Article
Engineering, Multidisciplinary
Mengchao Zhang, Kai Jiang, Shuai Zhao, Nini Hao, Yuan Zhang
Summary: This study proposes a deep-learning-based multioperation synchronous monitoring method for belt conveyors, which can simultaneously diagnose belt deviation, measure conveying load, identify idlers, and perform other tasks. The method effectively reduces the complexity and cost of monitoring, thereby avoiding environmental pollution caused by transportation accidents.
STRUCTURAL HEALTH MONITORING-AN INTERNATIONAL JOURNAL
(2023)
Article
Chemistry, Physical
Maciej Tabaszewski, Pawel Twardowski, Martyna Wiciak-Pikula, Natalia Znojkiewicz, Agata Felusiak-Czyryca, Jakub Czyiycki
Summary: This paper compares different intelligent system methods to identify tool wear and explores the application of machine learning in improving the quality of manufacturing industries. The results show that machine learning methods based on vibration acceleration signals can effectively predict the wear condition of tools.
Article
Computer Science, Artificial Intelligence
Zhan Gao, Qiguo Hu, Xiangyang Xu
Summary: This study proposes a multi-granularity feature extraction method based on GLCM and RF to improve the prediction accuracy of remaining useful life for turning tools.
NEURAL COMPUTING & APPLICATIONS
(2022)
Article
Computer Science, Information Systems
Ismail A. Adeleke, Nnamdi I. Nwulu, Omolola A. Ogbolumani
Summary: This paper explores the feasibility of using machine learning and the Internet of Things to monitor water quality at water storage stations. The study collects data on physical and chemical parameters using sensors and uses Artificial Neural Network and Support Vector Machine algorithms to predict water pollution levels. The results show that the ANN models have the highest accuracy and are most suitable for predicting water source and quality. The paper also introduces an automated corrective measure based on water contamination levels. Overall, AI and IoT are found to be more efficient in remotely monitoring water safety and hazards.
INTERNET OF THINGS
(2023)
Article
Computer Science, Information Systems
Gabriele Costa, Fabio Pinelli, Simone Soderi, Gabriele Tolomei
Summary: This paper demonstrates that federated learning systems can be transformed into covert channels for clandestine communications. Malicious actors can manipulate the global model to transmit information to a receiver without impacting overall model performance, posing a significant threat to FL infrastructures.
Article
Chemistry, Multidisciplinary
Maged Abdullah Esmail
Summary: The demand for network capacity has increased due to new digital applications and services that rely heavily on optical communication networks. To avoid the challenge of deploying fiber in certain scenarios, a hybrid all-optical fiber/FSO link is proposed. Machine learning techniques, specifically Gaussian process regression, are used to predict channel impairments in the hybrid all-optical fiber/FSO channels. The results show high accuracy in predicting most impairments, except for light turbulence parameters under strong ASE noise. The proposed approach provides a self-aware and self-adaptive communication system and has the potential to optimize network resources in the future.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Marine
Tatsuya Kaneko, Ryota Wada, Masahiko Ozaki, Tomoya Inoue
Summary: This article proposes a hybrid model combining a physics-based model and machine learning to estimate the behavior of unmeasurable parts in dynamic systems using measurable data in real-time. The effectiveness of the proposed model was verified through a case study on an offshore drilling system, showing that it outperformed conventional models and exhibited robustness against measurement errors. The proposed model is expected to be applicable to various dynamic systems.
Article
Engineering, Marine
Tatsuya Kaneko, Ryota Wada, Masahiko Ozaki, Tomoya Inoue
Summary: This study proposes a hybrid physics-based and machine learning model for real-time estimation of unmeasurable parts in dynamic systems. The performance of this hybrid model is verified for structural uncertainty through numerical experiments and an application case study. The introduction of interpretability and uncertainty concepts enhances the model's ability to explain the reality gap.
Article
Engineering, Industrial
Ihab Ragai, Abdallah S. Abdalla, Hussein Abdeltawab, Feng Qian, J. Ma
Summary: This paper investigates the correlation between turning process parameters and tool vibration, acoustic signal, and energy consumption using advanced monitoring technologies and machine learning methods. The results show that vibration signals are more sensitive to changes in spindle speed, power response is more sensitive to changes in feed rates, and both spindle speed and feed rates have combined effects on the acoustic signal.
JOURNAL OF MANUFACTURING SYSTEMS
(2022)
Article
Materials Science, Multidisciplinary
Sara Nasiri, Mohammad Reza Khosravani
Summary: This study introduces the application of machine learning in predicting the structural performance and fracture of additively manufactured components, with a focus on its use in predicting the mechanical behavior of 3D printed parts. Previous research on the application of ML in characterizing polymeric and metallic 3D-printed parts is reviewed and discussed to highlight potential limitations, challenges, and future perspectives for industrial applications of ML in additive manufacturing.
