Three Outstanding Paper Awards at the 2026 Korean Society of Civil Engineers Convention / Research Team Led by Professor Shin Ju-Young (Department of Construction Systems Engineering)

  • 26.11.18 / 홍유민

A research team led by Professor Shin Ju-Young from the Department of Construction Systems Engineering at the Graduate School of Kookmin University (President Jeong Seung Ryul) won three Outstanding Paper Awards at the “2026 Korean Society of Civil Engineers (KSCE) Convention.”
 
The recipients of these awards are the research team, including Ph.D. candidate Park Ji-yeon and master’s candidates Gil Jun-young and Jo Su-bin from the Department of Construction Systems Engineering. They conducted research utilizing artificial intelligence (AI) and hydraulic and hydrological analysis techniques to address the risks of extreme rainfall and urban flooding caused by climate change. Their research achievements were recognized in the fields of urban flooding characteristic analysis, high-resolution rainfall data generation, and sewer water level prediction, respectively.
 
First, the research team led by master’s student Gil Jun-young conducted a study titled “Analysis of Flood Response Characteristics in Urban Dual Drainage Systems in Response to Changes in Rainfall.” The team developed an analytical model linking the underground drainage network with the ground surface in the Mulgeum area of Yangsan City, South Gyeongsang Province, and analyzed urban flooding patterns according to rainfall intensity. The results revealed that while the burden on the drainage network increases continuously as rainfall intensity rises, surface flooding expands significantly only when rainfall exceeds a certain threshold. This finding demonstrates that urban flooding may not increase in simple proportion to rainfall intensity and is expected to be utilized in future assessments of urban flood risk based on rainfall conditions.
 
The research team led by master’s student Jo Su-bin received the award for their study titled “Evaluation of Extreme Rainfall Reproducibility through the Combination of Maximum Value Models for Deep Learning-Based Temporal Downscaling of Rainfall Time Series.” The team developed an AI-based model to convert hourly observed rainfall data into high-resolution 10-minute intervals and sought to enhance reproducibility by combining it with a model specifically trained on extreme rainfall events. Performance evaluations conducted at six rainfall observation stations nationwide confirmed that even when models trained on data from different observation sites were applied, they could reproduce rainfall characteristics at a similar level. This research is expected to contribute to estimating extreme rainfall and improving the accuracy of urban flooding analyses, even in areas where high-resolution observational data is scarce.
 
A research team led by Ph.D. candidate Park Ji-yeon received the Best Paper Award for their work titled “Development of a Sewer Water Level Prediction Technique Using a Gate-Based Heterogeneous Domain Fusion AI Model.” The team developed an AI model that effectively combines observational data with physical model predictions and applied it to predict water levels in the vertical shafts of the Sinwol Stormwater Retention and Drainage Facility in Seoul. As a result, they improved predictive performance compared to existing models that relied solely on observational data and demonstrated the potential to predict water levels using only observational data during the actual prediction stage by incorporating physical model information into the pre-training process. This is expected to be utilized for real-time water level prediction to address urban flooding and for the efficient operation of stormwater retention facilities.
 
The three award-winning studies are significant in that they each demonstrated the potential for advancing urban flooding response technologies through distinct approaches: the precise generation of rainfall data, the analysis of flooding characteristics in urban drainage systems, and the prediction of sewer water levels. In particular, by integrating artificial intelligence with existing hydraulic and hydrological analysis techniques, they laid the groundwork for more precise analysis and prediction of urban flooding risks caused by climate change.
 
Kookmin University plans to continue strengthening its research capabilities to address climate change and urban disasters through interdisciplinary research combining artificial intelligence with the fields of construction and the environment.

▲ Park Ji-yeon (Ph.D. candidate)

▲ Cho Su-bin (Master’s student)

▲ Gil Jun-young (Master’s student)

This content is translated from Korean to English using the AI translation service DeepL and may contain translation errors such as jargon/pronouns.

If you find any, please send your feedback to kookminpr@kookmin.ac.kr so we can correct them.

