A data-driven modelling and optimization framework for variable-thickness integrally stiffened shells

Li, Hongqing; Li, Zengcong; Cheng, Zhizhong; Zhou, Zhiyong; Wang, Gang; Wang, Bo; Tian, Kuo

doi:10.1016/j.ast.2022.107839

Cited by 16 publications

(6 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, the section parameters optimization under internal pressure load is carried out, and the shell cross-section of the pressurized capsule structure is determined. (2) Design of basic stiffener configuration of pressurized capsule [17,18,21,22,[37][38][39][40] According to the two-dimensional results obtained in the previous step, a threedimensional finite element model of the pressurized capsule shell structure is established. On this basis, the optimization design of the basic configuration is carried out, and the design scheme of the stiffened configuration is optimized.…”

Section: Research On the Design Methods Of Ippcsmentioning

confidence: 99%

“…Compared with the traditional modeling method, the modeling accuracy and efficiency are greatly improved. On this basis, Li et al [18] combined the Nonuniform Rational B-Splines (NURBS) method to establish a variable thickness integral stiffened shell modeling method, which can consider the variable thickness modeling and optimization of ribs and skins, and fully tap the potential of lightweight structure. Compared with the traditional curve design, NURBS [39,41] is the best representative form for the curve and surface.…”

Section: Research On the Design Methods Of Ippcsmentioning

confidence: 99%

“…), and the panel features are directly processed on the whole semifinished products with the final capsule shape. Since the design domain does not have the segmentation of the weld zone, the design space of the panel is extended to the global capsule, and a better design can be obtained through structural optimization [17][18][19][20][21][22]. The IPPCS not only realizes the global stiffeners' continuity of the capsule, but also the weight is reduced by replacing the frame with a local panel structure (as shown in Figure 9), which overcomes the shortcomings of the WPPCS and significantly improves the lightweight level of the pressurized capsule structure.…”

Section: The Emergence Of Ippcsmentioning

confidence: 99%

See 2 more Smart Citations

Progress in Lightweight Design Methods for Large-Size Panel Structures in Manned Pressurized Capsules

Zhou

Han

et al. 2023

Applied Sciences

Self Cite

View full text Add to dashboard Cite

The pressurized capsule structure provides the pressure environment for astronauts or payloads in space, which is thus considered as the most crucial structural component for manned spacecraft. The manned deep space exploration mission (MDSEM) brings new challenges to the pressurized capsule structure: extremely low structural weight, long service life, reusability and adaptability to the harsh deep space environment. The conventional welded panel pressurized capsule structure (WPPCS) is not able to meet these new requirements. To address the above challenges, this paper comprehensively expounds why the current WPPCS cannot meet the requirements of MDSEMs based on the analysis of the vibration environment and structural characteristics of the pressurized capsule structure. Furthermore, a new type of integrated panel pressurized capsule structure (IPPCS) is proposed, showing the lightweight advantage compared with WPPCS. Finally, the technical details and research results of the strength criterion, design method, material upgrading and structural integrity manufacturing process of the IPPCS are fully introduced. The conclusions drawn in this paper will provide useful and meaningful references for the future development of large-size, lightweight pressurized capsule structures.

show abstract

Section: Research On the Design Methods Of Ippcsmentioning

confidence: 99%

Section: Research On the Design Methods Of Ippcsmentioning

confidence: 99%

Section: The Emergence Of Ippcsmentioning

confidence: 99%

See 1 more Smart Citation

Progress in Lightweight Design Methods for Large-Size Panel Structures in Manned Pressurized Capsules

Zhou

Han

et al. 2023

Applied Sciences

Self Cite

View full text Add to dashboard Cite

show abstract

“…The results show that machine learning can accurately predict high-performance metal-organic frameworks and can improve the screening speed by 2-3 orders of magnitude. [13] Some researchers worked on machine learning methods or surrogate modeling methods for accelerating structure analysis, [14][15][16][17] Tian et al employed transfer learning to establish the variablefidelity surrogate model for shell buckling prediction, [15] demonstrating the potential of transfer-learning based variable-fidelity surrogate model in time-consuming prediction problems.…”

Section: Literature Reviewmentioning

confidence: 99%

Accelerating Analysis for Structure Design via Deep Learning Surrogate Models

Shao

Chen

Wang

et al. 2023

Advanced Intelligent Systems

View full text Add to dashboard Cite

Using computer simulation tools such as finite element analysis (FEA) to perform material stress analysis is a common design method in engineering practice. In order to model more realistic real‐world systems, simulation models have become more complex, and calculation becomes more expensive as a result. The rise of artificial intelligence technologies has made it possible to integrate deep learning methods and material stress analysis. Herein, FEA software is employed to obtain a large number of analysis cases as training samples and uses a fully connected neural network and long‐short‐term memory neural network as surrogate models, which can predict the stress distribution and stress sequence in the process of the bullet impacting target plates with different materials. These models can give results similar to FEA with 92.19% and 92.41% accuracy, respectively. The experimental results show that the deep learning surrogate models have great potential in material stress analysis.

show abstract

“…Curved surfaces are typically characterized by non-straight generatrix, making it challenging to model and optimize stiffeners on these curved surfaces. Tian et al [32] and Li et al [33] developed a novel mesh deformation method for data-driven modeling of stiffened curved shells based on RBF neural network machine learning methods and presented an optimization framework for stiffened curved shells. Zhang et al [34]- [35] proposed an effective B-spline parameterization method for the stiffener layout optimization of shell structures and extended it to handle stiffener layout optimization of thin-walled structures with complex surfaces using mesh parameterization.…”

Section: Introductionmentioning

confidence: 99%

Optimal design of non-uniform curved grid-stiffened shell with different stiffener patterns

Yu,

Xiaoang,

Yan

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

This paper presents a non-uniform curved grid-stiffened shell design method aiming to enhance structural performance using various stiffener patterns, allowing simultaneous optimization of stiffener thickness and stiffener layout. Firstly, the grid-stiffened cell description function is defined using quadratic polynomial functions, comprising the orthogrid, the triangle grid, the rotated triangle grid and the Kagome grid. Then, the non-uniform stiffener layout description function is established using the sawtooth function, while a filter function is employed to ensure the smooth and continuous of the stiffeners. Moreover, the analytical sensitivity is thoroughly derived, and the optimization problem is formulated. Finally, the effectiveness of the proposed method is demonstrated through three representative numerical examples: the cantilever beam, the special-shaped plate and the S-shape shell. The study concludes that the proposed method can optimize arbitrary flat plates by embedding the design domain into the background grid. Additionally, the proposed method can be extended to perform stiffener design on complex surfaces using mesh projection technology. Optimization results indicate that the non-uniform curved grid-stiffened shell design exhibits superior structural performance compared to the uniform grid-stiffened shell design.

show abstract

A data-driven modelling and optimization framework for variable-thickness integrally stiffened shells

Cited by 16 publications

References 56 publications

Progress in Lightweight Design Methods for Large-Size Panel Structures in Manned Pressurized Capsules

Progress in Lightweight Design Methods for Large-Size Panel Structures in Manned Pressurized Capsules

Accelerating Analysis for Structure Design via Deep Learning Surrogate Models

Optimal design of non-uniform curved grid-stiffened shell with different stiffener patterns

Contact Info

Product

Resources

About