Embedded Files
“Climate change necessitates innovations in computational modeling to improve the design and operation of infrastructure systems, which are impacted by, and in turn impact, climate change.”
Vision statement
Vision statement
Climate change, the defining challenge of our time, necessitates innovations in computational modeling to improve the design and operation of infrastructure systems, which are impacted by, and in turn impact, climate change. The increasing severity and frequency of extreme weather events directly affect safety and resilience. At the same time, the full life cycle of infrastructure—from design and material manufacturing to construction, usage, and demolition—accounts for approximately 40 percent of global emissions, making us one of the highest-emitting sectors.
The EISS Lab @ Hongik University is committed to advancing computational frameworks that support both climate adaptation and mitigation. Our mission is to enhance the resilience of infrastructure while accelerating the deployment of renewable energy systems. We develop efficient algorithms that improve the prediction and optimization of system lifespans and energy generation, all while reducing computational burden.
By integrating structural engineering, macro energy system modeling, and AI-driven enhancement and acceleration, we take a multidisciplinary approach centered on three key thrusts.
Thrust 1. Design for Climate
Thrust 1. Design for Climate
We develop advanced risk assessment methodologies to model nonstationary climate processes, enabling more accurate prediction of future climate-related risks compared to traditional approaches. Leveraging AI-driven data science techniques, we build computational frameworks that are both time- and data-efficient, allowing for rapid estimation of long-term environmental impacts and life cycle assessments (LCA) in large-scale renewable energy systems, such as floating offshore wind farms.
Thrust 2. O&M Optimization
Thrust 2. O&M Optimization
We investigate the use of partially observable Markov decision processes (POMDPs) integrated with reinforcement learning to optimize the scheduling of operations and maintenance (O&M) activities in multi-sectoral renewable energy systems. While integrated energy systems benefit from shared infrastructure and resources, their multifunctional nature introduces complexity in operational decision-making. Our approach aims to minimize O&M costs while maximizing crew safety through adaptive, data-driven scheduling strategies.
Thrust 3. IDS of MES
Thrust 3. IDS of MES
We develop computational models to assess the role of emerging clean energy technologies—such as floating offshore wind turbines and thermal energy storage systems—alongside established resources including solar, onshore wind, and hydropower. Applied within macro-energy systems modeling, these frameworks identify optimal configurations for generation assets and long-duration energy storage (LDES). In collaboration with the MIT Energy Initiative, our work focuses on evaluating the techno-economic feasibility of diverse technology portfolios to support data-driven decision-making for future large-scale energy system deployment in Korea.
Page updated
Google Sites
Report abuse