Numerical Analysis and Experiments of Butt Welding Deformations for Panel Block Assembly

Ryu, Hyun-Su; Kang, Seok Joo; Lee, Kwangkook

doi:10.3390/app10051669

Cited by 6 publications

(3 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sriskandarajah and Hall [19] have proven the NP-hardness of such problems when the number of processing machines exceeds two (m > 2). Currently, heuristic algorithms, metaheuristic algorithms [20,21], and reinforcement learning algorithms are the mainstream research approaches for solving this type of problem. The following is a summary of the current research for each method.…”

Section: Literature Reviewmentioning

confidence: 99%

Solving Panel Block Assembly Line Scheduling Problem via a Novel Deep Reinforcement Learning Approach

Zhou

Luo

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

The panel block is a quite important “intermediate product” in the shipbuilding process. However, the assembly efficiency of the panel block assembly line is not high. Therefore, rational scheduling optimization is of great significance for improving shipbuilding efficiency. Currently, the processing sequence of the panel blocks in the panel block assembly line is mainly determined using heuristic and metaheuristic algorithms. However, these algorithms have limitations, such as small problem-solving capacity and low computational efficiency. To address these issues, this study proposes an end-to-end approach based on deep reinforcement learning to solve the scheduling problem of the ship’s panel block assembly line. First, a Markov decision model is established, and a disjunctive graph is creatively used to represent the current scheduling status of the panel block assembly line. Then, a policy function based on a graph isomorphism network is designed to extract information from the disjunctive graph’s state and train it using Proximal Policy Optimization algorithms. To validate the effectiveness of our method, tests on both real shipbuilding data and publicly available benchmark datasets are conducted. We compared our proposed end-to-end deep reinforcement learning algorithm with heuristic algorithms, metaheuristic algorithms, and the unimproved reinforcement learning algorithm. The experimental results demonstrate that our algorithm outperforms other baseline methods in terms of model performance and computation time. Moreover, our model exhibits strong generalization capabilities for larger instances.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Solving Panel Block Assembly Line Scheduling Problem via a Novel Deep Reinforcement Learning Approach

Zhou

Luo

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…The transverse shrinkage and angular deformation of the joint also increase significantly, but the residual stress of the weld and fusion line decreases. Ryu, Hyunsu [12] et al established a quantitative relationship between welding deformation and temporary pieces setting of temporary weld parts through tests, and proposed a numerical method for welding deformation considering the setting of temporary parts, which can be used for rational use of temporary parts in the plate assembly stage and greatly improve productivity. Ma X-Q [13] et al proposed a method based on the three-dimensional thermal digital image correlation (DIC) method to measure the threedimensional full-field dynamic deformation of the curved surface of sheet metal, studied the evolution law of in-plane and out-of-plane deformation in the process of surfacing welding and cooling of the thin curved plate, and established the relationship between curvature of the curved plate and weld contraction.…”

Section: Introductionmentioning

confidence: 99%

Influence of Initial Structural Dimensions of Plates on Welding Distortion

Guo,

Zhang,

Tang

et al. 2023

Coatings

View full text Add to dashboard Cite

Aiming at the complex full-field deformation problem that easily occurs when welding plates, this paper adopts the elastic–plastic finite element method with heat-force coupling to study the deformation law of plates in different initial states. First, a rectangular plate finite element model with an initial radius and Gaussian heat source model was established to obtain the welding temperature field and deformation field of the plate; then, the method based on digital image correlation technology was used to detect the full-field dynamic deformation of the plate to verify the accuracy of the finite element model; finally, the influence of the initial structural dimensions of the plate on the weld deformation was investigated. The study shows the following: the thermoelastic–plastic finite element model proposed in this paper has high accuracy in both static and dynamic deformation; plates with the same curvature, and different lengths and widths of the initial structure of the plate welding deformation are saddle-shaped, and the edge effect of the welding of the plate is evident, independent of the length of the plate; and the maximum out-of-face deformation of the welding of the plate is linearly related to the length and the closer the aspect ratio of the plate is to 1, the smaller the out-of-face deformation is.

show abstract

“…While conducting this technique, a large amount of molten wire metal fills the groove between the structural components of the base material. As the heat input of the filler material is very high and its cooling rate is rapid afterwards, permanent plastic deformations and residual stresses occur in the weld area and its vicinity [1][2][3][4][5]. The plastic deformations disturb the external appearance of the welded structure making its assembly difficult.…”

Section: Introductionmentioning

confidence: 99%

Numerical Simulation and Experimental Investigation of Temperature and Residual Stress Distributions in a Circular Patch Welded Structure

Perić¹,

Nižetić

Tonković

et al. 2020

Energies

View full text Add to dashboard Cite

In this study, we performed a numerical simulation and experimental measurements on a steel circular patch welded structure to investigate the temperature and residual stress field distributions caused by the application of buried-arc welding technology. The temperature histories during the welding and subsequent cooling process were recorded for two locations, with the thermocouples mounted inside the plate close to the weld bead. On the upper surface of the welded model, the temperature-time changes during the cooling process were monitored using an infrared camera. The numerically calculated temperature values correlated well with the experimentally measured ones, while the maximum deviation of the measured and calculated temperatures was within 9%. Based on the numerical result analysis regarding circumferential and radial stresses after the completion of the welding process, it is concluded that both stresses are primarily tensile within the circular disk. Outside the disk, the circumferential stresses turn from tensile to compressive, while on the other hand the radial stresses disappear towards the ends of the plate.

show abstract

Numerical Analysis and Experiments of Butt Welding Deformations for Panel Block Assembly

Cited by 6 publications

References 21 publications

Solving Panel Block Assembly Line Scheduling Problem via a Novel Deep Reinforcement Learning Approach

Solving Panel Block Assembly Line Scheduling Problem via a Novel Deep Reinforcement Learning Approach

Influence of Initial Structural Dimensions of Plates on Welding Distortion

Numerical Simulation and Experimental Investigation of Temperature and Residual Stress Distributions in a Circular Patch Welded Structure

Contact Info

Product

Resources

About