Search

Injection moulding is a very popular technology for shaping plastics. Its history stretches back to the nineteenth century, and, as a consequence, it has developed outside the framework of digitisation. In order to fully implement the concepts of Industry 4.0, we need to update these legacy technologies so that they can fully benefit from the developments inherent in the "Internet of Things" and allow the process of injection moulding to take full advantage of digital optimisation so that it can fit effectively in the digital factory. In this work, we explore the quantitative use of X-ray scattering as a technology that can provide dynamic and multiscale feedback during the injection moulding cycle to be able to exploit digital twin technology as a means to optimise the operational parameters involved in injection moulding and to enable digital design of moulds in the fullest sense. This manuscript provides a way to mark future work and draw these possibilities to a wider audience. [ABSTRACT FROM AUTHOR]

Digital twins play an ever-increasing role in maximising the value of measurement and synthetic data by providing real-time monitoring of physical systems, integrating predictive models and creating actionable insights. This paper presents the development and implementation of the Aerosense digital twin for aerodynamic monitoring of wind turbine rotor blades. Employing low-cost, easy-to-install microelectromechanical (MEMS) sensors, the Aerosense system collects aerodynamic and acoustic data from rotor blades. This data is analysed through a cloud-based system that enables real-time analytics and predictive modelling. Our methodological approach frames digital twin development as a systems engineering problem and utilises design patterns, design thinking, and a co-design framework from applied category theory to aid in the development process. The paper details the architecture, deployment, and validation of a 'Digital Shadow'-type twin with simulation/prediction functionalities. The solution pattern is discussed in terms of its implementation challenges and broader applicability. By providing a practical solution to integrating all the digital twin components into a holistic system, we aim to help wind energy specialists learn how to transform a conceptual idea of a digital twin into a functional implementation for any application. [ABSTRACT FROM AUTHOR]

In response to the low operating speed and poor stability of energy harvesting systems in smart grids, an energy harvesting optimization method based on improved convolutional neural networks and digital twin technology is proposed in the experiment. Firstly, a smart grid data transmission framework integrating digital twin technology is proposed. A digital twin mapping method based on time, data, and topology structure is used to realize the digital twin mapping at the device level of power grid. Through data synchronization and interaction between the physical power grid and the digital twin model, the operational efficiency and reliability of the power grid are improved. Then, the classical convolutional neural network and attention mechanism are used to comprehensively analyze the physical topology data in the smart grid energy acquisition system. The improved lightweight target detection model is combined to monitor the equipment status of the smart grid and extract key features. Simultaneously utilizing convolutional attention mechanism to dynamically adjust the feature weights of channels or spaces, completing the preprocessing of energy harvesting data. Finally, combined with energy harvesting and power grid switching system, the process of energy harvesting and power grid operation are optimized together. On the training and validation sets, when the channels exceeded 60, the proposed method achieved a system energy efficiency of 55% during operation. The system energy efficiency of the other three comparative algorithms was all less than 40%. In practical applications, as the energy transfer loss increased to 1.0, the system throughput increased to 50 bits. The electricity needs of different users were met, and the difference between power allocation and optimal power allocation was small, which was very reasonable. This proves that the research has effectively optimized the energy harvesting system in the smart grid, improving the efficiency and reliability of the system in practical applications of the smart grid. At the same time, in the increasingly severe energy problem, this system can further provide technical references for the utilization of renewable energy and help achieve the goal of sustainable energy. [ABSTRACT FROM AUTHOR]

In the context of Industry 4.0, the most important trend in the shipbuilding industry is to efficiently manufacture intelligent and more environmentally friendly ships. This requires the shipbuilding industry to achieve technological innovations in all aspects of the ship's life cycle and accelerate the digital transformation of the industry. Thanks to the development of new-generation information technology, digital twins have received widespread attention and application in the manufacturing industry. The excellent performance of digital twins in monitoring, diagnosis and decision support also provides strong support for the digital transformation of the manufacturing industry. However, the application and development of digital twin technology are constrained when faced with the complex and discrete manufacturing processes in the shipbuilding industry. To address the constraints in the shipbuilding industry, promote the application of digital twin technology, and support the digital transformation of the shipbuilding industry, a statistical analysis of publications linked to the shipbuilding industry and digital twin using the Web of Science and the Scopus are done in this paper. The application and development of digital twin in the shipbuilding industry was systematically analyzed, the problems restricting the development of digital twin in the shipbuilding industry was summarized. Finally, the development trend of digital twin in shipbuilding industry prospects. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Academic Library & Information Science is the property of Anhui University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In the multi type load information of hybrid microgrids, data loss or incompleteness may occur due to network congestion, signal interference, equipment failures, and other reasons. Especially with the continuous generation of new load data, gradually incorporating these new data into the existing aggregation process to achieve continuous updating and optimization of load information. Therefore, this article proposes a digital twin model construction method for incremental aggregation of multi type load information in hybrid microgrids under integrity constraints. The Leida criterion and cubic exponential smoothing method are used to preprocess various load data of hybrid microgrids, remove abnormal data, reduce data fluctuations, and make the data more interpretable. Establish integrity constraints for multiple load data of hybrid microgrids and extract load characteristics of hybrid microgrids. Based on these, establish a digital twin model for the incremental aggregation of multiple load information in a hybrid microgrid, and solve the model using an improved K-means algorithm to achieve continuous updating and optimization of load information. The experimental results show that the data sharing delay of this method is 0.12 s, the load is basically consistent with the actual value, and the relative error of the load data is 4%. [ABSTRACT FROM AUTHOR]

The prediction of photovoltaic (PV) system performance has been intensively studied as it plays an important role in the context of sustainability and renewable energy generation. In this paper, a digital twin (DT) model based on a domain-matched transformer is proposed using convolutional neural network (CNN) for domain-invariant feature extraction, transformer for PV performance prediction, and domain adaptation neural network (DANN) for domain adaptation. The effectiveness of the proposed framework is validated using a PV power prediction dataset. The results indicate an accuracy improvement of up to 39.99% in model performance. Additionally, experiments with varying numbers of timestamps demonstrate enhanced PV power prediction performance as parameters are continuously updated within the DT framework, offering a reliable solution for real-time and adaptive PV power forecasting. [ABSTRACT FROM AUTHOR]

