture parameters maintain some definite interrelationship depending upon the specimen size and relative size of initial notch length.

• 한국해양공학회지
• /
• 제27권4호
• /
• pp.33-44
• /
• 2013

IIn this study, a wave-surge-tide coupling numerical model was developed to consider nonlinear interaction. Then, this model was applied and calculations were made for a storm surge on the southeast coast. The southeast coast was damaged by typhoon "Maemi" in 2003. In this study, we used a nearshore wind wave model called SWAN (Simulating WAves Nearshore). In addition, the Meyer model was used for the typhoon model, along with an ocean circulation model called POM (Princeton Ocean Model). The wave-surge-tide coupling numerical model could calculate exact parameters when each model was changed to consider the nonlinear interaction.

• 음성과학
• /
• 제9권2호
• /
• pp.135-146
• /
• 2002

Generally a speaker verification system improves its system recognition ratio by regularizing log likelihood ratio, using a speaker model and its background speaker model that are required to be verified. The speaker-based cohort method is one of the methods that are widely used for selecting background speaker model. Recently, Gaussian-based cohort model has been suggested as a virtually synthesized cohort model, and unlike a speaker-based model, this is the method that chooses only the probability distributions close to basic speaker's probability distribution among the several neighboring speakers' probability distributions and thereby synthesizes a new virtual speaker model. It shows more excellent results than the existing speaker-based method. This study compared the existing speaker-based background speaker models and virtual speaker models and then constructed new virtual background speaker model groups which combined them in a certain ratio. For this, this study constructed a speaker verification system that uses GMM (Gaussin Mixture Model), and found that the suggested method of selecting virtual background speaker model shows more improved performance.

• Educational Technology International
• /
• 제17권1호
• /
• pp.1-37
• /
• 2016

The purpose of this study was to develop a MOOC design model that would improve the current practice of MOOC development in Korea by specifying easy-to-use course development procedures and guiding strategies. Following Richey and Klein (2007)'s conceptual model development procedure, the first step was to perform critical review of relevant literature and observe typical MOOC development processes. As a result, the initial model was developed. The second step was to conduct the expert review with five educational technology and MOOC researchers to secure the internal validity of the model. Based on the experts' suggestions, the model was revised and once again reviewed by the same experts. This process resulted in the development of the 2^nd version of model. The third step was to carry out external validation research in order to test the effectiveness, efficiency, and usability of the model. A basic model may be confirmed or corrected based on examination of its results. Consequently, the model was elaborated as the final model. In the final model, 6 procedural phases and 9 specific steps were included. The six procedural phases are: Analysis (1st Iteration), Design, Development (Course Development), Implementation, Evaluation, and Analysis (2nd Iteration), a slight variation of ADDIE model. The specific steps include: 1) Goal Setting, 2) Environment Analysis, 3) Content Design, 4) Style Design, 5) Course Development, 6) Implementation Plan, 7) Course Implementation, 8) Summative Evaluation, and 9) Need Reflection. The study concluded with suggestions for further research and application of the MOOC design model.

• 대한조선학회논문집
• /
• 제52권3호
• /
• pp.180-186
• /
• 2015

This study proposed a new method of preparing granular ethylene glycol/aliphatic detergent (EG/AD) model ice that has both strength and uniform thickness. Various sheets of granular model ice prepared in ice tanks are surveyed and their preparation procedures are analyzed. We not only made a new granular model ice using the EG/AD solution but also measured its thickness, strength, and density. In addition, we found that the strength of the model ice could be controlled by varying the time and air temperature in the consolidation phase. Based on the results of this study, we verified that granular EG/AD model ice can be prepared more uniformly and effectively than columnar EG/AD model ice. This study is intended to contribute to reducing the time required for the ice model test and the operation of the ice model basin.