Application Architecture Research Articles

Charting new avenues in financial forecasting with TimesNet: The impact of intraperiod and interperiod variations on realized volatility prediction

This study evaluates TimesNet model for stock realized volatility forecasting, comparing its efficacy against traditional and contemporary models across key metrics: RMSE, MAE, MAPE, and QLIKE. Despite TimesNet’s competitive performance in RMSE and QLIKE, it faces challenges in MAE and MAPE, falling behind models like HAR, NBEATSx and NHITS. Highlighting the novel application of CNN architectures beyond image recognition, this research suggests TimesNet’s potential in long-term forecasting and extreme event prediction, yet indicates the necessity for further model enhancement. The findings motivates new avenues for future research into CNN-based forecasting models in stock realized volatility forecasting by improving the current version of the TimesNet model.

Comparison of modelling methodologies for large porous absorbers hanging in a volume

Porous material simulation methodologies are well established. However, most methods, like the Transfer-Matrix Method (TMM) make the assumption that there is a source side and a receiver side. In practice, this works for the majority of industrial noise control treatments. For architectural applications, this is not always the case. Hanging absorbers and curtains are common materials that add absorption to a space, but do not have a boundary condition at a wall or a well-defined source and receiver side. To evaluate the validity of existing simulation methods for this condition, a reverb room test was simulated with a large cube of melamine foam inside the volume. The effective absorption of the cube in the space was then calculated using various simulation methodologies, and compared to results from a measured reverb room absorption test.

Evaluation of alternative construction methods for architectural additive manufacturing applications

Evaluation of alternative construction methods for architectural additive manufacturing applications

The geometry and mechanics of the Chinese finger trap

We investigate the kinematics and mechanical mechanisms of cylindrically woven straps with a plain twill pattern, which form a tubular structure known as the Chinese Finger Trap. Our findings, derived from a combination of analytical, experimental, and numerical methods, demonstrate that when subjected to axial tension, the straps within the structure undergo reconfiguration, resulting in radial contraction. The geometry of the straps influences this contraction. As the structure compacts, the linear axial force transforms into an exponential force, resulting in a catenary-like profile in woven tubes due to the axially symmetric distribution of straps. Our study of this tension-induced contraction strategy proposes a straightforward approach to manufacturing morphable structures capable of efficiently converting axial elongation into radial contraction. This technique holds potential for medical, architecture, and soft robotics applications, offering accessible and controlled engagement and disengagement capabilities.

Towards a smart hospital: Smart infrastructure integration

In today’s business world, digital technologies have become vital for implementing processes in any company, including the healthcare industry. By utilizing the latest technologies to collect, process, analyze, and transmit data through smart devices and IT infrastructure, it is possible to implement trends in medicine as a value approach, personalization, predictability, and participation. Considering these trends, it is necessary to develop a methodology for creating a management system for medical organizations that can comprehensively and interconnectively develop various elements of the system. This includes business process systems, information systems and application architectures, hardware, and "smart" medical equipment that can fully participate in hospital information exchange. These elements should be developed within a single model. An architectural approach can be used to create a model of a medical organization. This paper examines the integration of smart infrastructure in medical organizations, develops requirements for smart infrastructure services, and creates architectural models based on the findings and requirements. Architecture models were developed for the following elements of the Smart Infrastructure: smart wards, wearable devices and telemedicine stands as the most necessary elements in everyday healthcare. The key discovery of the study is the architecture models of the mentioned elements and their integration with the medical organization IT architecture, described in the form of visual models and data exchange tables. The proposed models can be used both for further research or for practical implementation as the reference models to check up and adopt to each individual case. The solutions proposed in the paper give an example of open innovation philosophy, as they are developed on the basis of open information about the innovation activities of various medical organizations, and, at the same, time the results themselves are openly published and open for adaptation by the scientific and practicing medical community.