With the increasing focus on sustainable development in society, intelligent domain perception and digital twin technology can be used to evaluate and optimize the design of office furniture. This study analyzed sensor data through machine-learning and data-mining techniques to identify patterns and trends in the office environment. Simultaneously, a digital twin model of office furniture partition space was established to simulate the usage of furniture partition space throughout its full lifecycle. When 50% of nodes fail, the minimum transmission energy mode was significantly better than the maximum greedy forwarding mode in terms of cumulative throughput. The distributed, event-based, unsupervised clustering algorithm successfully reduced communication energy consumption, and the lightweight gradient boosting machine algorithm achieved the best design optimization rate, with an improvement of 0.53%. The ratio of value-added time to non-value-added time increased by 56.3%. The study aimed to provide innovative ideas for the development of intelligent office environments, promote the design of office furniture toward intelligence and sustainability, and improve the adaptability and efficiency of the work environment. [ABSTRACT FROM AUTHOR]

Energy harvesting is possible through capable energy transfer materials, and one such impressive material is graphene, which has exhibited promising properties like unprecedentedly high theoretical surface area, enhanced electrical conductivity, thermal conductivity, mechanical stability, flexibility, recyclability, and so on. Herein, for the sake of everyone desirous of contributing to the field of graphene materials for high-speed energy storage devices, the fundamentals, analytics, synthesis, prospects, and challenges of energy storage cell design for fast charging of electric vehicles have been reviewed. To overcome the limitations of mechanistic models for energy-storage devices, empiricism conjointly with data-driven machine learning has been quite effective in multiple domains. The limitations in modeling of energy storage devices, in terms of swiftness and accuracy in their state prediction can be surmounted by the aid of machine learning. Conclusively, in the context of energy management, we underscore the significant challenges related to modeling accuracy, performing original computations, and relevant big data generation. This review, by dint of its futuristic insights, will help researchers to develop digital twin approach for sustainable energy management using energy storage technology toward dependable, economic, and scalable optimization processes at the multiple stages of energy conversion and energy management. [ABSTRACT FROM AUTHOR]

We present an exciting journey from 3D point‐cloud data (PCD) to the state of the art in graph neural networks (GNNs) and their evolution with explainable artificial intelligence (XAI), and 3D geometric priors with the human‐in‐the‐loop. We follow a simple definition of a "digital twin," as a high‐precision, three‐dimensional digital representation of a physical object or environment, captured, for example, by Light Detection and Ranging (LiDAR) technology. After a digression into transforming PCD into images, graphs, combinatorial complexes and hypergraphs, we explore recent developments in geometric deep learning (GDL) and provide insight into the application of these network architectures for analyzing and learning from graph‐structured data. We emphasize the importance of the explainability of these models and recognize that the ability to interpret and validate the results of complex models is a crucial aspect of their wider adoption. This article is categorized under:Technologies > Artificial Intelligence [ABSTRACT FROM AUTHOR]

This paper offers a comprehensive overview of recent advancements in digital twin technology applied to additive manufacturing (AM), focusing on recent research trends, methodologies, and the integration of machine learning. By identifying emerging developments and addressing challenges, it serves as a roadmap for future research. Specifically, it examines various AM types, evolving trends, and methodologies within digital twin frameworks, highlighting the role of machine learning in enhancing AM processes. Ultimately, the paper aims to underscore the significance of digital twin technology in advancing smart manufacturing practices. A total of 133 papers were identified for analysis through IEEExplore, ScienceDirect, Web of Science, and Google Scholar and web resource. Approximately 74% of the papers are journals and 21% are conferences and proceedings. Moreover, 78% of the journal papers were Q1 journals. The paper identifies the potential benefits of digital twins at different levels, the existing problems associated with implementing digital twin in additive manufacturing, recent advancements, the existing approaches, and the framework. This review provides a comprehensive overview of the current landscape of research in digital twin technology for additive manufacturing, utilizing the latest resources to identify cutting-edge developments and methodologies. Through an exploration of potential benefits and implementation challenges, the review offers valuable insights to researchers and practitioners in the field. Additionally, it contributes to the discourse by offering a nuanced discussion on future research directions, paving the way for further advancements. [ABSTRACT FROM AUTHOR]

The conventional approach to appearance prediction for 3D printed parts is to print a thin slab of material and measure its reflectance or transmittance with a spectrophotometer. Reflectance works for opaque printing materials. Transmittance works for transparent printing materials. However, the conventional approach does not work convincingly for translucent materials. For these, we need to separate scattering and absorption. We suggest printing a collection of thin slabs of different thicknesses and using these in a spectrophotometer to obtain the scattering and absorption properties of the material. A model is fitted to the measured data in order to estimate the scattering and absorption properties. To this end, we compare the use of Monte Carlo light transport simulation and the use of an analytic model that we developed from the theory of radiative transfer in plane-parallel media. We assess the predictive capabilities of our method through a multispectral photo-render comparison based on the estimated optical properties. [ABSTRACT FROM AUTHOR]

There have recently been rapid developments in smart healthcare systems, such as precision diagnosis, smart diet management, and drug discovery. These systems require the integration of the Internet of Things (IoT) for data acquisition, Digital Twins (DT) for data representation into a digital replica and Artificial Intelligence (AI) for decision-making. DT is a digital copy or replica of physical entities (e.g., patients), one of the emerging technologies that enable the advancement of smart healthcare systems. AI and Machine Learning (ML) offer great benefits to DT-based smart healthcare systems. They also pose certain risks, including security risks, and bring up issues of fairness, trustworthiness, explainability, and interpretability. One of the challenges that still make the full adaptation of AI/ML in healthcare questionable is the explainability of AI (XAI) and interpretability of ML (IML). Although the study of the explainability and interpretability of AI/ML is now a trend, there is a lack of research on the security of XAI-enabled DT for smart healthcare systems. Existing studies limit their focus to either the security of XAI or DT. This paper provides a brief overview of the research on the security of XAI-enabled DT for smart healthcare systems. It also explores potential adversarial attacks against XAI-enabled DT for smart healthcare systems. Additionally, it proposes a framework for designing XAI-enabled DT for smart healthcare systems that are secure and trusted. [ABSTRACT FROM AUTHOR]

