Search

To address the problems of low accuracy in fault diagnosis of oil-immersed transformers, poor state perception ability and real-time collaboration during diagnosis feedback, a fault diagnosis method for transformers based on the integration of digital twins is proposed. Firstly, fault sample balance is achieved through Iterative Nearest Neighbor Oversampling (INNOS), Secondly, nine-dimensional ratio features are extracted, and the correlation between dissolved gases in oil and fault types is established. Then, sparse principal component analysis (SPCA) is used for feature fusion and dimensionality reduction. Finally, the Aquila Optimizer (AO) is introduced to optimize the parameters of the Kernel Extreme Learning Machine (KELM), establishing the optimal AO-KELM diagnosis model. The final fault diagnosis accuracy reaches 98.1013%. Combining transformer digital twin models, real-time interaction mapping between physical entities and virtual space is achieved, enabling online diagnosis of transformer faults. Experimental results show that the method proposed in this paper has high diagnostic accuracy and strong stability, providing reference for the intelligent operation and maintenance of transformers. [ABSTRACT FROM AUTHOR]

The advent of Internet of Things and artificial intelligence in the era of Industry 4.0 has transformed decision-making within production systems. In particular, many decisions that previously required significant human activity are now made automatically with minimal human intervention via so-called digital twins (DTs). In the context of maintenance and reliability modeling, this naturally calls for new paradigms that can be seamlessly integrated within DTs for decision-making. The input data for time to failure needed in reliability computations are directly collected from the work center in a digital setting and often do not satisfy a known distribution. A neural network (NN) is proposed here to bypass this difficulty within the DT. Further, an algorithm inspired from machine learning is employed to solve the underlying semi-Markov decision process, whose transition model is captured via the NN. Numerical studies are carried out to demonstrate the usefulness of the approach. Finally, convergence properties of the algorithm are analyzed mathematically. [ABSTRACT FROM AUTHOR]

The majority of Europe's building stock consists of facilities built before 2001, presenting a substantial opportunity for energy efficiency improvements during their operation and maintenance phase. Digitalizing these buildings with digital twin technology can significantly enhance their energy efficiency. Reviewing the applications and trends of digital twins in this context is beneficial to understand the current state of the art and the specific challenges encountered when applying this technology to older buildings. This study focuses on the application of digital twins in building operations and maintenance (O & M), emphasizing energy efficiency throughout the building lifetime. A systematic process to select 21 pertinent use-case studies was performed, complemented by an analysis of six enterprise-level digital twin solutions. This was followed by an overview of general characteristics, thematic classification, detailed individual study analyses, and a comparison of digital twin solutions with commercial tools. Five main applications of digital twins were identified and examined: component monitoring, anomaly detection, operational optimization, predictive maintenance and simulation of alternative scenarios. The paper highlights challenges like the reliance on Building Information Modeling (BIM) and the need for robust data acquisition systems. These limitations hinder the implementation of digital twins, in particular in existing buildings with no digital information available. It concludes with future research directions emphasizing the development of methods not solely reliant on BIM data, integration challenges, and potential enhancements through AI and machine learning applications. [ABSTRACT FROM AUTHOR]

Virtual and augmented reality digital twins are becoming increasingly prevalent in a number of industries, though the production of digital-twin systems applications is still prohibitively expensive for many smaller organisations. A key step towards reducing the cost of digital twins lies in automating the production of 3D assets, however efforts are complicated by the lack of suitable automated methods for determining the visual quality of these assets. While visual quality assessment has been an active area of research for a number of years, few publications consider this process in the context of asset creation in digital twins. In this work, we introduce an automated decimation procedure using machine learning to assess the visual impact of decimation, a process commonly used in the production of 3D assets which has thus far been underrepresented in the visual assessment literature. Our model combines 108 geometric and perceptual metrics to determine if a 3D object has been unacceptably distorted during decimation. Our model is trained on almost 4, 000 distorted meshes, giving a significantly wider range of applicability than many models in the literature. Our results show a precision of over 97% against a set of test models, and performance tests show our model is capable of performing assessments within 2 minutes on models of up to 25, 000 polygons. Based on these results we believe our model presents both a significant advance in the field of visual quality assessment and an important step towards reducing the cost of virtual and augmented reality-based digital-twins. [ABSTRACT FROM AUTHOR]

