The optimal design of steel plate girders has traditionally relied on meta-heuristic techniques, such as Genetic Algorithms (GA), to handle discrete design variables and complex non-linear constraints, including shear buckling and section classification. However, these methods suffer from high computational costs as they require repetitive re-optimization for every new load condition. To address this limitation, this study proposes a highly efficient Sequential Multi-Agent Reinforcement Learning (MARL) framework based on the Agent-Environment Cycle (AEC) architecture. Unlike parallel one-shot approaches, the proposed model effectively learns the dependencies between design variables by determining them sequentially. Furthermore, to maximize cost efficiency during the inference phase, we introduce an Adaptive Inference Chain combined with a deterministic DCR-based Shrink-Refine algorithm. Experimental results on 100 diverse load cases demonstrate that the proposed method achieves an average cost reduction of 8.2% compared to the GA baseline while maintaining 100% feasibility. With an inference time reduced to approximately 76 ms, the model demonstrates significant potential for real-time automated design. Additionally, an in-depth analysis of cases where the Demand-Capacity Ratio (DCR) fell short of the target clarifies the exploration limits within the discrete design space and validates the robustness of the algorithm.

A needle-free automatic injection syringe is a device that delivers drugs into the skin and tissues using a high-speed fluid jet without using an injection needle. This technology is attracting attention as an efficient means of vaccine delivery in the veterinary and livestock fields that reduce the risk of cross-infection and require mass vaccination. In particular, animal vaccination provides various advantages over conventional needle injection methods in terms of worker safety, inoculation speed, and maintenance cost. Among these drivers, Jet Injector Nozzle's flow path design is very important in needleless automatic injection syringes. This paper was conducted to solve the problem of pressure loss at the nozzle discharge side of the existing Jet Injector in designing the flow path of the animal vaccine-free automatic injection syringe nozzle. To this end, CAE was performed and the optimum design of the flow path required by the company was performed, and a large flow rate was possible in the optimal shape design, but this focuses on the nozzle flow path, which requires a review of design additions of cylinders and motors on the rear side.

To mitigate power transmission losses between power plants and consumers, Medium-Voltage Direct Current (MVDC) systems, which transmit power as DC and convert it back to AC at the destination, are increasingly being developed. However, these MVDC conversion systems, which are rich in semiconductor devices, generate significantly more heat than conventional transformers. Consequently, the thermal management capabilities of traditional air-cooling methods are proving insufficient, necessitating the adoption of liquid cooling systems. This study aims to optimize the energy efficiency of MVDC liquid cooling systems. Finally, we demonstrated a optimized system that minimizes the pump's power consumption by dynamically adjusting the coolant flow rate to each thermal module based on the automatic control valve system.

This study proposes a hierarchical optimization methodology for two-stage gear systems using Monte Carlo enhanced Genetic Algorithm (MCeGA). The approach integrates reliability-based design with genetic algorithms to overcome the inherent randomness of traditional GA methods. A two-phase optimization framework was developed. The system incorporates Unity engine for real-time 3D visualization and interactive design evaluation. Key design constraints including contact ratio, gear ratio, and meshing conditions were parameterized according to ISO 6336 and AGMA 2101 standards. The proposed framework enables application-specific optimal gear configurations through Pareto analysis and weighted optimization, providing engineers with practical design solutions for various industrial requirements.

In the present numerical study, the optimal design of cooling fans for cooling towers was investigated. The key design variables selected were the pitch angle and twist angle of the cooling fan, with the pitch angle ranging from 0° to 20°, and the twist angle from 0° to 10°. The objective was to develop a cooling fan capable of actively responding to varying operating conditions; to this end, the twist angle was implemented as a variable camber at the blade tip. The analysis results showed that, under identical pitch angle conditions, increasing the twist angle tended to reduce the stall phenomenon of the cooling fan. Optimization was performed using the pitch and twist angles as design variables, and lift, pressure drop and torque as response variables. Under the specified operating conditions, a combination of a 5° pitch angle and a 10° twist angle yielded the best performance.