Evaluating the performance of multilingual models in answer extraction and question generation

Multiple-choice test generation is one of the most complex NLP problems, especially in languages other than English, where there is a lack of prior research. After a review of the literature, it has been verified that some methods like the usage of rule-based systems or primitive neural networks have led to the application of a recent architecture, the Transformer architecture, in the tasks of Answer Extraction (AE) and Question Generation (QG). Thereby, this study is centred in searching and developing better models for the AE and QG tasks in Spanish, using an answer-aware methodology. For this purpose, three multilingual models (mT5-base, mT0-base and BLOOMZ-560 M) have been fine-tuned using three different datasets: a translation to Spanish of the SQuAD dataset; SQAC, which is a dataset in Spanish; and their union (SQuAD + SQAC), which shows slightly better results. Regarding the models, the performance of mT5-base has been compared with that found in two newer models, mT0-base and BLOOMZ-560 M. These models were fine-tuned for multiple tasks in literature, including AE and QG, but, in general, the best results are obtained from the mT5 models trained in our study with the SQuAD + SQAC dataset. Nonetheless, some other good results are obtained from mT5 models trained only with the SQAC dataset. For their evaluation, the widely used BLEU1-4, METEOR and ROUGE-L metrics have been obtained, where mT5 outperforms some similar research works. Besides, CIDEr, SARI, GLEU, WER and the cosine similarity metrics have been calculated to present a benchmark within the AE and QG problems for future work.

ENHANCING MACHINE LEARNING PERFORMANCE: THE ROLE OF GPU-BASED AI COMPUTE ARCHITECTURES

This paper advances the field of GPU-based embedded intelligence (EI) by providing a comprehensive review of current and emerging architectures and applications. It covers key paradigms in GPU-based EI, focusing on architecture, technologies, and practical applications. The paper is structured as follows: (1) An overview and classification of GPU-based EI research, providing a broad perspective and concise summary of the paper's scope; (2) An in-depth discussion of various architectural technologies for GPU-based deep learning techniques and applications; and (3) A detailed examination of architectural technologies for GPU-based machine learning techniques and applications. This paper aims to offer valuable insights into the research area, encouraging further development of GPU-based EI for practical deployment and applications.

Structural computational analysis of conventional and self-designed rover wheel architectures for extraterrestrial roving application

The technical advancements in robotics have closely been followed by developments in aerospace excavation robots. This research elucidates self-designed shape memory alloy wheels for their use in Lunar Excavations, which attributes to the new era of space exploration robots. The wheel design proves to be of paramount importance as crucial operations such as mobility, steering, and traction control build upon wheel design. A comparison has been made between Aluminium alloy, Structural Steel, and Shape Memory alloy (Nitinol) wheels for both conventional and self-designed wheel architectures. Wheel architecture tests and their results are articulated with the aid of simulation. The significant performance parameters, primarily Total Deformation, Equivalent Stress, and Shear stress-induced, were analyzed, which articulates the performance of the wheel at the time of landing. It was observed that for both wheel designs: self-designed and conventional, the maximum total deformation was 0.011154 mm and 1.3573 mm, respectively, when Aluminium alloy was chosen as the base material. For Steel and Nitinol, the self-designed wheel structure proved to perform better than the conventional wheel design with lower values of total deformation (0.0041129 mm and 0.010968 mm, respectively).

Preparation of Hollow Silica Nanoparticles with Polyacrylic Acid and Their Moisture Sorption Properties

Hollow silica nanoparticles (HSNPs) have hygroscopic properties because of their high specific surface area and surface hydroxyl groups. However, compared with other hygroscopic materials, their hygroscopic properties are relatively weak, which limits the further application of HSNPs. One feasible method to enhance their hygroscopic properties is by combining highly hygroscopic materials with hollow silica nanoparticles. To take advantage of the high hygroscopicity of polyacrylic acid (PAA) when combined with the high specific surface area of the hollow particles, PAA was coated on the inner and outer surfaces of the silica shell of the nanoparticles in this study to prepare hollow nanoparticles with a PAA/silica/PAA multilayer structure. The size of the PAA/silica/PAA multi-layer nanoparticles is about 85 nm, and the shell thickness is 25 nm. The specific surface area of the multi-layer nanoparticles is 58 m2/g. The water vapor adsorption capacity of multi-layer structure hollow nanoparticles was increased by 160% compared with the HSNPs (increased from 45.9 cm3/m2 to 109.1 cm3/m2). Meanwhile, at the same content of PAA, the PAA/silica/PAA-structured particles will adsorb 9% more water vapor than the PAA/silica-structured particles. This indicates that the high specific surface area structure of the hollow particles will enhance the adsorption ability of PAA toward water vapor. This novel structure of PAA-HSNPs is expected to be used as a humidity-regulating material for filler in environmental and architectural applications.