A digital twin is a digital representation that closely resembles or replicates a real world object by combining interdisciplinary knowledge and advanced technologies. Digital twins have been applied to various fields, including to the agricultural field. Given big data and systematic data management, digital twins can be used for predicting future outcomes. In this study, we endeavor to create an agricultural digital twin using mandarins as a model crop. We employ an Open API to aggregate data from various sources across Jeju Island, covering an area of approximately 185,000 hectares. The collected data are visualized and analyzed at regional, inter-orchard, and intra-orchard scales. We observe that the intra-orchard analysis explains the variation of fruit quality substantially more than the inter-orchard analysis. Our data visualization and analysis, incorporating statistical models and machine learning algorithms, demonstrate the potential use of agricultural digital twins in the future, particularly in the context of micro-precision and individualized agriculture. This concept extends the current management practices based on data-driven decisions, and it offers a glimpse into the future of individualized agriculture by enabling customized treatment for plants, akin to personalized medicine for humans. A digital twin represents a real world object using available data. Here, the authors develop a digital twin for mandaring orchards in Jeju island showing the value of individualized agriculture to predict fruit quality at tree level. [ABSTRACT FROM AUTHOR]

This paper focuses on the feasibility of deep neural operator network (DeepONet) as a robust surrogate modeling method within the context of digital twin (DT) enabling technology for nuclear energy systems. Machine learning (ML)-based prediction algorithms that need extensive retraining for new reactor operational conditions may prohibit real-time inference for DT across varying scenarios. In this study, DeepONet is trained with possible operational conditions and that relaxes the requirement of continuous retraining - making it suitable for online and real-time prediction components for DT. Through benchmarking and evaluation, DeepONet exhibits remarkable prediction accuracy and speed, outperforming traditional ML methods, making it a suitable algorithm for real-time DT inference in solving a challenging particle transport problem. DeepONet also exhibits generalizability and computational efficiency as an efficient surrogate tool for DT component. However, the application of DeepONet reveals challenges related to optimal sensor placement and model evaluation, critical aspects of real-world DT implementation. Addressing these challenges will further enhance the method's practicality and reliability. Overall, this study marks an important step towards harnessing the power of DeepONet surrogate modeling for real-time inference capability within the context of DT enabling technology for nuclear systems. [ABSTRACT FROM AUTHOR]

Demands for more accurate machine learning models have given rise to rethinking current modeling approaches that were deemed unsuitable, primarily due to their computational complexity and the lack of availability and accessibility to representative data. In Industry 4.0, rapid advancements in Digital Twin (DT) technologies and the pervasiveness of cost-effective sensor technologies have pushed the incorporation of artificial intelligence, particularly data-driven machine learning models, for use in smart manufacturing. However, the persistent issue with such models is their high sensitivity to the training data and the lack of interpretability in the outcomes, at times generating unrealistic results. The incorporation of knowledge into the machine learning pipeline has been earmarked as the most promising approach to address such issues. This paper aims to answer this call through a Knowledge-embedded Machine Learning (KML) framework for smart manufacturing, which embeds knowledge from experience and, or physics information into the machine learning pipeline, thus making the outcomes from these models more representative of real applications. The merits of KML were then presented through comparative studies showing its capability to outperform knowledge-based and data-driven models. This promising outcome led to the development of frameworks that can potentially incorporate KML for smart manufacturing applications such as Prognostics and Health Management (PHM) and DT, further supporting the usefulness of the proposed KML framework. [ABSTRACT FROM AUTHOR]

In this work, a soft sensor–based digital twin (DT) was developed to reduce the startup time in manufacturing plastic tubes and enable real-time product quality monitoring, i.e., the weight per unit length and the inner and outer diameters of the tube. An experimental campaign was conducted on a real tube extrusion line using three polyvinyl chloride (PVC) compounds and different process conditions, and machine learning regression algorithms were trained and tested to create the models of the extruder and the extrusion die the DT is based on. The characterization of the considered material, whose properties were given as input to the digital models, was carried out according to a procedure based only on the data collected by the production line. The DT was tested for the startup of the production of a single-layer tube and allowed to achieve the specified customer requirements (thickness and weight) in a few minutes. The proposed solution thus proved to be a valuable tool for reducing the setup time, thus increasing the efficiency of the process. [ABSTRACT FROM AUTHOR]

