DEM-CFD analysis and gas adsorption modeling of activated carbon filter for fuel cell electric vehicles
- 주제(키워드) 활성탄 필터(Activated carbon filter) , DEM-CFD 연계(DEM-CFD coupling) , 표면 밀도(Surface area density) , Freundlich isotherm-LDF 흡착 모델(Freundlich isotherm-LDF adsorption model) , 복합 목적 최적화(Multi- objective optimization)
- 발행기관 서강대학교 일반대학원
- 지도교수 강성원
- 발행년도 2026
- 학위수여년월 2026. 2
- 학위명 석사
- 학과 및 전공 일반대학원 기계공학과
- 실제URI http://www.dcollection.net/handler/sogang/000000082456
- UCI I804:11029-000000082456
- 본문언어 한국어
- 저작권 논문은 저작권에 의해 보호받습니다.
초록(요약문)
Activated carbon (AC) filters used in air-intake systems of fuel cell electric vehicles (FCEVs) desirably achieve high gas adsorption efficiency and small pressure drop, which are often conflicting requirements. In this study, a two-step numerical framework is proposed to evaluate and optimize the performance of a honeycomb-shaped gas filter cell with cylindrical AC pellets. First, a coupled discrete element method (DEM)-computational fluid dynamics (CFD) simulation approach is developed to investigate the effects of key geometric parameters of the AC pellets (AC volume fraction (VF), particle aspect ratio (AR), number of particles (NP) in a filter cell, and packing structure) on the pressure drop and total AC surface area that is proportional to the number of surface active sites. Second, a novel porous-media-based adsorption model is developed by combining the Freundlich isotherm and a modified linear driving force (LDF) model employing the AC surface area density (SAD) to reflect the effects of VF, AR, and NP on the breakthrough time. A parametric study with total 896 cases of DEM-CFD simulations shows that VF is the most dominant parameter on both the pressure drop and AC surface area. These two performance indices are significantly affected also by the particle shape parameters (AR, NP) and packing structure via the AC surface area, frontal area, and flow non-uniformity inside a filter cell. In general, both the pressure drop and AC surface area are proportional to AR and NP, which suggests a possible optimum between them. A multi-objective optimization problem is defined using a weighted composite function of the normalized pressure drop and AC surface area. In a range of the weight, it is found that there exists an optimal (AR, NP) combination, while a Pareto front analysis shows the optimum shifts markedly with the weight. The proposed adsorption model reflecting the geometric effects via SAD shows the effect of each model parameter reasonably well. The model parameters are tuned to reproduce an experimental NH3 breakthrough curve, showing both the breakthrough and saturation times with good accuracy. When this model is applied to a full-scale snorkel duct configuration, it reduces the computational cost by more than three orders of magnitude compared to a full-scale simulation resolving the AC particle geometry. The present framework provides a practical and scalable approach for optimization of honeycomb-shaped AC gas filters in FCEV air-intake systems.
more목차
1 Introduction 1
2 Simulation details 4
2.1 Computational domain and boundary conditions 4
2.2 Numerical methods and governing equations 6
2.2.1 DEM-CFD simulation approach 6
2.2.2 Gas adsorption modeling for porous-media-based simulation 12
2.2.3 Adsorption isotherm and LDF model 13
3 Results and discussion 15
3.1 Effects of AC pellet and packing geometry on pressure drop 15
3.2 Effects of AC pellet and packing geometry on total AC surface area 19
3.3 Multi-objective optimization 22
3.4 Parametric study of adsorption model 24
3.4.1 Effects of adsorption model parameters on breakthrough curve 24
3.4.2 Effects of AC geometric parameters on breakthrough curve 26
3.4.3 Model fitting of NH3 adsorption experimental data 28
3.5 Adsorption of multiple species 30
4 Conclusions 35