연안 및 근해를 운항하는 천수선박은 프로펠러의 잠김 깊이가 부족하여 추진 효율이 저하되는 문제가 발생할 수 있다. 이를 개선하기 위해 선미 하부에 Y형 또는 I형 스트럿 구조물을 설치하여 프로펠러 축을 지지하고, 프로펠러를 보다 깊이 잠기도록 설계하는 방법이 널리 적용되고 있다. 그러나 스트럿 베어링의 적용 방식과 위치는 추진축계의 유연성 및 진동 특성에 큰 영향을 미치므로, 초기 설계단계에서 적절한 축계 배치가 필수적이다. 본 연구에서는 스트럿 구조를 갖는 선박을 대상으로, 베어링의 수와 배치에 따른 추진축계 의 정렬 특성과 횡진동 거동을 분석하였다. Y형 스트럿 베어링과 선미관 베어링이 조합된 기존 축계는 높은 횡방향 유연성을 확보하였으 나, 프로펠러 블레이드 통과 주파수와 고유진동수가 근접하여 공진 위험이 존재함을 확인하였다. 이를 해결하기 위해 선미관 베어링을 제 거하고, I형 스트럿 베어링으로 대체하는 새로운 베어링 배치를 제안하였으며, 반력영향계수와 진동모드 해석을 통해 최적의 설치 위치를 도출하였다. 해석 결과, I형 스트럿 베어링을 FR.10~FR.11 구간에 배치하는 경우 정렬 안정성과 공진 회피 측면에서 가장 우수한 성능을 나타내었다. 본 연구는 스트럿 베어링을 갖는 천수선박의 추진축계 설계 시, 유연성과 진동 특성을 동시에 고려한 최적 베어링 배치 가이 드라인을 제시한다는 점에서 의의가 있다.

선박용 프로펠러는 선박 추진 성능과 연비에 직접적인 영향을 미치는 핵심 부품으로, 제작 과정에서 높은 정밀도가 요구된다. 사형주조는 복잡한 형상의 금속 부품 제작에 널리 사용되는 공정이지만, 주조 과정에서 발생하는 열적 팽창과 냉각 수축은 최종 치수 오 차와 가공 비용 증가를 초래하는 주요 원인이다. 본 연구에서는 사형주조 과정에서 발생하는 열팽창 및 수축 현상을 정밀하게 예측하고, 이를 고려한 최적의 치수 여유 설정을 통해 연마 작업을 최소화하는 설계 방안을 제안하였다. 알루미늄 청동 합금(ALBC3)을 사용한 프로 펠러를 대상으로 열팽창 공식과 유한요소해석(FEM)을 적용하여 블레이드, 허브, 전체 지름 등 각 부위별 변형을 정량적으로 분석하였다. 분석 결과, 블레이드 너비와 두께는 약 1.9%, 허브 직경은 1.5%, 전체 지름은 2.0%의 여유를 두는 것이 적절한 것으로 나타났다. 이러한 최 적 치수 여유를 적용한 결과, 최대 23kg의 재료 절감, 30만 원 이상의 제작 비용 절감, 작업 시간 50~60% 단축 등의 정량적 개선 효과가 확인되었다. 최적 설계를 적용함으로써 추가 연마 작업과 재료 손실을 줄일 수 있으며, 이에 따른 비용 절감 효과도 기대된다. 본 연구 결 과는 선박용 프로펠러 제작 과정의 품질 향상과 생산성 제고에 기여할 수 있을 것으로 판단된다.