 

View original article [click]

Three Outstanding Paper Awards at the 2026 Korean Society of Civil Engineers Convention / Research Team Led by Professor Shin Ju-Young (Department of Construction Systems Engineering)

A research team led by Professor Shin Ju-Young from the Department of Construction Systems Engineering at the Graduate School of Kookmin University (President Jeong Seung Ryul) won three Outstanding Paper Awards at the “2026 Korean Society of Civil Engineers (KSCE) Convention.”
 
The recipients of these awards are the research team, including Ph.D. candidate Park Ji-yeon and master’s candidates Gil Jun-young and Jo Su-bin from the Department of Construction Systems Engineering. They conducted research utilizing artificial intelligence (AI) and hydraulic and hydrological analysis techniques to address the risks of extreme rainfall and urban flooding caused by climate change. Their research achievements were recognized in the fields of urban flooding characteristic analysis, high-resolution rainfall data generation, and sewer water level prediction, respectively.
 
First, the research team led by master’s student Gil Jun-young conducted a study titled “Analysis of Flood Response Characteristics in Urban Dual Drainage Systems in Response to Changes in Rainfall.” The team developed an analytical model linking the underground drainage network with the ground surface in the Mulgeum area of Yangsan City, South Gyeongsang Province, and analyzed urban flooding patterns according to rainfall intensity. The results revealed that while the burden on the drainage network increases continuously as rainfall intensity rises, surface flooding expands significantly only when rainfall exceeds a certain threshold. This finding demonstrates that urban flooding may not increase in simple proportion to rainfall intensity and is expected to be utilized in future assessments of urban flood risk based on rainfall conditions.
 
The research team led by master’s student Jo Su-bin received the award for their study titled “Evaluation of Extreme Rainfall Reproducibility through the Combination of Maximum Value Models for Deep Learning-Based Temporal Downscaling of Rainfall Time Series.” The team developed an AI-based model to convert hourly observed rainfall data into high-resolution 10-minute intervals and sought to enhance reproducibility by combining it with a model specifically trained on extreme rainfall events. Performance evaluations conducted at six rainfall observation stations nationwide confirmed that even when models trained on data from different observation sites were applied, they could reproduce rainfall characteristics at a similar level. This research is expected to contribute to estimating extreme rainfall and improving the accuracy of urban flooding analyses, even in areas where high-resolution observational data is scarce.
 
A research team led by Ph.D. candidate Park Ji-yeon received the Best Paper Award for their work titled “Development of a Sewer Water Level Prediction Technique Using a Gate-Based Heterogeneous Domain Fusion AI Model.” The team developed an AI model that effectively combines observational data with physical model predictions and applied it to predict water levels in the vertical shafts of the Sinwol Stormwater Retention and Drainage Facility in Seoul. As a result, they improved predictive performance compared to existing models that relied solely on observational data and demonstrated the potential to predict water levels using only observational data during the actual prediction stage by incorporating physical model information into the pre-training process. This is expected to be utilized for real-time water level prediction to address urban flooding and for the efficient operation of stormwater retention facilities.
 
The three award-winning studies are significant in that they each demonstrated the potential for advancing urban flooding response technologies through distinct approaches: the precise generation of rainfall data, the analysis of flooding characteristics in urban drainage systems, and the prediction of sewer water levels. In particular, by integrating artificial intelligence with existing hydraulic and hydrological analysis techniques, they laid the groundwork for more precise analysis and prediction of urban flooding risks caused by climate change.
 
Kookmin University plans to continue strengthening its research capabilities to address climate change and urban disasters through interdisciplinary research combining artificial intelligence with the fields of construction and the environment.

▲ Park Ji-yeon (Ph.D. candidate)

▲ Cho Su-bin (Master’s student)

▲ Gil Jun-young (Master’s student)

This content is translated from Korean to English using the AI translation service DeepL and may contain translation errors such as jargon/pronouns.

If you find any, please send your feedback to kookminpr@kookmin.ac.kr so we can correct them.

 

View original article [click]

TOP