Article
Computer Science, Interdisciplinary Applications
Yi Xing, Liyong Tong
Summary: In this work, a machine learning-assisted structural optimization (MLaSO) scheme is proposed to accelerate the computational speed of structural optimization. A new machine learning model is used to predict the update of the optimization quantity during the optimization process, eliminating the need for finite element analysis and sensitivity analysis. The MLaSO scheme can be easily integrated into different structural optimization methods and does not require additional training datasets.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2022)
Review
Computer Science, Information Systems
Ziyue Guo, Zongyang Zhu, Yizhi Li, Shidong Cao, Hangyue Chen, Gaoang Wang
Summary: This article explores the integration of enhanced personalization and seamless multimodal interfaces in the field of fashion design and recommendation. It discusses the increasing demand for personalized fashion experiences and the potential of multimodal interfaces in facilitating effective communication between designers and users. By leveraging user preferences, body measurements, and style choices, AI systems can deliver highly personalized fashion recommendations. The integration of various input modalities enables designers and users to communicate their design ideas with ease. The primary results highlight the transformative potential of enhanced personalization and seamless multimodal interfaces in empowering designers and consumers to co-create unique and personalized designs.
Article
Oncology
Dalia Fahmy, Heba Kandil, Adel Khelifi, Maha Yaghi, Mohammed Ghazal, Ahmed Sharafeldeen, Ali Mahmoud, Ayman El-Baz
Summary: This manuscript discusses the applications of artificial intelligence (AI) in lung segmentation, pulmonary nodule segmentation, and classification. It highlights the importance of early detection of pulmonary nodules for lung cancer treatment and saving lives, and how AI technology can improve diagnostic accuracy.
Article
Automation & Control Systems
Minsik Seo, Seungjae Min
Summary: This paper introduces a deep neural network-based method for accelerating topology optimization in irregular design domains, which can predict (near-)optimal topologies.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Computer Science, Interdisciplinary Applications
Chao Qian, Wenjing Ye
Summary: This study utilizes artificial neural networks as efficient surrogate models for forward and sensitivity calculations in topology optimization, achieving faster design processes and improved accuracy. Dual-model artificial neural networks are used to enhance sensitivity analysis accuracy and are integrated into the Solid Isotropic Material with Penalization (SIMP) method, showing performance gains in two benchmark design problems.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2021)
Review
Computer Science, Interdisciplinary Applications
Rebekka V. Woldseth, Niels Aage, J. Andreas Baerentzen, Ole Sigmund
Summary: The question of how artificial intelligence methods can improve traditional frameworks for topology optimization has gained attention in the past few years. While different model variations have been proposed with varying levels of success, few significant breakthroughs have been achieved. The literature tends to have a strong belief in the magical capabilities of artificial intelligence, leading to misunderstandings about its limitations. This article presents a critical review of the current state of research in this field and provides recommendations for future scientific progress.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2022)
Review
Computer Science, Artificial Intelligence
Sara Kaviani, Ki Jin Han, Insoo Sohn
Summary: In recent years, the improvement of medical images and the performance of deep learning networks have led to the application of deep learning approaches in medical image classification and segmentation. However, there are concerns about the security and accuracy of healthcare systems, as well as the vulnerability of medical deep learning networks to adversarial attacks. This paper reviews the proposed adversarial attack methods and defense techniques for medical imaging DNNs and discusses future directions for improving neural network's robustness.
EXPERT SYSTEMS WITH APPLICATIONS
(2022)
Article
Computer Science, Interdisciplinary Applications
Ren Kai Tan, Chao Qian, Kangjie Li, Dan Xu, Wenjing Ye
Summary: Topology optimization is a systematic approach for obtaining optimal performance in structural design, but it can be computationally expensive and deep learning models lack generalizability. This work proposes an adaptive, scalable deep learning-based model-order-reduction method using MapNet to accelerate large-scale topology optimization. The method allows simulations to be performed at a coarser mesh, reducing computational time, and introduces domain fragmentation to improve the method's transferability and scalability.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2022)
Article
Engineering, Multidisciplinary
Minsik Seo, Seungjae Min
Summary: DL-MSTO+ is a deep learning-based multi-scale topology optimization framework that improves the efficiency of multi-scale topology optimization by reducing the dimensionality of design variables and predicting homogenized material properties. The framework includes two distinct deep neural networks for learning the low-dimensional representation of material microstructures and predicting the homogenized elasticity matrix. The proposed method demonstrates higher efficiency than the conventional multi-scale approach in numerical experiments and provides connectable multi-scale designs.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2023)
Article
Engineering, Multidisciplinary
Martin Ohrt Elingaard, Niels Aage, Jakob Andreas Baerentzen, Ole Sigmund
Summary: This paper presents a deep learning-based de-homogenization method for structural compliance minimization, showing excellent generalization properties and performance within 7-25% of homogenization-based solutions at a fraction of the computational cost, while being robust and insensitive to domain size, boundary conditions, and loading.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2022)
Article
Nanoscience & Nanotechnology
Rohit Unni, Kan Yao, Xizewen Han, Mingyuan Zhou, Yuebing Zheng
Summary: The research introduces a tandem optimization model that combines a mixture density network (MDN) and a fully connected network to inversely design practical thin-film high reflectors. This model can retrieve the reflectance spectra of 20-layer thin-film structures and demonstrate improved designs with extended high-reflectance zones. By combining the efficiency advantage of DL with optimization-enabled performance improvement, efficient and on-demand inverse design for practical applications is enabled.