The increasing proportion of bifacial photovoltaic modules (Bi-PVM) in new projects makes the operation of photovoltaic system (PVS) more complicated, and it is difficult to accurately predict the power of the PVS. To solve this problem, this paper proposes a new power prediction method for PVS based on Bi-PVM. Firstly, the equal proportion digital twin model of the example project is constructed. The superposition principle is used to analyze the factors affecting the power generation performance of Bi-PVM, and the characteristic project is constructed according to the analysis results. Secondly, the bifacial correction coefficient is introduced to reduce the parameter error caused by Bi-PVM to the prediction model. On this basis, a power prediction machine learning model based on bidirectional gated recurrent unit (Bi-GRU) network is established. Finally, a simulation experiment is carried out on the TensorFlow machine learning platform. With the actual operation data of a PV power station in Jiuquan, China, the simulation analysis is carried out under four weather types, namely, sunny day, rainy day, snowy day and complex and changeable day, respectively, which verified the correctness and excellence of the proposed method. [ABSTRACT FROM AUTHOR]

This document discusses the emergence of deep learning as a transformative tool in the field of operations research (OR). Deep learning, a specialized category of machine learning algorithms, uses multiple layers to uncover hidden patterns and insights from large datasets. It has gained popularity due to its superior predictive performance compared to traditional machine learning models. The integration of deep learning applications into OR has the potential to enhance decision-making, productivity, sustainability, and optimization processes. However, there are challenges to address, such as data availability, interpretability, and computational cost. The document highlights the need for practical research and provides future research directions in areas such as supply chain management, demand forecasting, inventory optimization, and decision-making processes. Overall, deep learning has the potential to transform the field of OR, and this document aims to provide new perspectives for further research and practical applications. [Extracted from the article]

Muscle electrophysiology has emerged as a powerful tool to drive human machine interfaces, with many new recent applications outside the traditional clinical domains, such as robotics and virtual reality. However, more sophisticated, functional, and robust decoding algorithms are required to meet the fine control requirements of these applications. Deep learning has shown high potential in meeting these demands, but requires a large amount of high-quality annotated data, which is expensive and time-consuming to acquire. Data augmentation using simulations, a strategy applied in other deep learning applications, has never been attempted in electromyography due to the absence of computationally efficient models. We introduce a concept of Myoelectric Digital Twin - highly realistic and fast computational model tailored for the training of deep learning algorithms. It enables simulation of arbitrary large and perfectly annotated datasets of realistic electromyography signals, allowing new approaches to muscular signal decoding, accelerating the development of human-machine interfaces. Muscle electrophysiology is a promising tool for human-machine approaches in medicine and beyond clinical applications. The authors propose here a model simulating electric signals produced during human movements and apply this data for training of deep learning algorithms. [ABSTRACT FROM AUTHOR]

Modern manufacturing has to cope with dynamic and changing circumstances. Market fluctuations, the effects caused by unpredictable material shortages, highly variable product demand, and worker availability all require system robustness, flexibility, and resilience. To adapt to these new requirements, manufacturers should consider investigating, investing in, and implementing system autonomy. Autonomy is being adopted in multiple industrial contexts, but divergences arise when formalizing the concept of autonomous systems. To develop an implementation of autonomous manufacturing systems, it is essential to specify what autonomy means, how autonomous manufacturing systems are different from other autonomous systems, and how autonomous manufacturing systems are identified and achieved through the main features and enabling technologies. With a comprehensive literature review, this paper provides a definition of autonomy in the manufacturing context, infers the features of autonomy from different engineering domains, and presents a five-level model of autonomy — associated with maturity levels for the features — to ensure the complete identification and evaluation of autonomous manufacturing systems. The paper also presents the evaluation of a real autonomous system that serves as a use-case and a validation of the model. [ABSTRACT FROM AUTHOR]