Tuned Mass Dampers (TMDs) are widely used to mitigate structural vibrations in buildings and bridges. However, conventional optimization methods often struggle to achieve optimal performance due to the complexity of structural dynamics. This study proposes the NN-L-BFGS-B algorithm, which combines Artificial Neural Networks (ANNs) for global exploration and L-BFGS-B for local exploitation to efficiently optimize TMD parameters. A ten-story shear-building model with a TMD is used for validation. The proposed method achieves the lowest H₂ norm compared to previous studies, demonstrating improved optimization performance. Additionally, NN-L-BFGS-B effectively balances computational efficiency and accuracy, making it adaptable to various engineering optimization problems.

Water-soluble substances like hydrogen fluoride, generated in semiconductor manufacturing, pose serious health and environmental risks, underscoring the need for effective capture devices. Vertical liquid capture devices help by aggregating and discharging hazardous substances with water, but their design can lead to backflow during abnormal operations, causing unintended releases and impacting efficiency and safety. This study seeks to improve a vertical liquid collection device’s containment performance by optimizing its geometry. The vertical wall was rotated at various angles and directions, and turbulent kinetic energy and streamline distribution were analyzed to assess vortex formation and flow characteristics. These structural modifications identify optimal conditions to control hazardous substance migration, offering insights for future pollutant removal device designs.

The multi-local resonance metamaterial is based on the local resonance mechanism of resonators, effectively blocking wave propagation within multiple resonant frequency ranges, a phenomenon known as band gaps. In practical applications for vibration reduction, the goal is to achieve wide-band vibration attenuation at low frequencies. Therefore, this study aims to improve the vibration reduction performance of multi-local resonance metamaterials by lowering the band gap frequency and expanding the band gap width. To achieve this, an objective function was formulated in the optimization problem, considering both the frequency and width of the band gap, with the geometric shapes of the multiple local resonators selected as design variables. The Sequential Quadratic Programming (SQP) technique was employed for optimization. The results confirmed that the band gap was generated at lower frequencies and that the band gap width was expanded.

전 세계적으로 보전 성과 및 효과평가 필요성, 신뢰도 높은 지표 측정 방법 등이 강조되고 있다. 식물체 높이는 재도입 등 보전이입 및 생태계 복원 시 중요한 지표로 활용된다. 정확한 모집단 추정을 위한 세밀한 표본 설계는 연구 진행에 앞서 가장 중요한 과정 중 하나이다. 이 연구는 멸종위기 야생생물 백양더부살이의 전국 분포, 개체군 현황, 생장 및 번식 특성, 식물체 높이의 지역 개체군 내 또는 개체군 간 변이를 파악하고, 보전·복원사업 효과평가에 필요한 효율적이고도 신뢰도 높은 생장 지표(즉, 식물체 높이) 모니터링 기반을 구축하고자 수행되었다. 백양더부살이 는 탐색한 전국 16곳 중 7곳에서 확인되었으며, 정읍 1지역 개체군이 국내 최대 규모 개체군이었다(N = 636). 이중 3개 지역(정읍 2지역, 제주 1지역)에 대해 식물체 높이 전수조사를 수행한 결과(N = 947), 높이의 범위는 3-50 cm였고 중앙값은 20 cm였다. 제주 지역 개체군의 평균 높이(22.1 ± 6.7 cm)는 가장 컸고, 무리(clump)당 개체수(중앙값 = 5)도 가장 많았다. 전수조사 자료를 바탕으로 표본추출 강도별 두 가지 방법(1. 집락표본추출, 2. 이단표본추출)으로 부트스트랩 복원추출을 수행한 결과, 집락표본추출 결과가 변동성, 정확도, 정밀도 측면에서 우수했으며, 무리를 표본추 출단위로 하고 표본추출 강도를 60% 이상으로 할 때 신뢰도 높은 모집단 추정이 가능했다. 이 연구는 멸종위기 야생생 물 보전·복원사업 이전에 충분한 기초정보 수집과 명확한 표본 설계를 통한 모니터링 체계 구축의 중요성을 강조한다.

