• Structural Engineering and Mechanics
• /
• 제84권5호
• /
• pp.605-618
• /
• 2022

Glass fiber reinforced polymer (GFRP) elastic gridshells consist of long continuous GFRP tubes that form elastic deformations. In this paper, a method for the form-finding of gridshell structures is presented based on the interpretable machine learning (ML) approaches. A comparative study is conducted on several ML algorithms, including support vector regression (SVR), K-nearest neighbors (KNN), decision tree (DT), random forest (RF), AdaBoost, XGBoost, category boosting (CatBoost), and light gradient boosting machine (LightGBM). A numerical example is presented using a standard double-hump gridshell considering two characteristics of deformation as objective functions. The combination of the grid search approach and k-fold cross-validation (CV) is implemented for fine-tuning the parameters of ML models. The results of the comparative study indicate that the LightGBM model presents the highest prediction accuracy. Finally, interpretable ML approaches, including Shapely additive explanations (SHAP), partial dependence plot (PDP), and accumulated local effects (ALE), are applied to explain the predictions of the ML model since it is essential to understand the effect of various values of input parameters on objective functions. As a result of interpretability approaches, an optimum gridshell structure is obtained and new opportunities are verified for form-finding investigation of GFRP elastic gridshells during lifting construction.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2021년도 가을 학술논문 발표대회
• /
• pp.4-5
• /
• 2021

In irregular curtain walls, the nominal stress required for each member varies greatly depending on the shape, so it is inefficient to design members based on the maximum required stress. Then, built-up members are absolutely necessary, but built-up members manufactured by Manpower-welding cannot be constructed in an irregular curtain wall building because it' not precise. In order to address the problems, this paper presents why Robotic-laser-welding should be used in irregular curtain walls using Gwanggyo Galleria Department Store involving 3D printing as an example. Results verify the performance of Robot-Laser-Welding as an efficient solution for precise steel curtain wall members.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2018년도 추계 학술논문 발표대회
• /
• pp.8-9
• /
• 2018

3D printing has the unique advantage of the ability to customize freeform product even in small quantity. However, we need to select and apply the only necessary parts of it because of the high cost of the manufacturing technology. It is of critical importance in irregular-shaped curtain walls to ensure precision of construction as well as quality fo finish. Complex shape that have structural members at varying angles can have nodes of different shapes making it unfeasible to construct using a general node connection detail. Therefore, this study aims to utilize smart node system using 3D printing as a solution to complex irregular-shaped curtain wall design.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2021년도 봄 학술논문 발표대회
• /
• pp.6-7
• /
• 2021

3D printing has the unique advantage of the ability to customize freeform products even in small quantities. It was recently applied successfully in the node connection system for the irregular shaped glass design of Gwanggyo Galleria Department Store. This was achieved by applying 3D printing technology for an innovative smart node design. The novel system offers flexibility to address various design challenges in addition to maximizing construction quality and reducing construction period. In this paper, we aim to examine the use of 3D printing based innovative technologies in the construction industry. With this aim in mind, we present the case of the Gwanggyo Galleria Department Store's smart nodes. The key objective of this study is creating awareness in the construction industry of the need to utilize fourth industrial revolution technology in architecture.