We present a data-driven, in situ proximal multi-sensor digital soil mapping approach to develop digital twins for multiple agricultural fields. A novel Digital Soil CoreTM (DSC) Probe was engineered that contains seven sensors, each of a distinct modality, including sleeve friction, tip force, dielectric permittivity, electrical resistivity, soil imagery, acoustics, and visible and near-infrared spectroscopy. The DSC System integrates the DSC Probe, DSC software (v2023.10), and deployment equipment components to sense soil characteristics at a high vertical spatial resolution (mm scale) along in situ soil profiles up to a depth of 120 cm in about 60 s. The DSC Probe in situ proximal data are harmonized into a data cube providing vertical high-density knowledge associated with physical–chemical–biological soil conditions. In contrast, conventional ex situ soil samples derived from soil cores, soil pits, or surface samples analyzed using laboratory and other methods are bound by a substantially coarser spatial resolution and multiple compounding errors. Our objective was to investigate the effects of the mismatched scale between high-resolution in situ proximal sensor data and coarser-resolution ex situ soil laboratory measurements to develop soil prediction models. Our study was conducted in central California soil in almond orchards. We collected DSC sensor data and spatially co-located soil cores that were sliced into narrow layers for laboratory-based soil measurements. Partial Least Squares Regression (PLSR) cross-validation was used to compare the results of testing four data integration methods. Method A reduced the high-resolution sensor data to discrete values paired with layer-based soil laboratory measurements. Method B used stochastic distributions of sensor data paired with layer-based soil laboratory measurements. Method C allocated the same soil analytical data to each one of the high-resolution multi-sensor data within a soil layer. Method D linked the high-density multi-sensor soil data directly to crop responses (crop performance and behavior metrics), bypassing costly laboratory soil analysis. Overall, the soil models derived from Method C outperformed Methods A and B. Soil predictions derived using Method D were the most cost-effective for directly assessing soil–crop relationships, making this method well suited for industrial-scale precision agriculture applications. [ABSTRACT FROM AUTHOR]

Deep, site-specific research is needed to understand ecosystem complexity. Improving conservation outcomes should be a driving force for the development of a holistic understanding of ecosystems. Holistic ecosystem models and digital twin ecosystems could guide conservation interventions in real-time. New digital and sensor technology provides a huge opportunity to revolutionise conservation, but we lack a plan for deploying the technologies effectively. I argue that environmental research should be concentrated at a small number of 'super-sites' and that the concentrated knowledge from super-sites should be used to develop holistic ecosystem models. These, in turn, should be morphed into digital twin ecosystems by live connecting them with automated environmental monitoring programmes. Data-driven simulations can then help select pathways to achieve locally determined conservation goals, and digital twins could revise and adapt those decisions in real-time. This technology-heavy vision for 'smart conservation' provides a map toward a future defined by more flexible, more responsive, and more efficient management of natural environments. [ABSTRACT FROM AUTHOR]

Background: Digital twin technology heralds a transformative era in Otorhinolaryngology (ORL), merging the physical and digital worlds to offer dynamic, virtual models of physical entities or processes. Purpose: These models, capable of simulating, predicting, and optimizing real-world counterparts, are evolving from static replicas to intelligent, adaptive systems. Methods: Fueled by advancements in communication, sensor technology, big data analytics, Internet of Things (IoT), and simulation technologies, artificial intelligence (AI), digital twins in ORL promise personalized treatment planning, virtual experimentation, and therapeutic intervention optimization. Despite their potential, the integration of digital twins in ORL faces challenges including data privacy and security, data integration and interoperability, computational demands, model validation and accuracy, ethical and regulatory considerations, patient engagement, and cost and accessibility issues. Results: Overcoming these challenges requires robust data protection measures, seamless data integration, substantial computational resources, rigorous validation studies, ethical transparency, patient education, and making the technology accessible and affordable. Looking ahead, the future of digital twins in ORL is bright, with advancements in AI and machine learning, omics data integration, real-time monitoring, virtual clinical trials, patient empowerment, seamless healthcare integration, longitudinal data analysis, and collaborative research. Conclusion: These developments promise to refine diagnostic and treatment strategies, enhance patient care, and facilitate more efficient and tailored ORL research, ultimately leading to more effective and personalized ORL management. [ABSTRACT FROM AUTHOR]

A digital twin-driven lifecycle management method is proposed to monitor the lifespan of motorized spindle. The lifecycle management system is designed and developed based on the combined programming of MATLAB, ANSYS, and LabVIEW. The prediction models of lifespan are built according to the digital twin (DT) data of thermal characteristics of the motorized spindle. The DT for thermal characteristics is realized through correcting the thermal boundary conditions according to the correction models which are established according to the measured and simulated temperatures at thermal key points. The temperature domain and threshold models of the cooling channel are constructed using the exponential fitting method to monitor the working status of the cooling system. The experimental results show that the DT-driven lifecycle management system can effectively monitor the status and remaining useful life of the motorized spindle, which provides a basis for the lifecycle management of the motorized spindle. [ABSTRACT FROM AUTHOR]

The manufacturing process of light-curing additive manufacturing leads to the inevitable presence of steps on the surface of the produced models, which affects the surface quality of the product. In the context of mass production, evaluating the surface quality of a model requires a large number of experimental samples to obtain data, which leads to material waste and extended production cycles. To facilitate the acquisition of surface feature data of a solid model, a digital twin system is proposed in this study. The system is specifically applied to light-curing additive manufacturing to generate a digital model with the surface features of a solid model. It aims to realize the replacement of the solid model with a virtual model for surface quality inspection. To realize the concept of generating high-fidelity digital models, a digital twin light-curing molding process was designed. It simulates the principle of light-curing molding and generates a digital model containing the surface features of a solid product based on the slice profile data. Case studies are conducted to build a digital twin system for the liquid crystal display light-curing additive manufacturing molding process to verify the feasibility of the proposed scheme and the fidelity of the digital model. The results of the comparison test between the digital model and the solid model show that the digital model produced by the digital twin system of the LCD light-curing additive manufacturing molding process contains the key features of the solid model. The digital model serves as the foundational data to verify whether the physical model meets the requirements. [ABSTRACT FROM AUTHOR]

