General Architecture Research Articles

Maintenance architecture design of equipment operation and maintenance system based on digital twins

The digital twins process of equipment operation and maintenance involves massive data and complex business logic. Through data awareness of equipment and workshop environment. According to the characteristics of equipment operation and maintenance, the data characteristics and requirements are analysed, an equipment operation and maintenance management system based on the micro-service framework is proposed, a set of equipment operation and maintenance digital twins are designed and a general reference software architecture is designed and by checking the underlying in-heap pattern reduces high-impact portions of memory to provide accurate follow-up complex digital twins modelling provides the architectural foundation.

Geology-constrained dynamic graph convolutional networks for seismic facies classification

Knowing a land’s facies type before drilling is an essential step in oil exploration. In seismic surveying, subsurface images are analyzed to segment and classify the facies in that volume. With the recent developments in deep learning, multiple works have utilized deep neural networks to classify facies from subsurface images. Unlike natural images, seismic data have different patterns and structures, which means that although general deep learning architectures can work with seismic data, it would be more effective if these architectures were optimized and refined specifically for such types of data. Most of the works in the seismic domain focus on convolution neural networks as the main backbone for the architectures, and more recently transformers started becoming more common in seismic data processing. Proposing a different approach that can capture unique correlations in the data, we introduce the use of dynamic graph convolutional networks as a method for capturing long-term dependencies for seismic facies classification. The proposed architecture combines the use of convolution neural networks and graph convolution networks to capture both global and local structures of the data. The performance of the model was evaluated on a facies classification dataset, and the proposed method provided state-of-the-art results while significantly reducing the number of parameters in the model compared to other architectures. Code is available at https://github.com/swaidan/Geology-Restricted-DGCNN.

Robust and Secure Federated Learning Against Hybrid Attacks: A Generic Architecture

Robust and Secure Federated Learning Against Hybrid Attacks: A Generic Architecture

GLPanoDepth: Global-to-Local Panoramic Depth Estimation.

Depth estimation is a fundamental task in many vision applications. With the popularity of omnidirectional cameras, it becomes a new trend to tackle this problem in the spherical space. In this paper, we propose a learning-based method for predicting dense depth values of a scene from a monocular omnidirectional image. An omnidirectional image has a full field-of-view, providing much more complete descriptions of the scene than perspective images. However, fully-convolutional networks that most current solutions rely on fail to capture rich global contexts from the panorama. To address this issue and also the distortion of equirectangular projection in the panorama, we propose Cubemap Vision Transformers (CViT), a new transformer-based architecture that can model long-range dependencies and extract distortion-free global features from the panorama. We show that cubemap vision transformers have a global receptive field at every stage and can provide globally coherent predictions for spherical signals. As a general architecture, it removes any restriction that has been imposed on the panorama in many other monocular panoramic depth estimation methods. To preserve important local features, we further design a convolution-based branch in our pipeline (dubbed GLPanoDepth) and fuse global features from cubemap vision transformers at multiple scales. This global-to-local strategy allows us to fully exploit useful global and local features in the panorama, achieving state-of-the-art performance in panoramic depth estimation.

PartialHD: Toward Efficient Hyperdimensional Computing by Partial Processing

Hyperdimensional (HD) computing is a brain inspired learning method that is widely employed in resource-constrained applications such as the internet of things (IoT) regarding its lightweight computation. Although HD computing has high efficiency in the IoT applications, it suffers from high computation due to the large vector size. Thus, several studies are proposed to speedup and increase the efficiency of HD computing. In this work, we propose a method to process HD computing called PartialHD. Our method divides a long hypervector into multiple partial vectors, and process each vector separately. In the retraining phase, this method improves accuracy up to 1.93%. Moreover, by employing our proposed method in the retraining and inference, only a few partial vectors participate and this causes the computational overhead reduces in these phases. The evaluation shows that our method processes on average 22.2% and 36.23% of the entire hypervectors with negligible accuracy loss in inference and retraining, respectively. Furthermore, we propose two general architectures (light-weight and high-speed), which accelerate partial vector computations on different FPGA platforms. The result shows that the light-weight architecture can accelerate HD computing on resource-constrained FPGAs such as Artix while the high-speed architecture has 4.72× higher throughput than light-weight architecture.

Frequency Analysis of Pulse-Width Modulation on All-Digital Transmitters

This work investigates the frequency spectrum in the modulator and up-conversion outputs of a generic all-digital transmitter architecture. Considering the modulator is implemented as a pulse-width modulation (PWM), it is verified that this up-conversion process introduces a series of impulses resembling the spectrum of a unipolar return-to-zero (RZ) waveform in addition to the PWM impulses. Simulations show the minimum bit resolution of PWM signals where the impulses introduced are almost unnoticed, thus their influence is not distinguished from random background noise. An all-digital transmitter is designed on a field-programmable gate array to verify it experimentally. Experimental results show that reducing power from PWM harmonics leads to a power increase in the RZ harmonics, and vice-versa. These harmonic’s power present an energy trade-off relation, a novelty demonstrated in this paper. Finally, but not least, it is observed that filtering the carrier wave out causes a considerable amount of power to be lost.