Editorial Material
Engineering, Electrical & Electronic
Samuel K. Moore, David Schneider, Eliza Strickland
Summary: In an artificial neural network, each neuron calculates its output by summing its inputs and applying an activation function, similar to how neurons in the brain transmit signals between each other via synapses.
Article
Computer Science, Interdisciplinary Applications
Ren Kai Tan, Chao Qian, Michael Wang, Wenjing Ye
Summary: The study proposes a solution to reduce the training cost of artificial-neural-network (ANN)-based surrogate models by reducing the number of numerical simulations during training data generation. The solution utilizes a Mapping Network to map a coarse field to a fine field, generating fine-scale training data.
STRUCTURAL AND MULTIDISCIPLINARY OPTIMIZATION
(2022)
Article
Engineering, Multidisciplinary
Weisheng Zhang, Yue Wang, Zongliang Du, Chang Liu, Sung-Kie Youn, Xu Guo
Summary: A machine-learning assisted topology optimization approach is proposed for architectural design with artistic flavor. Neural style transfer technique is adopted to measure and generate prior knowledge from a reference image with concerned artistic flavor. The measured knowledge is integrated into pixel-based topology optimization as a formal similarity constraint. The effectiveness of the proposed approach is illustrated through solving 2D and 3D problems, achieving systematic inheritance of artistic heritage.
COMPUTER METHODS IN APPLIED MECHANICS AND ENGINEERING
(2023)
Article
Engineering, Electrical & Electronic
Thomas Falconer, Letif Mones
Summary: Machine learning assisted optimal power flow (OPF) reduces computational complexity by using offline training and investigating different neural network architectures (FCNN, CNN, GNN). Results show limited usefulness of CNN and GNN compared to FCNN for fixed grid topology, but GNN outperforms FCNN and CNN when considering variable topology (transmission line contingency).
IEEE TRANSACTIONS ON POWER SYSTEMS
(2023)
Article
Engineering, Mechanical
Stephanie Kirmse, Lucio Flavio Campanile, Alexander Hasse
Summary: By utilizing compliant mechanisms, controlled deformations can be achieved, but mechanisms with distributed compliance may deviate from desired behavior. Selective compliant mechanisms are advanced structures that can approximate ideal behavior. A new synthesis approach, which incrementally linearizes the optimization problem, is presented for the stable synthesis of mechanisms with multiple output degrees of freedom.
MECHANISM AND MACHINE THEORY
(2021)
Article
Acoustics
Alexander Nowak, L. Flavio Campanile, Alexander Hasse
Summary: Semi-active vibration reduction techniques control system parameters like stiffness to dissipate energy more efficiently, with localized stiffness changes yielding larger vibration attenuation compared to spatially homogeneous modulation, which results in a pseudo-active effect.
JOURNAL OF SOUND AND VIBRATION
(2021)
Article
Materials Science, Multidisciplinary
Lucio Flavio Campanile, Stephanie Kirmse, Alexander Hasse
Summary: Compliant mechanisms are alternative to traditional mechanisms that use elastic strain to achieve desired deformation. They lack a strict definition of kinematics and can be seen as a fuzzy property. This paper discusses the concepts of accuracy and precision in compliant systems, proposing a quantitative method based on eigenvalue analysis of hinge stiffness to determine these qualities.
JOURNAL OF INTELLIGENT MATERIAL SYSTEMS AND STRUCTURES
(2022)
Article
Materials Science, Multidisciplinary
Jan Papuga, Martin Nesladek, Alexander Hasse, Eva Cizova, Lukas Suchy
Summary: This paper presents a comparison between six recently introduced multiaxial fatigue strength estimation criteria and four already published methods. The results of each newer method are analyzed and discussed. The comparison shows that only Bohme's criterion reaches a similar estimation quality to the best performing criteria. Additionally, the validation using the derived AMSD25 data set allows for a reduction in the number of evaluated test cases while preserving the worst cases that highlight the weaknesses of different estimation methods.