Digital twins (DT) have increasingly garnered attention in today's manufacturing industry for their potential to enhance productivity, efficiency, and decision-making processes. This study presents a human-centric digital twin framework (HCDT) specifically designed for a three-axis vertical milling machine (VMC3), following the ISO 23247 standard. Our system seeks to offer a comprehensive virtual depiction of a tangible milling machine, facilitating immediate monitoring, analysis, and enhancement of machining operations. This study targets specifically the prediction of power consumption in three-axis vertical machining using advanced machine learning. Power consumption during modified Kakino toolpath cutting was examined through a series of trials that involved adjusting spindle speed, depth of cut, and feed rate. The analysis was conducted utilizing physical sensors, mathematical modeling, and machine learning. The utilization of machine learning methods, specifically random forest, exhibited encouraging outcomes, as evidenced by a mean absolute error (MAE) of 17.30. In addition, virtual simulations were performed to forecast power usage. The proposed applied digital twin (ADT) incorporates human intelligence and artificial intelligence to effectively integrate physical and virtual environments, offering a unique approach that adheres to the ISO 23247 framework. The cross-system entity of the digital twin showcases that the calculated power has large variations with the experimental power. However, the proposed ADT approach has resulted in higher similarity to the calculated, experimental, and predicted power consumption respectively in the proposed digital twin scenario. Also, the addition of explainability to the result has developed the trustworthiness in the digital twin environment. [ABSTRACT FROM AUTHOR]

Robotic manipulators play an innovative role as a new method for high-precision, large-scale manufacturing of composite components. However, machining composite materials with these systems presents unique challenges. Unlike traditional monolithic materials, composites exhibit complex behaviour and inconsistent results during machining. Additionally, robotic manipulator as a machine tool often associates with stiffness and vibration issues which adds another layer of complexity to this approach. By employing a comprehensive analysis and a combination of quantitative and qualitative review methodology, this review paper aims to survey diverse properties of composite materials by different categories and their interaction with machining processes. Subsequently, a survey of manufacturing techniques for composite machining following with a review in various modeling practices to capture material machining behaviour under a systematic framework is presented. Thereafter, the reviewed literature examines the errors inherent in robotic systems, alongside ongoing research efforts in modeling to characterise robot behaviour and enhance its performance. Afterward, the paper explores the application of data-driven modelling methods, with a primary focus on digital twins, in enabling real-time monitoring and process optimisation. Finally, this paper aims to identify the gap in this field and suggests the potential routes for future research and application as well as their challenges. [ABSTRACT FROM AUTHOR]

Offsite construction (OSC) techniques are argued to provide superior quality and shorter schedules compared with traditional techniques. Nonetheless, the pace of OSC implementation has been slow due to the influence of several barriers. In recent years, virtual reality (VR) applications have been used to address many of these barriers and promote the implementation of OSC projects. However, a comprehensive and coherent literature review that establishes the current state and categorizes VR applications in OSC projects is still lacking. To address this research gap, this study provides a state-of-the-art review of VR applications in OSC (VR–OSC) using the scientometric and systematic review methods. This study characterizes the synthesis between VR and OSC and identifies research trends and gaps that can be studied in future VR–OSC research. The scientometric review focuses on identifying the main topics of both research domains separately and combined based on the collected articles. The systematic review, meanwhile, qualitatively evaluates these articles, highlighting the existing research gaps and anticipating future research frontiers. The scientometric results indicate that VR applications in OSC can be organized into a number of clusters, such as Crane Operations and Onsite Planning, Educational Applications, Safety and Ergonomics, and Evaluation of Design Alternatives. The qualitative analysis identifies several future research directions to advance the field of VR–OSC, including (1) multiuser VR models in crane operation planning, (2) consideration of the role of human emotions in VR safety training by adopting biometric sensors, (3) decentralized web-VR platforms for remote OSC planning, and (4) VR-solutions for modeling robotic movements in OSC factories. This study can serve as a useful point of reference for VR–OSC researchers and provides a sound foundation for future research on VR–OSC. Practical Applications: As the market share of OSC techniques continues to grow, it is vital for construction professionals to stay up to date with the latest advances and technologies so that they are well-positioned to identify and address adoption barriers. This study provides a comprehensive analysis of the two topics (i.e., OSC and VR) both independently and integrally. The review of VR applications and their categorization into several OSC subdomains, such as crane operations management, safety management, processes planning, and educational practices, provides OSC practitioners with a holistic understanding on the basis of which to employ VR solutions more efficiently and effectively in their projects. Furthermore, the integration of VR with other technologies, such as eye-tracking technologies, motion capture sensors, and robotic tools, illustrates how VR tools can be used to address many OSC-related problems, including complex production lines, inefficient safety training, and overlapping crane operations. Finally, the improved understanding of current gaps in the domain of VR–OSC can help to galvanize industry support for more in-depth research on the use of VR in OSC. [ABSTRACT FROM AUTHOR]

A Distributed Intelligent System (DIS) encompasses a set of intelligent subsystems and components that collaborate to perform tasks and solve problems. Given the advancements of paradigms such as the Internet of Things, along with the advancements of technologies such as Machine Learning and Digital Twins, DISs are on the rise. These systems are increasingly integrating components that perform intelligent functions, and these intelligent functions are increasingly heterogeneous and varied. Moreover, there is no standardized framework to help researchers and practitioners adequately address DISs. As a result, the complexity, interoperability issues, and development time and costs of these systems are growing. However, Model-Driven Development (MDD) can help to address these challenges by providing a Domain-Specific Language (DSL) for developing DISs. In this work, a DSL for the design, validation, generation, and deployment of DISs is proposed. Firstly, the proposed DSL captures in a metamodel the key and high-level abstract concepts of the distinct DISs documented in the literature. Then, it allows modeling of DISs conforming to this metamodel. Subsequently, the DSL enables formal validation of the modeled systems. Lastly, it allows the generation and deployment of all DISs into production. Therefore, the work undertaken in this communication provides a methodological, formal, and standardized approach to defining and developing DISs from a high level of abstraction. This work allows users to address DISs by facilitating agility, minimizing manual tasks, and reducing the number of defects introduced in their development. To illustrate the applicability of the proposed DSL, a real case study of an agricultural digital twin is presented. [ABSTRACT FROM AUTHOR]

Copyright of e & i Elektrotechnik und Informationstechnik is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of GeSec: Revista de Gestao e Secretariado is the property of Sindicato das Secretarias e Secretarios do Estado de Sao Paulo (SINSESP) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

This study presents a safety management model for hydrogen refueling stations, integrating digital twin technology and artificial intelligence (AI) to enhance operational safety. Given the risks associated with high-pressure gas handling and potential fires from hydrogen leaks, real-time safety monitoring is crucial. The proposed model is based on a digital twin, a virtual replica of the physical system using real-time data, including temperature, pressure, and state of charge, collected from an actual hydrogen refueling station in Samcheok, Gangwon Province. Out of nine tested machine learning and deep learning algorithms, the convolutional neural network (CNN) demonstrated the highest performance (accuracy: 1, F1 score: 0.993) for risk prediction. Using AI libraries like Scikit-Learn and TensorFlow, the model achieved prediction times of 68 milliseconds, enabling decision-making at intervals of 1 s. Developed with the Unity 3D modeling tool, the digital twin visualizes predicted risk situations, allowing users to quickly identify and respond to potential hazards. This approach offers a robust solution for improving the safety of hydrogen refueling stations. [ABSTRACT FROM AUTHOR]