Agile C-states: A Core C-state Architecture for Latency Critical Applications Optimizing both Transition and Cold-Start Latency

Latency critical applications running in modern datacenters exhibit irregular request arrival patterns and are implemented using multiple services with strict latency requirements (30 μs –250 μs ). These characteristics render existing energy saving idle CPU sleep states ineffective due to the performance overhead caused by the state’s transition latency. Besides the state transition latency, another important contributor to the performance overhead of sleep states is the cold-start latency, or in other words, the time required to warm-up microarchitectural state (e.g., cache contents, branch predictor metadata) that is flushed or discarded when transitioning to a lower-power state. Both the transition latency and cold-start latency can be particularly detrimental to the performance of latency critical applications with short execution times. While prior work focuses on mitigating the effects of transition and cold-start latency by optimizing request scheduling, in this work, we propose a redesign of the Core C-state architecture for latency-critical applications. In particular, we introduce C6Awarm a new Agile Core C-state that drastically reduces the performance overhead caused by idle sleep state transition latency and cold-start latency, while maintaining significant energy savings. C6Awarm achieves its goals by implementing 1) medium-grained power gating, 2) preserving the microarchitectural state of the core and 3) by keeping the clock generator and PLL active and locked. Our analysis for a set of microservices based on an Intel Skylake server, shows that C6Awarm manages to reduce the energy consumption by up to \(70\% \) with limited performance degradation (at-most \(2\% \) ).

Magnetic hollow fibers of covalent organic frameworks (COF) for pollutant degradation and adsorptive removal

The shaping of covalent organic frameworks (COFs) from non-processible powder forms into applicable architectures with additional functionality remains a challenge. Using pre-electrospun polymer fibers as a sacrificial template, herein, we report a green synthesis of an architecture in the form of COF hollow fibers with an inner layer of peroxidase-like iron oxide nanoparticles as a catalytic material. When compared to peroxidase-like pristine iron oxide nanoparticles, these COF hollow fibers demonstrate higher catalytic breakdown of crystal violet due to their peroxidase-like activity via advanced oxidation process. Furthermore, as a potential adsorbent, hollow COF fibers exhibit significantly effective adsorption capacity and removal efficiency of organic solvent and oil from water. Because of their magnetic nature, COF hollow fibers can be easily recovered and have exhibited high recycling stability for both catalytic dye degradation and organic solvent removal from water.

I-Health: SDN-Based Fog Architecture for IIoT Applications in Healthcare.

The Industrial Internet of Things (IIoT) has been introduced in an era of increasingly broad potentials in the medical industry. In recent years, IIoT-based healthcare applications have grown in popularity, with the majority of them relying on Wireless Body Area Network (WBAN) for flexibility. There have been a few recent works that have investigated SDN-based fog architecture for constructing smart healthcare systems. However, the best fog node from the fog layer must be identified and limit the transmission of unnecessary data. To address this issue, the Intelligent Software-defined Fog Architecture (i-Health) is developed in this work. Based on the prior data pattern of each patient, the controller will decide whether to send the data to the fog layer. Furthermore, we introduced the Fog Ranking Service (FRS) and Fog Probing Service (FPS) to select the best fog node. The performance comparison reveals that the proposed i-Health outperforms existing benchmark approaches.

Sandwiched WS2/MoTe2/WS2 Heterostructure with a Completely Depleted Interlayer for a Photodetector with Outstanding Detectivity.

Photodetectors based on two-dimensional van der Waals (2D vdW) heterostructures with high detectivity and rapid response have emerged as promising candidates for next-generation imaging applications. However, the practical application of currently studied 2D vdW heterostructures faces challenges related to insufficient light absorption and inadequate separation of photocarriers. To address these challenges, we present a sandwiched WS2/MoTe2/WS2 heterostructure with a completely depleted interlayer, integrated on a mirror electrode, for a highly efficient photodetector. This well-designed structure enhances light-matter interactions while facilitating effective separation and rapid collection of photocarriers. The resulting photodetector exhibits a broadband photoresponse spanning from deep ultraviolet to near-infrared wavelengths. When operated in self-powered mode, the device demonstrates an exceptional response speed of 22/34 μs, along with an impressive detectivity of 8.27 × 1010 Jones under 635 nm illumination. Additionally, by applying a bias voltage of -1 V, the detectivity can be further increased to 1.49 × 1012 Jones, while still maintaining a rapid response speed of 180/190 μs. Leveraging these outstanding performance metrics, high-resolution visible-near-infrared light imaging has been successfully demonstrated using this device. Our findings provide valuable insights into the optimization of device architecture for diverse photoelectric applications.