Among the diverse challenges in machining processes, chatter has a significant detrimental effect on surface quality and tool life, and it is a major limitation factor in achieving higher material removal rate. Early detection of chatter occurrence is considered a key element in the milling process automation. Online detection of chatter onset has been continually investigated over several decades, along with the development of new signal processing and machining condition classification approaches. This paper presents a review of the literature on chatter detection in milling, providing a comprehensive analysis of the reported methods for sensing and testing parameter design, signal processing and various features proposed as chatter indicators. It discusses data-driven approaches, including the use of different techniques in the time–frequency domain, feature extraction, and machining condition classification. The review outlines the potential of using multiple sensors and information fusion with machine learning. To conclude, research trends, challenges and future perspectives are presented, with the recommendation to study the tool wear effects, and chatter detection at dissimilar milling conditions, while utilization of considerable large datasets—Big Data—under the Industry 4.0 framework and the development of machining Digital Twin capable of real-time chatter detection are considered as key enabling technologies for intelligent manufacturing. [ABSTRACT FROM AUTHOR]

Multi-access edge computing (MEC) has unique interests in processing intensive and delay-critical computation tasks for vehicular application through computation offloading. However, due to the spatial inhomogeneity and dynamic mobility of connected vehicles (CVs), the edge servers (ESs) must dynamically adjust their resource allocation schemes to effectively provide computation offloading services for CVs. In this case, we propose a MEC framework supporting the collaboration of CVs, and incorporate digital twins (DTs) into wireless network to mitigate the unreliable long-distance communication between CVs and ESs. To solve the contradiction between the task change requirements of CVs and ES resources, we proactively balance the computation resources load of ESs by appropriately cooperative route planning of CVs, and achieve cross-domain load balancing between traffic flow and edge cloud resources domains. Furthermore, we jointly formulate route planning and resource allocation to balance the travel and service time delay by considering the mobility of CVs, distributed resources of ESs and the deadline sensitive vehicular tasks comprehensively. Besides, considering the coupled relationship between route planning and resource allocation, an alternating optimization algorithm is proposed to solve the formulated problem. we decompose it into two sub-problems. Firstly, a reinforcement learning method is used to optimize the route planning of CVs with fixed resource allocation. Then, an online learning and iterative algorithm is used to optimize the resource allocation strategy of edge cloud with fixed route selection. In order to demonstrate that our suggested scheme is more effective than other comparison schemes, a comprehensive series of experiments are conducted. [ABSTRACT FROM AUTHOR]

This paper presents a predictive computational framework for surrogate modeling of pressure field and optimization of pressure sensor placement for wind engineering applications. Firstly, a machine learning-derived surrogate model, trained by high-fidelity simulation data using finite element-based CFD and informed by a turbulence model, is developed to construct the full-field pressure from scattered sensor measurements in near real-time. Then, the surrogate pressure model is embedded in another neural network (NN) for optimizing pressure sensor placement. The goal of the NN-based optimizer is to learn the best layout of a fixed number of pressure sensors over the structural surface to deliver the most accurate full-field pressure prediction for various inflow wind conditions. We deploy the model to a representative low-rise building subjected to different wind conditions. The performance of the proposed framework is assessed by comparing the predicted results with finite element-based CFD simulation results. The framework shows excellent accuracy and efficiency, which could be potentially integrated with structural health monitoring to enable digital twins of civil structures. [ABSTRACT FROM AUTHOR]

Self-configuration is the preparation required to facilitate smart-manufacturing (SM) with the inputs derived without user intervention for engineering applications. Thus, it is vital for achieving the highest maturity level of SM technologies. In context, digital twin (DT) is an advanced virtual factory with simulation as its core technical functionality. However, the requirement of several inputs limits the implementation of DT on a physical asset without user intervention. Moreover, surpassing this limitation requires extraction methods for deriving the necessary inputs for DT application. Therefore, this study proposes information fusion and systematic logic library (SLL)-generation methods to facilitate the self-configuration of an autonomous DT. The information fusion aggregates and extracts the information elements required for DT application from heterogeneous information sources. In addition, the SLL generation method created the SLL required for reflecting the functional units of agents within the physical asset. Both methods were proposed from available SM standards such as ISA-95, automation markup language, and OPC unified architecture. Furthermore, an autonomous DT-supporting framework was designed by analyzing the relationship between asset description and SM standards, which facilitated the artificial intelligence-based extraction of the asset description object and SLL. Additionally, the core functional engines within this framework were designed using machine learning and process-mining techniques. Consequently, the proposed methods reduced the input pre-processing time required for constructing and synchronizing an autonomous DT to aid the application of autonomous DT on the physical asset without user intervention. [ABSTRACT FROM AUTHOR]