Global challenges related to sustainable development are increasingly focusing on the use of digital twin technology as a universal tool for optimizing and monitoring renewable energy installations. This article discusses digital twin technology as a support for sustainable development based on the analysis of microgrid structures. Digital twins allow the creation of virtual models of physical systems. This capability facilitated the accurate replication of the microgrid model at Kielce University of Technology using ETAP (Electrical Transient Analyzer Program) software (version 22.5). The operational parameters of the microgrid structure were analyzed for the examined power range of the photovoltaic installation to determine the possibilities of expanding the existing installation. The impact of the photovoltaic installation's power on the operational parameters of the microgrid structure was visualized, and final conclusions were formulated. Moreover, the integration of digital twin technology into renewable energy systems not only enhances operational efficiency but also plays a pivotal role in advancing sustainability objectives. Through real-time monitoring and predictive maintenance, digital twin technology facilitates the optimization of energy production and distribution, thereby reducing waste and contributing to the overall sustainability of energy systems. This technology enables the simulation of various scenarios, such as fluctuations in energy demand or the integration of new renewable sources, which can inform more sustainable decision-making processes. In the context of microgrids, digital twin technology ensures that energy production is closely aligned with consumption patterns, minimizing energy losses and enhancing grid resilience. Furthermore, digital twin technology supports the sustainable expansion of renewable installations by providing detailed insights into potential environmental impacts and the long-term sustainability of various energy configurations. As the demand for clean energy continues to grow, digital twin technology will be indispensable in achieving a balance between energy needs and environmental stewardship, ensuring that the expansion of renewable energy sources contributes positively to global sustainability objectives. [ABSTRACT FROM AUTHOR]

This research investigates the use of digital twin (DT) technology to improve building energy management and analyse occupant behaviour. DTs perform the function of acting as virtual replicas of physical assets, which facilitates real-time monitoring, predictive maintenance, and data-driven decision-making. Consequently, energy performance and occupant comfort can be enhanced. This study evaluates the efficiency of DTs in optimising energy usage by a mix of a systematic literature review and scientometric analysis of 466 articles from the Scopus database. Among the main obstacles noted are interoperability issues, privacy and data quality difficulties, and the requirement for a more thorough integration of digital and physical interactions. The results highlight the necessity of standardised frameworks to direct DT implementations and suggest areas for further study, especially in the areas of improving cybersecurity and incorporating occupant behaviour into DT models. This research makes practical recommendations for using DT technology to increase energy efficiency and sustainability in the built environment. [ABSTRACT FROM AUTHOR]

Air exchange in pressurized water pipelines is an essential but complex aspect of pipeline modeling and operation. Implementing effective air management strategies can yield numerous benefits, enhancing the system's energy efficiency, reliability, and safety. This paper comprehensively evaluates an irregular profile pipeline filling procedure involving air-release through an air valve. The analysis includes real-time data tests and numerical simulations using Computational Fluid Dynamics (CFD). A Digital Twin model was proposed and applied to filling maneuvers in water installations. In particular, this research considers an often-overlooked aspect, such as filling a pipe with an irregular profile rather than a simple straight pipe. CFD simulations have proven to capture the main features of the transient event, which are suitable for tracking the air-water interface, the unsteady water flow, and the evolution of the trapped air pocket. Thus, they provide thorough and reliable information for real-time operational processes in the industry, focusing on the filling pressure and geometry of the air-valve hydraulic system. Additionally, this study provides details regarding the application of an efficient Digital Twin CFD approach, demonstrating its feasibility in optimizing the filling procedure in pipes with irregular profiles. [ABSTRACT FROM AUTHOR]

This article deals with the design and creation of digital twins of robotic stations. A literature review of digital twins, robot programming methods and laser tracker applications is presented. This paper shows that the construction of digital twins is closely related to one of the most popular methods of robot programming, i.e., off-line programming. In the case of digital twins of robotic stations, modeling accuracy and two-way communication with the real station proved to be crucial. The article proposes a methodology for solving the basic problem of off-line robot programming, i.e., the limited accuracy of the representation of the station and the details. The algorithm of proceeding in the case when the station already exists and its digital model is built and the case when the digital model is first created and the real solution is built on its basis is shown. According to the developed methodology, a digital twin of a real robotic station was created and the possibilities arising from the use of virtual tools were shown. The developed digital twin has the ability to communicate with advanced Matlab 2021-type tools, uses cloud solutions and virtual and augmented reality for training, simulates physical phenomena and provides the ability to accurately program robots off-line. [ABSTRACT FROM AUTHOR]

The dynamic performances of an innovative robotic total station (RTS) are assessed while tracking a moving target; also, the results of these performance evaluations are exploited to design its control system. The proposed methodology is novel and uses ad hoc virtual experiments implemented on the digital twin of the studied RTS. This methodology is general and is also applicable for the design of other mechatronic systems. The proposed ad hoc virtual experiments constitute a benchmark for assessing the tracking performances of pointing systems. The application of this methodology to the studied RTS will help complete the RTS's design by determining the most suitable actuators and the control system constants necessary to obtain the assigned dynamic performances. A 3D-printed prototype of the designed RTS was built and is shown in this paper. [ABSTRACT FROM AUTHOR]

The aim of this paper is to present examples and original solutions related to the application of the digital twin concept in designing safety systems for robotic stations. This paper includes a review of publications on robot safety systems and digital twins. Based on this review, it was concluded that further work in this area is justified. This paper demonstrates the use of a digital model of a robotic casting mold preparation station to design safety components for an industrial cell. A key element of this paper is the presentation of developed algorithms and their applications in building digital twins of existing robotic stations. By characterizing advanced safety systems used in robotic stations, an example of using a digital twin of a robotic station to create safety zones and so-called restricted zones for the robot was developed. As part of the research conducted, a real, comprehensive example of creating safety zones based on the robot's TCP paths was carried out. [ABSTRACT FROM AUTHOR]