Real-Time Monitoring of Cable Sag and Overhead Power Line Parameters Based on a Distributed Sensor Network and Implementation in a Web Server and IoT.

Based on the need for real-time sag monitoring of Overhead Power Lines (OPL) for electricity transmission, this article presents the implementation of a hardware and software system for online monitoring of OPL cables. The mathematical model based on differential equations and the methods of algorithmic calculation of OPL cable sag are presented. Considering that, based on the mathematical model presented, the calculation of cable sag can be done in different ways depending on the sensors used, and the presented application uses a variety of sensors. Therefore, a direct calculation is made using one of the different methods. Subsequently, the verification relations are highlighted directly, and in return, the calculation by the alternative method, which uses another group of sensors, generates both a verification of the calculation and the functionality of the sensors, thus obtaining a defect observer of the sensors. The hardware architecture of the OPL cable online monitoring application is presented, together with the main characteristics of the sensors and communication equipment used. The configurations required to transmit data using the ModBUS and ZigBee protocols are also presented. The main software modules of the OPL cable condition monitoring application are described, which ensure the monitoring of the main parameters of the power line and the visualisation of the results both on the electricity provider's intranet using a web server and MySQL database, and on the Internet using an Internet of Things (IoT) server. This categorisation of the data visualisation mode is done in such a way as to ensure a high level of cyber security. Also, the global accuracy of the entire OPL cable sag calculus system is estimated at 0.1%. Starting from the mathematical model of the OPL cable sag calculation, it goes through the stages of creating such a monitoring system, from the numerical simulations carried out using Matlab to the real-time implementation of this monitoring application using Laboratory Virtual Instrument Engineering Workbench (LabVIEW).

Research on the Application of Cloud Edge Collaboration Architecture in Power System

Research on the Application of Cloud Edge Collaboration Architecture in Power System

Identifying Reference Architecture Types for Stakeholder Groups in Industry 4.0

AbstractNew developments in the area of the Industrial Internet‐of‐Things (IIoT) and Industry 4.0 offer huge potential for a more efficient and flexible industrial production, but are also accompanied by rising system complexity. Consequently, to deal with the increased system complexity, novel approaches, such as reference architectures, are emerging. However, most of these approaches are not yet mature and rather theoretical than ready‐to‐use. Thus, companies need to be provided with frameworks that actively support the transformation of their systems towards Industry 4.0. One of those frameworks has been introduced with Reference Architecture Model Industry 4.0 (RAMI 4.0), which counteracts the mentioned complexity and can be used for various use cases. However, as most of its concepts are too general to be applied directly to actual systems, the need for directly applicable reference architectures emerges. Therefore, this paper proposes a method to derive more detailed reference architectures based on RAMI 4.0 by making use of model‐based systems engineering (MBSE), which target single manufacturing domains rather than the whole industry. Therefore, relevant stakeholders are analyzed and different types of reference architectures targeting their concerns are identified. The resulting reference architectures should be ready‐to‐use for interested manufacturers and thus, enhance the acceptance of RAMI 4.0 as well as improve systems engineering in industrial manufacturing. Finally, the developed reference architecture is evaluated in a proof‐of‐concept case study of a flexible production system.

Dynamic Schwarz Meta-Foams: Customizable Solutions for Environmental Noise Reduction.