In this article, a digital twin approach is proposed for modeling a pharmaceutical drying process using machine learning techniques, driven by data from different sensors captured in-line. The current difficulty with the drying process is mainly due to the manual operator control for choosing the end-point for terminating the drying step. This results in significant variability, depending on which human operator is supervising, and due to the multi-tasking nature of the job, generally allows a longer processing time to be sure the material is completely dry. The objective is to automate the end-point identification, thus optimizing the processing duration and therefore the overall energy consumption of the process. The point at which the drying is complete is indicated by the temperature difference between the ingoing and outgoing air flow. However, the stochastic nature of the process makes the data modeling a challenge. Firstly, a wide selection of supervised statistical and machine learning algorithms was benchmarked to find the one (CatBoost) which gave the best performance with the data. Next, the set of hyper-parameters was found for CatBoost which gave the optimum performance. This gave a best performance of 0.788 (R2) fitting of the drying end-point prediction with the real values, for a large number of batches (over 700 K records). This is considered a good result taking into account the high residual of data models for these data and the stochastic nature of the process. The approach has been deployed in a real setting digital twin to control the drying process cutoff in the production plant. The results show the viability of the approach for modeling the process and automatically identifying the optimum end-point for the drying process, thus achieving significant energy savings which have been quantified as approximately 3.7 MWh per year for the pharmaceutical company. [ABSTRACT FROM AUTHOR]

This paper develops an approach to improve a CNC machine's tool performance and slow down its degradation rate automatically in the Pre-Failure stage. A Deep Reinforcement Learning (DRL) agent is developed to optimize the machining process performance online during the Pre-Failure interval of the tool's life. The Pre-Failure agent that is presented in the proposed approach tunes the feed rate according to the optimal policy that is learned in order to slow down the tool's degradation rate, while maintaining an acceptable Material Removal Rate (MRR) level. The machine learning techniques and pattern recognitions are implemented to monitor and detect the tool's potential failure level. The proposed mechanism is applied to a CNC machine when turning Titanium Metal Matrix Composites (TiMMC). A CNC machine Digital Twin (DT) is developed to emulate the physical machine in the digital environment. It is validated with the physical machine's measurements. The proposed pre-failure mechanism is a model-free approach, which can be implemented in any machining process with fewer online computational efforts. It also validated on a wide range of cutting speeds, up to 15,000 RPM. Deployment of the proposed machine learning approach for the particular case study improves the tool's Time to Failure (T2F) by 40% and the MRR by 6%, on average, compared to the classical approach. [ABSTRACT FROM AUTHOR]

ANPR cameras allow the automatic detection of vehicle license plates and are increasingly used for law enforcement. However, also statistical data generated by ANPR cameras are a potential source of urban insights. In order for this data to reach its full potential for policy-making, we research how this data can be shared in digital twins, with researchers, for a diverse set of machine learning models, and even Open Data portals. This article's key objective is to find a way to anonymize and aggregate ANPR data in a way that it still can provide useful visualizations for local decision making. We introduce an approach to aggregate the data with geotemporal binning and publish it by combining nine existing data specifications. We implemented the approach for the city of Kortrijk (Belgium) with 43 ANPR cameras, developed the ANPR Metrics tool to generate the statistical data and dashboards on top of the data, and tested whether mobility experts from the city could deduct valuable insights. We present a couple of insights that were found as a result, as a proof that anonymized ANPR data complements their currently used traffic analysis tools, providing a valuable source for data-driven policy-making. [ABSTRACT FROM AUTHOR]