This paper presents a critical review of literature on the emerging technology known as digital twin and its application in structural integrity management for marine structures. The review defines digital twin in relation to structural integrity management as a virtual representation of a physical structure that mirrors the same structural conditions in real time. Twinning is a dynamic process that involves reducing the discrepancy between the virtual representation and physical structure, which is achieved with the aid of monitored data. Regarding the state-of-the-art concerning marine structure applications, all require the creation of a finite element model to represent the physical structure. Several practical schemes for physical to virtual interconnection have been proposed, but few researchers have concentrated on virtual to physical feedback. In addition, most works have focused only on assessing the current states of structures. To address this, a digital twin-based monitoring framework is proposed and three key enabling technologies, namely model updating, real-time simulation, and data-driven forecasting are demonstrated using a numerical case study. Such technologies enable structural diagnostics, as well as prognostics, to support decision making such as inspection/maintenance planning. Based on the case study, the opportunities and associated challenges of digital twin are discussed. For instance, to fully exploit the potential of digital twin, challenges related to monitoring systems such as standardisation, enhanced redundancy for long-term application, and monitored data quality assurance need to be addressed. Further, because digital twin can avail a vast amount of data, a dedicated data mining capability should also be incorporated. [ABSTRACT FROM AUTHOR]

The Digital Twin (DT) technology has shown promise in transforming complex engineered systems. However, its adoption in the Architecture, Engineering, Construction, and Operation (AECO) field, particularly for Built Cultural Heritage (BCH) conservation, is still in its early stages. This study presents a systematic literature review of 85 academic publications to evaluate the current state of DT implementation in heritage building conservation and identifies areas for optimising preventive management. In addition, this review explores interpretations of the DT concept in this field, addressing overlaps with Historical/Heritage Building Information Modelling (HBIM), discusses DT functionalities and presents existing frameworks. The findings demonstrate the potential of DT to revolutionise BCH conservation through a holistic approach. However, further focus is needed on features and tools for enhancing performance-based management with targeted strategies and advanced data analysis. Future research should prioritise developing these aspects to fully leverage the potential of the DT paradigm in BCH conservation. [ABSTRACT FROM AUTHOR]

In order to improve the power generation efficiency of the gas turbine,and to solve the problem of the high cost of the operation and maintenance (O&M) of the compressor water washing system, a digital twin-based O&M decision study of the compressor was conducted. And a health management framework for gas turbine in power plants based on digital twin was proposed. Based on that the compressor operation data was processed, and the extreme gradient boosting algorithm was used to build a prediction model for the washing cycle, and some of the parameters within the dataset were selected as inputs to the model, with the gas consumption as the output quantity, analyzed the change rule and its relationship with the input quantity. the water washing cycle and water washing recovery rate were calculated and compared, and the appropriate water washing cycle for O&M decision of the compressor was derived. The model prediction results showed that: the average R2 _score of the eight water washing gas consumption prediction reached 0. 98, and the prediction results were accurate. Among the eight times of water washing, the second and third water washing cycles were appropriate, and the third water washing recovery rate was optimal, resulting in the guiding hours of gas turbine compressor water washing cycle of 1 824 hours. Compared with the average water washing cycle of the power plant in the actual implementation, the cost of water washing could be reduced by 21. 9 million yuan per time. [ABSTRACT FROM AUTHOR]

Wind power generation has become an indispensable part of the power supply side of the power grid. Due to the intermittent and uncertain characteristics of wind energy, short-term wind power prediction plays an important role in the stable operation of power system. By constructing the digital twin model, real-time and high-precision prediction of wind energy is realized. First, the genetic algorithm-support vector machine (GA-SVM) algorithm is used to build the model. Multidimensional sensors and meteorological stations of the virtual wind power generation system collected the meteorological data of the environment and updated the meteorological history database at the same time. Second, based on the collected meteorological data, the preliminary prediction results are obtained, and by searching in the historical database, the predicted value and the actual output value of wind turbines or wind farms under similar conditions are obtained. Finally, the prediction results of the GA-SVM are modified to obtain the predicted value of the digital twin model. The prediction method can greatly improve the accuracy of the short-term forecast of wind energy. [ABSTRACT FROM AUTHOR]

This teaching case examines how engineering consultants Helbling Technik AG overcame challenges in implementing a digital twin approach to mechatronics product development, saving costs, and reducing time to market. It challenges students to understand digital twins in mechatronic system development and their implications for processes, use cases, and service models. This real-world example is relevant for students and educators interested in the challenges and opportunities of digital twin implementation. [ABSTRACT FROM AUTHOR]

Numerical simulations are crucial for predicting outcomes in forging processes but often neglect dynamic interactions within forming tools and presses. This study proposes an approach for achieving accurate real-time prediction of forging outcomes. Initially, a simulation-based surrogate model is developed to replicate key process characteristics related to the billet, enabling prediction of geometry, deformation field, and forging load after an upsetting operation. Subsequently, this model is integrated with a mass-spring-damper model representing the behavior of forging machine and tools. This integration enables the prediction of blow efficiency and energy distribution after each blow, including plastic, elastic, damping, and frictional energy of the upsetting operation. The approach is validated by comparing predictions with experimental results. The coupled model outperformed Finite Element Method (FEM) predictions, exhibiting mean absolute errors (MAE) below 0.1 mm and mean absolute percentage errors (MAPE) below 1% in geometry predictions. Deformation field predictions showed errors below 0.05 mm/mm, and load-displacement curves closely matched experimental data. Blow efficiency predictions aligned well with experimental results, demonstrating a mean absolute error below 1.1%. The observed energy distribution correlated with literature findings, underscoring the model’s fidelity. The proposed methodology presents a promising approach for accurate real-time prediction of forging outcomes. [ABSTRACT FROM AUTHOR]

Regenerative chatter is a prominent form of vibration in any machining process caused by the waviness of the workpiece surface. During internal turning processes, regenerative chatter constitutes a significant concern due to the highly flexible nature of boring bars and workpieces. As a result, the surface finish is poor, tool wear is accelerated, and machining at a high depth of cut is challenging. Typically, stability lobe diagrams (SLDs), plots of spindle speeds versus depth of cuts, are employed to select stable machining parameters. In the present work, a numerical linear stability analysis of a two-degree-of-freedom finite element model of a boring tool–workpiece system is carried out in the frequency domain to construct SLDs that are utilized to investigate how changing dynamics of the workpiece affect machining stability. The numerical model is validated using an analytical model, and the results of both models are found to be in agreement. Moreover, the results show that the depth of cut of a tool–workpiece system with moderately low and high workpiece stiffness is 11.8% and 16.2% lower than that of a single degree-of-freedom (DOF) system that does not take the workpiece into account. The SLDs of two DOF systems with workpieces of high and very high stiffness are observed to overlap with that of a single DOF system. It is also inferred that the damping ratio of the workpiece has no significant effect on the stability lobes. The machining experiments are conducted on a boring tool–workpiece system, and it is found that the experimental results align with the analytical and numerical findings. Additionally, a digital twin framework based on the computational model is suggested to facilitate the real-time monitoring and optimization of boring processes. [ABSTRACT FROM AUTHOR]

