Deep learning-based inverse design for engineering systems: multidisciplinary design optimization of automotive brakes

Kim, Seongsin; Jwa, Minyoung; Lee, Soonwook; Park, Sunghoon; Kang, Namwoo

doi:10.1007/s00158-022-03386-8

Cited by 6 publications

(5 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ongoing research is exploring the potential of deep learning to enhance the performance of topology optimization-based generative design in design exploration [15], [28], [31]. Studies [32], [33] employed convolution filters of deep belief networks and reduced-order models to generate various topology designs. In research [34], proposed generative design based on reinforcement learning eliminates the need for preoptimized topological iteration.…”

Section: Generative Models For 3d Vehicle Wheelsmentioning

confidence: 99%

“…With hierarchical data features, CNNs' basic architecture emulates human visual processing abilities, allowing them to recognize, classify, and interpret environmental information [18]. Deep neural networks (DNNs) are commonly utilized for dimensionality reduction in unsupervised learning [33]. With the same sizes for the input and output layers in the autoencoder architecture, autoencoders, in particular, compress high-dimensional input data into a low-dimensional latent space.…”

Section: Computer-aided Engineering (Cae) and Deep Learning Integrationmentioning

confidence: 99%

“…In the domain of automotive engineering, the infusion of deep learning-based algorithms into CAE marks a transformative shift. Deep learning, a subset of artificial intelligence (AI), renowned for its prowess in image and data analysis, natural language processing, and pattern recognition, holds the potential to revolutionize CAE practices [33], [57]. Extending these capabilities to the specific domain of CAE, especially in simulating 3D vehicle wheels, promises unparalleled accuracy and efficiency.…”

Section: Cae Reinforcement Learningmentioning

confidence: 99%

See 2 more Smart Citations

A Review of Generative Models for 3D Vehicle Wheel Generation and Synthesis

Akande,

Alabi,

Oyinloye

2024

J. Comput. Theor. Appl.

View full text Add to dashboard Cite

Integrating deep learning methodologies is pivotal in shaping the continuous evolution of computer-aided design (CAD) and computer-aided engineering (CAE) systems. This review explores the integration of deep learning in CAD and CAE, particularly focusing on generative models for simulating 3D vehicle wheels. It highlights the challenges of traditional CAD/CAE, such as manual design and simulation limitations, and proposes deep learning, especially generative models, as a solution. The study aims to automate and enhance 3D vehicle wheel design, improve CAE simulations, predict mechanical characteristics, and optimize performance metrics. It employs deep learning architectures like variational autoencoders (VAEs), convolutional neural networks (CNNs), and generative adversarial networks (GANs) to learn from diverse 3D wheel designs and generate optimized solutions. The anticipated outcomes include more efficient design processes, improved simulation accuracy, and adaptable design solutions, facilitating the integration of deep learning models into existing CAD/CAE systems. This integration is expected to transform design and engineering practices by offering insights into the potential of these technologies.

show abstract

Section: Generative Models For 3d Vehicle Wheelsmentioning

confidence: 99%

Section: Computer-aided Engineering (Cae) and Deep Learning Integrationmentioning

confidence: 99%

Section: Cae Reinforcement Learningmentioning

confidence: 99%

See 1 more Smart Citation

A Review of Generative Models for 3D Vehicle Wheel Generation and Synthesis

Akande,

Alabi,

Oyinloye

2024

J. Comput. Theor. Appl.

View full text Add to dashboard Cite

show abstract

“…Derivative-Free Global Optimization (DFGO) problems that require significant computational resources can be found in a wide range of fields, including robotics (Hauser, 2017;Wang et al, 2020), engineering design (Lin et al, 2022;Kim et al, 2022b), economics (Liu et al, 2022;Kim et al, 2022a), tourism (Liao et al, 2021;Paulavičius et al, 2023), and many others (Floudas et al, 2013;Moret et al, 2016;Grigaitis et al, 2007;. These problems often involve black-box functions that require expensive simulations or experiments for evaluation.…”

Section: Introductionmentioning

confidence: 99%

Review and Computational Study on Practicality of Derivative-Free DIRECT-Type Methods

Stripinis,

Paulavičius

2024

Informatica

View full text Add to dashboard Cite

Derivative-free DIRECT-type global optimization algorithms are increasingly favoured for their simplicity and effectiveness in addressing real-world optimization challenges. This review examines their practical applications through a systematic analysis of scientific journals and computational studies. In particular, significant challenges in reproducibility have been identified with practical problems. To address this, we conducted an experimental study using practical problems from reputable CEC libraries, comparing DIRECT-type techniques against their state-of-the-art counterparts. Therefore, this study sheds light on current gaps, opportunities, and future prospects for advanced research in this domain, laying the foundation for replicating and expanding the research findings presented herein.

show abstract

“…Currently, the technology for generative AI is very active and progressing at a very fast pace. [1] This direction is also being applied to industrial companies to reflect generative design [2] , and this research is being conducted through a research project as described later.…”

Section: Introductionmentioning

confidence: 99%

Case Study of Product Development through Generative Design according to Anemometer Replacement Cycles

CHOI,

Shin,

Lee

2024

PHME_CONF

View full text Add to dashboard Cite

Product Lifecycle Management (PLM) systems are commonly used to manage various product data generated throughout the product lifecycle. This paper explains the results obtained by multiple participants using commercial software within the PLM environment to perform structural and vibration analyses of an Anemometer. Generative design techniques were employed for 3D CAD modeling of the Anemometer, and the commercial analysis software NASTRAN was used for simulation analyses. The open-source PLM system ARAS Innovator's project and workflow management modules were utilized to manage the generated design data, allocate tasks among participants, and control schedules. Through this approach, we propose a method to predict and manage the replacement cycle of Anemometer.

show abstract

Deep learning-based inverse design for engineering systems: multidisciplinary design optimization of automotive brakes

Cited by 6 publications

References 47 publications

A Review of Generative Models for 3D Vehicle Wheel Generation and Synthesis

A Review of Generative Models for 3D Vehicle Wheel Generation and Synthesis

Review and Computational Study on Practicality of Derivative-Free DIRECT-Type Methods

Case Study of Product Development through Generative Design according to Anemometer Replacement Cycles

Contact Info

Product

Resources

About