The purpose of this study was to derive an optimal mix design for bridge deck pavements that can compensate for the limitations of latexmodified concrete (LMC). To address the limitations of LMC, this paper proposes the incorporation of silica fume into LMC. Concrete mixtures with varying ratios of latex and silica fume were prepared, and tests for compressive strength, flexural strength, and chloride-ion penetration resistance were conducted to compare and analyze the fundamental performance of each mix. Latex contributed to the improvement of the initial pore structure and significantly affected the chloride-ion penetration resistance in the early stages of curing. However, its influence gradually diminished over time. In contrast, silica fume induced additional C-S-H formation and further improved the pore structure through pozzolanic reactions as time progressed, thus exerting a greater impact in the later stages of curing. The L7-SF8 variable demonstrated the best performance in terms of compressive strength and chloride-ion penetration resistance. Given the characteristics of bridge-deck pavements, this variable is considered the most suitable for ensuring long-term durability. Therefore, this paper proposes a mixture of 7% latex and 8% silica fume as the optimal mix design.

본 논문에서는 15차 bézier 곡선을 사용하여 기존의 연구보다 더 유연한 빔 형상을 설계하고, 더 넓은 설계 공간에서 최적 설계를 수 행하여 최적의 열전도도를 갖는 빔 형상을 설계한다. 설계 공간이 넓어지면 그 만큼 계산양이 증가하게 되는데, 고차원 변수 공간에서 효율적으로 작동하는 인공신경망을 사용하여 최적 설계를 가속화하여 계산 한계를 극복하였다. 더 나아가 최적의 탄성계수를 갖는 빔의 형상과 비교하였으며 열전도와 탄성학 사이의 수학적 유사성을 이용하여 빔 형상을 설명한다. 본 연구에서는 인공지능을 활용 한 형상 최적설계를 통해 기존의 한계를 뛰어넘는 격자구조의 빔 형상을 제안한다. 먼저, SC(Simple Cubic), BC(Body Centered Cubic) 격자 구조 빔 형상을 bézier 곡선으로 모델링하고 bézier 곡선의 제어점 좌표를 무작위로 설정하여 학습데이터를 확보하였다. NN(Neural Network) 및 GA(Genetic Algorithm)를 통해 우수한 유효 열전도도를 가진 빔 형상을 생성하여 최적의 빔 형상을 설계하였 다. 본 연구를 통해 추후 다양한 열 조건에서 격자구조의 적절한 구조적 해답을 제시할 수 있을 것으로 기대된다.

Recent advances in computer technology have made it possible to solve numerous challenges but require faster hardware development. However, the size of the classical computer has reached its physical limit, and researchers' interest in quantum computers is growing, and it is being used in various engineering fields. However, research using quantum computing in structural engineering is very insufficient. Therefore, in this paper, the characteristics of qubits, the minimum unit of quantum information processing, were grafted with the crow search algorithm to propose QCSA (quantum crow search algorithm) and compare the convergence performance according to parameter changes. In addition, by performing the optimal design of the example truss structure, it was confirmed that quantum computing can be used in the architectural field.

In this study, we propose an optimal design method by applying the Prefabricated Buckling Restrained Brace (PF-BRB) to structures with asymmetrically rigidity plan. As a result of the PF-BRB optimal design of a structure with an asymmetrically rigidity plan, it can be seen that the reduction effect of dynamic response is greater in the case of arrangement considering the asymmetric distribution of stiffness (Asym) than in the case of arrangement in the form of a symmetric distribution (Sym), especially It was confirmed that at an eccentricity rate of 20%, the total amount of reinforced PF-BRBs was also small. As a result of analyzing the dynamic response characteristics according to the change in eccentricity of the asymmetrically rigidity plan, the distribution of the reinforced PF-BRB showed that the larger the eccentricity, the greater the amount of damper distribution around the eccentric position. Additionally, when comparing the analysis models with an eccentricity rate of 20% and an eccentricity rate of 12%, the response reduction ratio of the 20% eccentricity rate was found to be large.