JOURNAL OF MATERIALS RESEARCH AND TECHNOLOGY-JMR&T
(2021)
Article
Automation & Control Systems
Viktor Kulisek, Petr Kolar, Pavel Vrba, Jan Smolik, Miroslav Janota, Milan Ruzicka, Martin Machalka
Summary: This paper explores the potential of hybrid materials combining steel or cast iron with fibre or particle composites for lightweight machine tool structural design with high damping ratio. Fiber composites and particle composites were successfully used to reduce component mass and improve damping ratio on single structural parts, while the improvement in damping ratio of hybrid components in assembly configuration is less significant. The importance of damping caused in the connecting interfaces was also highlighted in the study.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
S. Nandhini, S. Parthasarathy, S. Saravanan
Summary: The objective of this study is to develop a new technique that can collect data from different sensors on the ship and identify early signs of defective behavior. A Hybrid SVELM-GOLF technique is proposed for estimating the condition of the ship's cylinder. Simulation results demonstrate that the proposed method achieves higher prediction accuracy and precision compared to other approaches.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Review
Materials Science, Multidisciplinary
Sujeet Kumar Chaubey, Kapil Gupta
Summary: This paper presents a review of previous research on turning miniature cylindrical bars using thermoelectric-erosion based turning processes, highlighting process parameters, performance evaluation, and key findings.
Article
Green & Sustainable Science & Technology
Malte Toetzke, Nicolas Banholzer, Stefan Feuerriegel
Summary: A machine learning framework is developed to categorize and monitor global development aid activities based on their textual descriptions. This allows for the identification of unexplored topics and spatio-temporal disparities in aid, enabling evidence-based decisions targeting the Sustainable Development Goals.
NATURE SUSTAINABILITY
(2022)
Article
Chemistry, Analytical
Xiaoqiang Ji, Zhi Rao, Wei Zhang, Chang Liu, Zimo Wang, Shuo Zhang, Butian Zhang, Menglei Hu, Peyman Servati, Xiao Xiao
Summary: A point-of-care system for healthcare on flights is proposed, which monitors electrocardiogram, breathing, and motion signals through wearable devices. By processing the data and using LSTM-RNN for sleep apnea-hypopnea syndrome (SAHS) classification, the model achieves an accuracy of up to 84-85%.
Article
Automation & Control Systems
Carlos E. H. Ventura, Frederico C. Magalhaes, Alexandre M. Abrao, Berend Denkena, Bernd Breidenstein
Summary: This study investigates the impact of geometric parameters on cutting tool performance when turning hardened steel, finding that parameters such as hone radius projection, form factor, and perimeter ratio have significant effects on cutting force and temperature. The experimental and numerical results provide insights into the relationship between edge preparation and cutting forces, as well as chip temperature trends.
INTERNATIONAL JOURNAL OF ADVANCED MANUFACTURING TECHNOLOGY
(2021)
Article
Engineering, Mechanical
Filipe Figueiredo dos Santos, Sandro da Costa Silva, Alexandre Mendes Abrao, Berend Denkena, Bernd Breidenstein, Kolja Meyer
Summary: Surface integrity is crucial for the functional performance of mechanical components, and deep rolling is an important surface treatment method. The study found that deep rolling significantly improves surface finish, but excessive pressure and feed can increase roughness. Increasing the number of rolling passes, however, helps improve surface finish.
JOURNAL OF TRIBOLOGY-TRANSACTIONS OF THE ASME
(2021)
Article
Engineering, Manufacturing
Berend Denkena, Marc-Andre Dittrich, Siebo Stamm, Marcel Wichmann, Soren Wilmsmeier
Summary: This passage discusses the challenges faced by production systems, bio-inspired production systems, and the concepts of "Gentelligence" and "process-DNA," as well as their performance in different applications.
CIRP JOURNAL OF MANUFACTURING SCIENCE AND TECHNOLOGY
(2021)
Article
Engineering, Manufacturing
Berend Denkena, Fritz Schinkel, Jonathan Pirnay, Soren Wilmsmeier
Summary: This paper introduces a new method for process-parallel Flexible Job Shop Scheduling based on quantum computing optimization, showcasing its good performance and practicality through a scientific benchmark and application to a real use-case. A managerial insight demonstrates how this approach can be integrated into existing production planning and control IT infrastructure.
CIRP JOURNAL OF MANUFACTURING SCIENCE AND TECHNOLOGY
(2021)
Review
Engineering, Manufacturing
Berend Denkena, Benjamin Bergmann, Alexander Schmidt
Summary: This study investigates the capability of sensor fusion based on principal component analysis to monitor preload loss of single nut ball screws, by studying features of different preload levels of the ball screw through selecting different ball diameters. The results show that this method can reliably detect preload levels.