Development of digital twins is emerging rapidly in geotechnical engineering, and it often requires real-time updating of numerical models (e.g., finite element model) using multiple sources of monitoring data (e.g., settlement and pore water pressure data). Conventional model updating, or calibration, often involves repeated executions of the numerical model, using monitoring data from a specific source or at limited spatial locations only. This leads to a critical research need of real-time model updating and predictions using a numerical model improved continuously by multi-source monitoring data. To address this need, a physics-informed machine learning method called multi-source sparse dictionary learning (MS-SDL) is proposed in this study. Originated from signal decomposition and compression, MS-SDL utilizes results from a suite of numerical models as basis functions, or dictionary atoms, and employs multi-source monitoring data to select a limited number of important atoms for predicting multiple, spatiotemporally varying geotechnical responses. As monitoring data are collected sequentially, no repeated evaluations of computational numerical models are needed, and an automatic and real-time model calibration is achieved for continuously improving model predictions. A real project in Hong Kong is presented to illustrate the proposed approach. Effect of monitoring data from different sources is also investigated. [ABSTRACT FROM AUTHOR]

Airports have ambitious decarbonization objectives, with many EU airports pledging to reach net-zero carbon emissions by 2050 for all emissions fully under their control. This requires setting a baseline and defining roadmaps composed of decarbonization measures that intend to reduce energy demand and carbon emissions from that baseline. However, there is substantial uncertainty in defining the impact that can be achieved through these roadmaps, which can be substantially reduced through the use of a digital twin. In this paper, we present the creation of a baseline digital twin of the built environment of Brussels Airport, which will be used to analyze the impact of decarbonization measures before they are deployed. To create the digital twin, an extensive data collection exercise was carried out to ensure that the created model is a realistic representation of the core airport systems and that its simulations are accurate in forecasting results. Utilizing the IES Apache physics-based engine, the digital twin predictions have been validated against actual energy consumption data. The present work demonstrates the methodology and results achieved, offering insights on how to tackle the complex challenge of creating a digital twin of the built environment of an entity as complex as an airport while focusing the baselining efforts on the most significant aspects, with the ultimate objective of reducing uncertainty of future decarbonization investments. The digital twin has been used to calculate the energy consumption and CO2 emission baseline, and it will be utilized to model key decarbonization measures that are part of the Brussels Airport net zero carbon roadmap. [ABSTRACT FROM AUTHOR]

This paper explores the telemedicine possibilities available to astronauts during spaceflight, focusing on the possible utilization of medical avatars for continuous health monitoring. Various types and sources of telemedicine are examined, along with their applications in deep space travel to address the challenges faced by astronauts and facilitate timely intervention in case of injuries and illnesses. The use of medical avatars presents a promising solution to mitigate risks and ensure effective healthcare delivery in the absence of immediate physical medical support. Furthermore, potential terrestrial applications of telemedicine and virtual medical avatars are discussed, highlighting the wider implications and benefits beyond space exploration. By leveraging advanced technologies and innovative approaches, telemedicine in conjunction with medical avatars offers a compelling avenue for enhancing healthcare practices both in space and on Earth. [ABSTRACT FROM AUTHOR]

A digital process was developed to facilitate additive manufacturing for ceramic materials using digital light processing (DLP). A numerical model that predicts DLP sample properties can be generated from manufacturing inputs to forecast the effect of resin age on mechanical strength of the printed part based on data collected from experiments. Key parameters for printing the green bodies included determining the depth of cure, layer thickness, material composition, and solids loading. Thermogravimetric analyses were used to develop debinding and sintering curves. Debinding is used to remove the volatile organics comprising the photopolymer resin. Sintering is performed after debinding to increase density and mechanical strength of the printed parts. The sintered parts were then subjected to characterization and mechanical testing. The ensemble of data for various DLP-printed ceramic materials were added to a database. A design of experiments can be generated from the manufacturing process defined in the database with selected changeable parameters randomized over a range. Because the database is defined with an architecture to capture manufacturing processes, it can persist as a more generic platform for manufacturing digital twins. This can ease the development of future digital twins and can grow as a common repository for the insights gained from manufacturing research. Creating a digital twin of a DLP system for 3D printing parts enables manufacturers to simulate and assess the impact of resin age on printing parameters and part quality, facilitating optimization, predictive maintenance, and cost reduction. [ABSTRACT FROM AUTHOR]

AbstractThe 28 cooling towers of the French Nuclear Power Plants (NPP) fleet have been studied using a digital twin automated approach that allows a better understanding of their behaviour and includes on-site measurements such as settlement displacements and predicted corrosion as inputs in a linear Finite Element Model (FEM). Advanced simulations are then built upon this digital twin, providing a more precise simulation of the ultimate mechanical behaviour of the tower, with a non-linear FEM allowing steel and concrete damage. Finally, increasing wind load is applied until the structure reaches a near-collapse state, allowing us to conclude on the mechanical margin and the need for reinforcement. [ABSTRACT FROM AUTHOR]

As the power industry becomes increasingly complex, Digital Twin (DT) technology has emerged as a crucial tool for enhancing grid resilience and operational efficiency by creating dynamic digital replicas of physical systems. These replicas enable accurate simulations and proactive management, but the vast amount of data generated by DT systems poses significant challenges for processing and analysis. Cloud computing offers a flexible solution by offloading these computational tasks to distributed resources, allowing for real-time analysis and scalable operations. However, this approach introduces complexities in task distribution and maintaining quality of service. Recent efforts have applied Deep Reinforcement Learning (DRL) to address these challenges, primarily using single-agent methods. However, these methods struggle with scalability and performance in increasingly complex cloud environments. To overcome these limitations, we propose an efficient task scheduling framework based on Multi-Agent Deep Q-Network (MADQN) principles, specifically designed to optimize both response times and operational costs. We provide a comprehensive design overview of our approach and conduct a thorough evaluation of its performance. The experimental results clearly indicate that our approach can considerably reduce response times amd lower operational costs compared to current methods, including state-of-the-art single-agent approaches. [ABSTRACT FROM AUTHOR]