In an era marked by increasing environmental challenges affecting human well-being, traditional acoustic materials struggle to effectively handle the diverse and multi-frequency nature of harmful environmental noises. This has spurred a demand for innovative acoustic metamaterial solutions by utilizing sustainable design strategies. This research introduces tunable Schwarz metamaterial capable of transforming into a soft meta-foam to solve the complex problems of varying environmental noises. This study primarily focuses on adjusting single to multiple sound-blocking bandgaps mechanism using a multi-layered approach, incorporating the Schwarz P-type triply periodic minimal surface (TPMS) and its elective soft foam counterpart, known as tunable Schwarz meta-foams (TSMF-x). The tunable design parameters of the unit cell, multi-layered TPMS, and soft programmable TSMF-lichen version are comprehensively explored including a fire-safety test. The results demonstrate these enhanced flame retardant meta-foam families have the potential to be used for mid-to-high-frequency environmental noises in industrial equipment and smart homes for sustainable architecture and environmental health applications.

Natural fibre pultruded profiles: Illustration of optimisation processes to develop high-performance biocomposites for architectural and structural applications

The selection of materials in the construction industry plays a pivotal role in advancing sustainability goals. Traditional materials derived from natural resources face inherent constraints linked to geographic limitation, growth time, and geometric inconsistency and therefore recent attention has shifted towards developing novel bio-based materials. Composites, offering varying properties and geometries, are becoming increasingly popular for customising materials for specific applications. Pultrusion, a technology for manufacturing linear fibre-reinforced composites, is a well-established and reliable method. This study delves into optimising pultrusion technology, which traditionally relies on synthetic fibres, by exploring the potential of natural alternatives, specifically hemp bast fibres. Additionally, it presents a customised formulation based on a plant-based resin and additives. This formulation is tailored for pultrusion to produce high-performance biocomposites for use as load-bearing components in structural applications, with an initial focus on bending structures. The study elaborates on the material composition and performance of these newly developed natural fibre pultruded profiles, showcasing their mechanical capabilities through rigorous experimentation and testing. The results demonstrate the material's mechanical capabilities showcasing a flexural strength of 260 MPa with a bending modulus of 21 GPa and a bending radius reaching 0.5 m. While this study focuses on the material formulation tested on laboratory-scale pultrusion, the findings will be later applied in an upscaled production at an industrial level, aiming to enhance overall sustainability in the construction industry.

LeichtPRO-profiles: development and validation of novel linear biocomposite structural components fabricated from pultruded natural flax fibres with plant-based resin for sustainable architectural applications

Reconsidering the materials used in construction is crucial within the building industry, particularly in the context of sustainability. Recently, there has been a growing interest in exploring novel materials, with fibre-reinforced composites emerging as a prominent choice with biocomposites standing out as promising for advancing sustainability goals. This paper introduces the development of LeichtPRO-Profiles, continuous linear biocomposite profiles fabricated using the pultrusion technology. A primary focus is the application of these profiles in structural systems as load-bearing elements, emphasising the significance of understanding their mechanical properties. Specifically, an original application involves active-bending structures, necessitating a focus on the material’s bending behaviour. This study discusses the methods employed in developing the pultruded biocomposite profiles which are made from natural flax fibres and an optimised matrix formulation based on a plant-based resin system. This research also outlines the optimisation of the fabrication process of these biocomposite profiles using bio-based ingredients. The results demonstrate the material’s mechanical capabilities through extensive experiments and mechanical tests, revealing a compression strength of 31.2 kN and a flexural strength of 300 MPa, with a bending radius of up to 2.4 m, indicating its suitability for structural applications. Concepts of applications in several systems across different scales and contexts are also presented. The versatility and adaptability of this product make it suitable for a wide range of applications spanning various scales and thematic contexts.

Cement-Based Composites with Additive Manufacturing Technologies for Sustainable Architecture: A Review

To minimize environmental damage caused by architecture and contribute to sustainability, digital fabricationtechnologies, which have lower CO2 emissions and energy consumption compared to traditional productiontechniques, have been applied. One of the processes of digital fabrication is additive manufacturing (AM). AMtechnology is becoming increasingly popular due to its ability to give the designer freedom of form and its eco-friendlynature. Considering the cement-based composites applied with AM technologies, CO2 emissions can be reduced byapproximately 32%. This paper aims to systematically summarize the content, improved properties, andenvironmental advantages of cement-based composite materials applied with AM technology and to determine thedistribution of the number of studies by year. It has been observed that the studies have started to increasesignificantly since 2017. This indicates that cement used in architectural applications through various reinforcementsand, AM technology for environmental sustainability has become more and more widespread.