최민호 교수
Minho Choi
KAIST 전산학부 · 공학
연구실 소개
최민호 교수의 연구실은 고체 상태에서의 이온 도전 메커니즘과 비평형 열역학적 현상에 대한 비파erturbative 고전적 모델링을 중심으로, 미세한 상호작용에서 비롯되는 에너지 소산과 열화학적 평형화 메커니즘을 수치적·이론적 방법으로 탐구합니다. 특히 입자-진동자 시스템에서의 에너지 분산과 열화상태 도달 과정을 분석하며, 이는 나노스케일 에너지 소산 및 열전도 메커니즘 이해에 기여합니다. 또한, 최근에는 대규모 언어 모델의 장기 일관성 문제를 해결하기 위한 그래프 기반 메모리 구조와 소형 언어 모델을 활용한 효율적 지식 추출 기법 등, 인공지능과 물리 모델링의 융합 연구도 진행 중입니다.
연구 현황
연구 성과 추이
표시된 성과는 수집된 데이터 기준으로 산출되며, 일부 차이가 있을 수 있습니다.
주요 논문
6Using a nonperturbative classical model, we numerically investigate the dynamics of mobile particles interacting with an infinite chain of harmonic oscillators, an abstraction of ionic conduction through solid-state materials. We show that coupling between the mobile particles and a single mass of the chain is sufficient to induce dissipation of the mobile particles' energy over a wide range of system parameters. When we introduce thermal fluctuations in the position of the chain mass, the mobil
While Large Language Models (LLMs) have advanced open-domain dialogue systems, maintaining long-term consistency remains a challenge due to inherent limitations in long-context reasoning and the inefficiency of processing extensive raw text. Existing approaches typically rely on either unstructured memory storage, which is prone to information loss, or computationally expensive LLMs that incur high latency. To address these limitations, we propose G-Long, a graph-enhanced framework that utilizes
Scalar dissipation rate (SDR) evolution in a stopping turbulent jet was analysed using direct numerical simulations and a theoretical approach. After the jet is stopped, a deceleration wave for the SDR propagates downstream with a speed similar to that for axial velocity. Upstream of the deceleration wave, mean centreline SDR becomes proportional to axial distance, and inversely proportional to the square of time. After passing of the deceleration wave, normalised radial profiles of SDR and its
Using a non-perturbative classical model, we numerically investigate the dynamics of mobile particles interacting with an infinite chain of harmonic oscillators, an abstraction of ionic conduction through solid-state materials. We show that coupling between the mobile particles and a single mass of the chain is sufficient to induce dissipation of the mobile particles' energy over a wide range of system parameters. When we introduce thermal fluctuations in the position of the chain mass, the mobi
This study examines how recess length affects supercritical kerosene/oxygen combustion in a bi-swirl injector using Reynolds-averaged Navier–Stokes simulations at 200 bar. Five recess ratios (RR = 1.3–3.1) are analyzed to assess their impact on flame stabilization, swirl decay, pressure drop, thermal loading, and combustion efficiency. Real-fluid behavior is captured using the SRK equation of state with high-pressure transport-property corrections. Results show that RR 1.3–2.5 support a diffusio
대표 연구 분야
최민호 교수의 연구를 Nubint에서 더 깊이 살펴보세요
이 연구실의 논문을 앱에서 열어 AI와 함께 읽고, 핵심을 요약하고, 내 글에 인용하세요.