A new method was proposed to address fault diagnosis by applying the digital twin (DT) high-fidelity behavior and the deep learning (DL) data mining capabilities. Subsequently, the proposed fault distribution GAN (FDGAN) was built to map virtual and physical entities for the data from the established test platform. Finally, the MobileViG was employed to validate the model and diagnose faults. The accuracy of the proposed method with training samples of 600 and 800 were 88.4% and 99.5%, respectively. These accuracies surpass those of other methods based on CycleGAN (98.86%), CACGAN (94.92%), ACGAN (86.45%), ML1D-GAN (82.33%), and transfer learning (99.38%). Therefore, with the integration of global connectivity, an innovative network structure, and training methods, FDGAN can effectively address challenges such as network degradation, limited feature extraction in small windows, and insufficient model robustness. [ABSTRACT FROM AUTHOR]

Proliferative vitreoretinopathy (PVR) is a pathological process characterized by the formation of fibrotic membranes that contract and lead to recurrent retinal detachment. Pars plana vitrectomy (PPV) is the primary treatment, but recurrence rates remain high, as surgery does not address the underlying molecular mechanisms driving fibrosis. Despite several proposed pharmacological interventions, no approved therapies exist, partly due to challenges in conducting preclinical and in vivo studies for ethical and safety reasons. This review explores the potential of computational models and Digital Twins, which are increasingly gaining attention in medicine. These tools could enable the development of progressively complex PVR models, from basic simulations to patient-specific Digital Twins. Nintedanib, a tyrosine kinase inhibitor targeting PDGFR, VEGFR, and FGFR, is presented as a prototype for computational models to simulate its effects on fibrotic pathways in virtual patient cohorts. Although still in its early stages, the integration of computational models and Digital Twins offers promising avenues for improving PVR management through more personalized therapeutic strategies. [ABSTRACT FROM AUTHOR]

Human induced pluripotent stem cell‐derived cardiomyocytes (iPSC‐CMs) offer potential as an in vitro model for studying drug cardiotoxicity and patient‐specific cardiovascular disease. The inherent electrophysiological heterogeneity of these cells limits the depth of insights that can be drawn from well‐designed experiments. In this review, we provide our perspective on some sources and the consequences of iPSC‐CM heterogeneity. We demonstrate the extent of heterogeneity in the literature and explain how such heterogeneity is exacerbated by patch‐clamp experimental artifacts in the manual and automated set‐up. Finally, we discuss how this heterogeneity, caused by both intrinsic and extrinsic factors, limits our ability to build digital twins of patient‐derived cardiomyocytes. [ABSTRACT FROM AUTHOR]

We developed and validated digital twins (DTs) for contrast sensitivity function (CSF) across 12 prediction tasks using a data-driven, generative model approach based on a hierarchical Bayesian model (HBM). For each prediction task, we utilized the HBM to compute the joint distribution of CSF hyperparameters and parameters at the population, subject, and test levels. This computation was based on a combination of historical data (N = 56), any new data from additional subjects (N = 56), and “missing data” from unmeasured conditions. The posterior distributions of the parameters in the unmeasured conditions were used as input for the CSF generative model to generate predicted CSFs. In addition to their accuracy and precision, these predictions were evaluated for their potential as informative priors that enable generation of synthetic quantitative contrast sensitivity function (qCSF) data or rescore existing qCSF data. The DTs demonstrated high accuracy in group level predictions across all tasks and maintained accuracy at the individual subject level when new data were available, with accuracy comparable to and precision lower than the observed data. DT predictions could reduce the data collection burden by more than 50% in qCSF testing when using 25 trials. Although further research is necessary, this study demonstrates the potential of DTs in vision assessment. Predictions from DTs could improve the accuracy, precision, and efficiency of vision assessment and enable personalized medicine, offering more efficient and effective patient care solutions. [ABSTRACT FROM AUTHOR]

AbstractIn this work, we introduce a novel method to develop whole system digital twins (DTs) for advanced nuclear reactors. This method treats a complex reactor system as a heterogeneous graph: with the system components as different types of graph nodes and their physical interconnections as edges. Based on the heterogeneous graph, a graph neural network combining graph convolution and temporal node attention is developed as the DT, facilitating a comprehensive understanding of the system’s dynamic behavior. By utilizing the System Analysis Module (SAM) code for simulating various operational transients, we develop a graph-based database that trains the DT. This DT is characterized by two primary functions: It can infer the entire system’s status using sparse node information, and it can predict the progress of transients based on current and historical system information. Our approach is validated through case studies on the Experimental Breeder Reactor II (EBR-II) system and a generic Fluoride-salt-cooled High-temperature Reactor (gFHR), demonstrating the DT’s accuracy in forecasting operational transients. The DT’s rapid computation capabilities enhance its potential for supporting advanced reactor operations, offering benefits in intelligent simulation, autonomous control, and anomaly detection, paving the way for improved safety analysis and intelligent component health management for advanced reactor systems and reducing their operations and maintenance cost. [ABSTRACT FROM AUTHOR]

Accurate defect detection is crucial for ensuring the trustworthiness of intelligent railway systems. Current approaches either rely on static ensembling of Convolutional Neural Networks (CNNs) or on single deep-learning CNN models. Traditional methods use a large volume of data to identify underlying patterns. However, training a new defect classifier with limited samples often results in overfitting and poor performance on unseen images. To overcome these challenges, we introduce a dynamic weight adaptation mechanism within an ensemble learning framework, enhanced by state-of-the-art transfer learning models (VGG-19, MobileNetV3, and ResNet50). We obtained a good test and validation accuracy of 99%, accompanied by a precision of 0.99, and recall and F1-score of 0.98. Unlike previous methods, our approach demonstrates consistency. Particularly, techniques that involve selecting predictions based on minimum loss previously required a significant number of epochs to stabilize. Data fusion techniques are even more demanding, requiring a substantially higher number of epochs for stabilization. Conversely, our proposed method achieves stable performance and rapid convergence within just 10 epochs, and it shows minimal fluctuation in the training curve (0–5 epochs). This visibly contrasts with earlier methods, which, despite reaching similar levels of accuracy, required significantly more epochs and exhibited greater fluctuations in their training curves. Through these findings, we anticipate that the proposed dynamic weight adaptation-based ensemble approach will further research and development of reusable AI-enabled solutions for rail defect classification. [ABSTRACT FROM AUTHOR]