CIRP JOURNAL OF MANUFACTURING SCIENCE AND TECHNOLOGY
(2021)
Article
Engineering, Industrial
Berend Denkena, Patrick Ahlborn
Summary: This paper presents a novel linear-rotary direct drive for machine tools, which combines linear and rotary movement in one drive to enhance the overall dynamics of the machine and reduce installation space.
CIRP ANNALS-MANUFACTURING TECHNOLOGY
(2022)
Article
Engineering, Industrial
Carolin Kellenbrink, Nicolas Nuebel, Andre Schnabel, Philipp Gilge, Joerg R. Seume, Berend Denkena, Stefan Helber
Summary: This paper presents a cyber-physical system demonstrator for the maintenance, repair, and overhaul (MRO) of high-pressure turbine blades of aircraft engines. By using a virtual layer and a virtual twin, the system handles the variability in damage patterns and achieves individual, flexible, and economically optimized MRO actions. It showcases the combination and innovation of research results from multiple disciplines.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2022)
Article
Engineering, Multidisciplinary
Lars Schoenemann, Oltmann Riemer, Bernhard Karpuschewski, Per Schreiber, Heinrich Klemme, Berend Denkena
Summary: The research has found that ultra-precision cutting based on digital surface twins can accurately predict surface features and characteristics, thus supporting the development of tool offset compensation methods and improving productivity.
PRECISION ENGINEERING-JOURNAL OF THE INTERNATIONAL SOCIETIES FOR PRECISION ENGINEERING AND NANOTECHNOLOGY
(2022)
Article
Engineering, Manufacturing
B. Denkena, A. Kroedel, M. Wilckens
Summary: The use of larger CBN grains in grinding can improve material removal rates for hardened steel components and achieve higher depths of cut. Grinding with coarse grains results in lower process forces, higher residual stress, and rougher surfaces, with minimal wear observed. In some cases, using larger grains can enhance tool performance and allow for higher feed rates.
PRODUCTION ENGINEERING-RESEARCH AND DEVELOPMENT
(2021)
Article
Engineering, Electrical & Electronic
Berend Denkena, Benjamin Bergmann, Matthias Witt
Summary: To achieve increased automation and flexibility in production, it is necessary to monitor component-specific characteristics and evaluate signals highly correlated with process quality. The use of material-specific cutting force improves the sensitivity of confidence limits to process errors, allowing the force-sensitive machine to substitute the dynamometer for process monitoring.
Article
Engineering, Manufacturing
Berend Denkena, Marc-Andre Dittrich, Hai Nam Nguyen, Konrad Bild
Summary: Self-optimizing process planning uses machine learning models to correlate process parameters with surface quality, automatically adjusting optimal parameters to achieve target roughness.
PRODUCTION ENGINEERING-RESEARCH AND DEVELOPMENT
(2021)
Article
Engineering, Manufacturing
Berend Denkena, Bernd-Arno Behrens, Benjamin Bergmann, Malte Stonis, Jens Kruse, Matthias Witt
Summary: This study predicts variations in dimension and cavities during cross-wedge rolling of shafts based on measured tool pressure. Multi-linear regression models are developed to determine the resulting diameters of the shaft shoulder, showing better prediction accuracy than models based on meta-data. The sensor concept for a new cross-wedge rolling machine and the approach for monitoring machining processes of workpieces with dimensional variations are presented for upcoming studies.
PRODUCTION ENGINEERING-RESEARCH AND DEVELOPMENT
(2021)
Article
Engineering, Manufacturing
Berend Denkena, Alexander Kroedel, Steffen Heikebruegge, Kolja Meyer, Philipp Pillkahn
Summary: This study investigates the impact of machining parameters on surface topography after deep rolling, and introduces a novel tool concept to explore the predictability of surface topography for milled specimens. By adjusting parameters gradually, the minimum pressure and lateral displacement conditions were identified to ensure accuracy in predicting surface topography.
PRODUCTION ENGINEERING-RESEARCH AND DEVELOPMENT
(2021)
Article
Engineering, Manufacturing
B. Denkena, O. Pape, A. Kroedel, V. Boss, L. Ellersiek, A. Mucke
Summary: This study presents a dynamic multi-dexel based material removal simulation for additive manufacturing processes, which is able to predict high-resolution surface topography and stable parameters to improve efficiency and accuracy in repair operations.
PRODUCTION ENGINEERING-RESEARCH AND DEVELOPMENT
(2021)
Article
Engineering, Manufacturing
B. Denkena, B. Breidenstein, A. Kroedel, V. Prasanthan
Summary: The requirements for massive high-performance components are constantly increasing, and by joining different materials in one component, these contradictory requirements can be met. The final step of machining in the process chain of manufacturing hybrid components has a decisive influence on the surface and subsurface properties.
PRODUCTION ENGINEERING-RESEARCH AND DEVELOPMENT
(2021)
