AI-driven automation for structural design has been actively studied in structural engineering. In particular, reinforcement learning (RL) has attracted attention as a framework in which an agent interacts with an environment to autonomously search for optimal design solutions in complex design spaces. This study proposes an automated design model for rectangular reinforced-concrete (RC) columns based on a multi-agent Double Deep Q-Network (Double DQN). Extending prior RL-based automation developed for RC beam design to column members, the proposed environment explicitly incorporates key column-specific behaviors, including axial force–bending moment (P–M) interaction and moment magnification due to column buckling. Four agents independently determine the section width (b), section depth (h), number of longitudinal bars (n), and bar size. The reward function combines (i) penalty terms for violations of ACI 318-19 design constraints and (ii) an economic reward defined relative to an approximate optimal cost predicted by a quadratic regression model. After training for approximately 10,000 episodes, the proposed multi-agent Double DQN consistently generated ACI-compliant column designs across all test load cases and produced solutions with improved cost efficiency compared with the approximate optimal baseline. These results demonstrate the feasibility and practical potential of multi-agent RL for automated RC column section design.

During the reign of King Sejong in the Joseon Dynasty (1433-1438), the Daegyupyo (large gnomon) was produced. The Daegyupyo, with a crossbar (horizontal bar), was used to observe the length of the gnomon’s shadow cast by the sun passing at the meridian. The shadow of this crossbar can be obtained using a measurable device called the Yeongbu (shadow definer). These Daegyupyo and Yeongbu are described in detail in the “Treatise on Astronomy” of Yuan History or “Celestial Spheres and Globes” of Jega-Yeoksang-Jjp (Collected Discourses on the Astronomy and Calendrical Science of the Chinese Masters). According to Jega-Yeoksang-Jjp, the Yeongbu had a structure similar to a door attached to its frame. A pinhole is located in the center of a copper leaf corresponding to the door of the Yeongbu. The image of the sun’s meridian transit and the shadow of the crossbar through the pinhole are projected onto the surface of the Daegyupyo’s ruler stone. Unlike the width and length of the Yeongbu, the height of the Yeongbu is not recorded. This research analyzed the height of the Yeongbu required to maintain the constant distance from the pinhole to the ruler stone surface. Based on these assumptions, it was estimated that 8 to 13 Yeongbu of different heights would be needed for observations using the Daegyupyo in Seoul. To accommodate the need for Yeongbu of various heights, this study proposed a model for a stackable Yeongbu with an adjustable height.

Automated structural design methods for reinforced concrete (RC) beam members have been widely studied with various techniques to date. Recently, artificial intelligence has been actively applied to various engineering fields. In this study, machine learning (ML) is adopted to make automated structural design model for RC beam members. Among various machine learning methods, a supervised learning was selected. When a supervised learning is applied to development of ML-based prediction model, datasets for training and test are required. Therefore, the datasets for rectangular and t-shaped RC beams was constructed by commercial structural design software of MIDAS. Five supervised learning algorithms, such as Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor (KNN), Artificial Neural Networks (ANN), eXtreme Gradient Boosting (XGBoost) were used to develop the automated structural design model. Design moment (Mu), design shear force (Vu), beam length, uniform load (wu) were used for inputs of structural design model. Width and height of the designed section, diameter of top and bottom bars, number of top and bottom bars, diameter of stirrup bar were selected for outputs of structural design model. Performance evaluation of the developed structural design models was conducted using metrics sush as root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE), and coefficient of determination (R2). This study presented that random forest provides the best structural design results for both rectangular and t-shaped RC beams.

본 논문에서는 상용 프로그램 MIDAS GEN을 활용하여 플랜트 시설물의 특성을 반영한 골조와 단일 부재의 비선형 동적 해석을 수 행하였으며 이에 따른 결과를 분석하였다. 플랜트에 배치되는 일반적인 구조 부재의 크기와 재료적 특성을 고려하였으며, 수치해석 방법 중 뉴마크 평균 가속도법, 재료 비선형을 고려하기 위한 소성 힌지를 적용하였다. 플랜트 폭발의 대표적 유형인 증기운 폭발의 폭 발하중을 산정하였으며, 이를 골조 및 단일 부재에 적용하여 비선형 동적 해석을 수행하였다. 동적 거동의 결과는 고유주기와 하중지 속시간의 비율, 최대변위, 연성도, 회전각으로 정리하였으며 골조를 단일 부재로 해석할 수 있는 조건과 범위를 분석 및 확인하였다. 보-기둥 강성비가 0.5, 연성도가 2.0 이상인 NSFF는 FFC로 단순화할 수 있으며, 보-기둥 강성비가 0.5, 연성도가 1.5 이상인 NSPF는 FPC로 단순화하여 해석할 수 있다. 본 연구의 결과는 플랜트 시설물의 내폭설계 가이드라인으로 활용될 수 있다.

The purpose of this study is to suggest structural model and analyze design factors for the development of small greenhouse standardization model. The average dimensions of small greenhouse desired by urban farmers were 3.3m in width, 1.9m in eaves height, 2.7m in ridge height, 5.7m in length. The cladding materials for small greenhouse were preferred to glass, PC board and plastic film, framework to aluminum alloy and steel, and heating method in electrical energy. In addition, it was analyzed that small greenhouses need to develop structural model by dividing them into entry-level type and high-level type. The roof type that was used for entry-level type was arch shape, framework was steel pipe, cladding material was plastic film. On the other hand, high-level type was used in even span or dutch light type, framework with square hollow steel, cladding materials with glass or PC board. In consideration of these findings and practicality, this study developed four types of small greenhouses. The width, eaves height, ridges height, and length of the small greenhouses of even span type, which were covered with 5mm thick glass and 6mm thick PC board were 3m, 2.2m, 2.9m, and 6m, respectively. The small greenhouse of dutch light type covered with 5mm thick glass was designed with 3.8m in with, 2.2m in eaves height, 2.9m in ridges height, and 6m in length. The width, eaves height, ridges height, and length of the arch shape small greenhouse covered with a 0.15mm PO film were 3m, 1.5m, 2.8m, and 6m, respectively.

본 논문에서는 건물 구조 통합 구조설계 시스템의 구현을 위한 설계모델인 설계 객체 모델을 제안하였다. 건물 구조에 대한 구조 설계 정보를 단계(초기구조설계, 해석, 상세설계) / 계층(시스템, 서브시스템, 콤퍼넌트)별로 분류 모델링한 후, 제시된 요구조건에 대한 세부관점별 해결방법을 고려하여 설계 객체 모델을 개발하였다. 이와 같은 방법론을 통하여 시스템 구현을 고려한 설계 객체 모델의 체계적 분석과 모델링이 가능하였다. 제시된 설계 객체 모델은 계획 설계 측면의 설계정보 표현을 통하여 효율적인 설계정보의 관리가 가능하며, 위상 설계 객체에 의한 공간상 구조부재의 인식이 용이하고, 해석 관련 설계정보를 이해하기 용이한 표현으로 관리할 수 있게 한다.