Article
Materials Science, Multidisciplinary
Kristian Mauser, Martin Sjarov, Alexander Nowak, Lucio Flavio Campanile, Alexander Hasse
Summary: This paper investigates compliant mechanisms with variable stiffness behavior, achieved through targeted application of prestressing forces. It presents an efficient optimization method based on linear programming, which is numerically and experimentally validated to precisely vary the stiffness behavior of a compliant mechanism. The study demonstrates the potential for precise variation of stiffness behavior with the presented approach.
JOURNAL OF INTELLIGENT MATERIAL SYSTEMS AND STRUCTURES
(2022)
Article
Materials Science, Multidisciplinary
Stephanie Seltmann, Lucio Flavio Campanile, Alexander Hasse
Summary: This paper presents a novel optimization algorithm that can synthesize different types of compliant mechanisms. The algorithm addresses the issues of load-case dependency and desired deformations by minimizing the former and imposing the latter.
JOURNAL OF INTELLIGENT MATERIAL SYSTEMS AND STRUCTURES
(2023)
Article
Engineering, Manufacturing
Roman Funke, Jonathan Schanner, Alexander Hasse, Andreas Schubert
Summary: Increasing the coefficient of static friction in frictional connections can help meet current demands like increasing power density or lightweight design. The study found that microstructures with a PTA of 88° were most effective in increasing the COF. However, even slight changes in the feed per tooth had a strong effect on both surface topography and COF.
JOURNAL OF MANUFACTURING PROCESSES
(2022)
Article
Engineering, Mechanical
Sebastian Vetter, Erhard Leidich, Alexander Hasse
Summary: This paper investigates the possibility of quantifying the fatigue-strength scatter in high-cycle and very-high-cycle fatigue regimes. Experimental tests were carried out on shafts to determine the scatter and a probabilistic model was developed. The fatigue strength distribution was determined by analyzing fictitious nominal stress levels and calculated failures of randomly generated shafts.
ENGINEERING FAILURE ANALYSIS
(2023)
Article
Engineering, Mechanical
Denny Knabner, Sven Hauschild, Lukas Sucny, Sebastian Vetter, Erhard Leidich, Alexander Hasse
Summary: The lower fretting-fatigue strengths compared to plain fatigue strengths is a significant characteristic in strength assessments of component connections. This is due to the additional tribological stress caused by contact pressure, slip amplitude and material pairing. A double-actuated flat-pad test bench is used to analyze these variables separately. The study focuses on steel materials commonly used in shaft-hub connections. The proposed local approach, based on the FKM guideline, considers fretting factors to ensure fail-safe design in fretting fatigue cases.
INTERNATIONAL JOURNAL OF FATIGUE
(2022)
Article
Engineering, Manufacturing
Jonathan Schanner, Roman Funke, Andreas Schubert, Alexander Hasse
Summary: The coefficient of friction is an important parameter for mechanical engineers designing frictional connections. Surface microstructuring of the harder friction partner increases the coefficient of friction and reduces sensitivity to lubricant contamination.
JOURNAL OF MANUFACTURING AND MATERIALS PROCESSING
(2022)
Proceedings Paper
Engineering, Mechanical
Denny Knabner, Sebastian Vetter, Lukas Suchy, Alexander Hasse
Summary: According to the literature, the location of failure in fretting fatigue problems depends on the material pairing and contact parameters. Previous calculation methods were inaccurate in determining the failure location. This study conducted experimental tests and numerical modeling on realistic component connections and fretting pads to evaluate local stresses. The results showed significant differences in failure locations for different material pairings, and the McDiarmid criterion gave the smallest error.
PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON FRACTURE, FATIGUE AND WEAR, FFW 2021
(2022)
Proceedings Paper
Construction & Building Technology
Sebastian Vetter, Erhard Leidich, Nils Becker, Berthold Schlecht, Alexander Hasse
Summary: This paper investigates the possibility of quantifying the scatter of fatigue strength in the high-cycle fatigue region of shafts using a probabilistic model. The model is based on identifying the parameters influencing the damage mechanism, and statistically modeling the scatter-influencing parameters. A local strength approach is used in combination with non-scattering parameters. By developing a probabilistic model using Monte Carlo simulations, shafts and their surfaces can be randomly generated and subjected to local strength verifications. The probability distribution of fatigue strength and its scatter can be determined by statistically evaluating the generated shafts. The model is validated and compared with experimental results.
4TH INTERNATIONAL CONFERENCE ON STRUCTURAL INTEGRITY (ICSI 2021)
(2022)