A contribution for dismantling of nuclear facilities: a functional pattern for dismantling operations and indicators design and management

Optimizing nuclear installation decommissioning and dismantling operations is an ongoing quest. Faced with the complexity of this activity, Model Based System Engineering promotes relevant principles and modeling techniques. It motivated the definition of a functional generic pattern model of the waste package production line and the decommissioning of the facility. It proposes a global and generic functional architecture of such a system aiming to reduce the level of the pollutant. This pattern is coupled with a process of logistics. Six functions are combined to define this functional pattern. The application of this pattern model to a case of waste recovery in a pit shows the relevance of the model-based system engineering approach, reducing the weight of the history in the development of scenarios by optimizing the control means for nuclear safety and product quality.

Comparative Performance Evaluation of One-Stage and Two-Stage Object Detectors for Screw Head Detection and Classification in Disassembly Processes

The process of manually detecting and removing screws during disassembly can be time-consuming and demanding. Automated systems utilizing imaging and deep learning can rapidly and accurately identify screws based on features like head shape and size. Different deep learning methods have varying screw detection capabilities. Choosing the right method significantly impacts system effectiveness, robustness and efficiency. Several generic deep learning architectures have been proposed to address this problem. However, there is little guidance on their suitability for specific scenarios. This paper aims to provide qualified guidance for selecting suitable deep learning methods for screw head detection. The proposed approach involves researching existing object detection methods and choosing two representatives – one employing a one-stage approach, the other using two-stage. These selected methods will be implemented identically and their performance evaluated via rigorous analysis and comparison. The dataset comprises synthetic and real images of 6 screw types across 3 products. YOLOv5 was selected for one-stage detectors, Faster R-CNN for two-stage. Results show the one-stage YOLOv5 generally outperforms Faster R-CNN in precision, recall, speed and training times, while Faster R-CNN delivered superior recall with smaller training datasets.

OSNet and MNetO: Two Types of General Reconstruction Architectures to Transform DBP Images for Linear Computed Tomography in Multi-Scenarios

OSNet and MNetO: Two Types of General Reconstruction Architectures to Transform DBP Images for Linear Computed Tomography in Multi-Scenarios

Advancing 6-DoF Instrument Pose Estimation in Variable X-Ray Imaging Geometries.

Accurate 6-DoF pose estimation of surgical instruments during minimally invasive surgeries can substantially improve treatment strategies and eventual surgical outcome. Existing deep learning methods have achieved accurate results, but they require custom approaches for each object and laborious setup and training environments often stretching to extensive simulations, whilst lacking real-time computation. We propose a general-purpose approach of data acquisition for 6-DoF pose estimation tasks in X-ray systems, a novel and general purpose YOLOv5-6D pose architecture for accurate and fast object pose estimation and a complete method for surgical screw pose estimation under acquisition geometry consideration from a monocular cone-beam X-ray image. The proposed YOLOv5-6D pose model achieves competitive results on public benchmarks whilst being considerably faster at 42 FPS on GPU. In addition, the method generalizes across varying X-ray acquisition geometry and semantic image complexity to enable accurate pose estimation over different domains. Finally, the proposed approach is tested for bone-screw pose estimation for computer-aided guidance during spine surgeries. The model achieves a 92.41% by the 0.1·d ADD-S metric, demonstrating a promising approach for enhancing surgical precision and patient outcomes. The code for YOLOv5-6D is publicly available at https://github.com/cviviers/YOLOv5-6D-Pose.

Electronic isomerism in a heterometallic nickel-iron-sulfur cluster models substrate binding and cyanide inhibition of carbon monoxide dehydrogenase.

The nickel-iron carbon monoxide dehydrogenase (CODH) enzyme uses a heterometallic nickel-iron-sulfur ([NiFe4S4]) cluster to catalyze the reversible interconversion of carbon dioxide (CO2) and carbon monoxide (CO). These reactions are essential for maintaining the global carbon cycle and offer a route towards sustainable greenhouse gas conversion but have not been successfully replicated in synthetic models, in part due to a poor understanding of the natural system. Though the general protein architecture of CODH is known, the electronic structure of the active site is not well-understood, and the mechanism of catalysis remains unresolved. To better understand the CODH enzyme, we have developed a protein-based model containing a heterometallic [NiFe3S4] cluster in the Pyrococcus furiosus (Pf) ferredoxin (Fd). This model binds small molecules such as carbon monoxide and cyanide, analogous to CODH. Multiple redox- and ligand-bound states of [NiFe3S4] Fd (NiFd) have been investigated using a suite of spectroscopic techniques, including resonance Raman, Ni and Fe K-edge X-ray absorption spectroscopy, and electron paramagnetic resonance, to resolve charge and spin delocalization across the cluster, site-specific electron density, and ligand activation. The facile movement of charge through the cluster highlights the fluidity of electron density within iron-sulfur clusters and suggests an electronic basis by which CN- inhibits the native system while the CO-bound state continues to elude isolation in CODH. The detailed characterization of isolable states that are accessible in our CODH model system provides valuable insight into unresolved enzymatic intermediates and offers design principles towards developing functional mimics of CODH.

DCG-GAN: design concept generation with generative adversarial networks

Abstract Generative adversarial networks (GANs) have recently been proposed as a potentially disruptive approach to generative design due to their remarkable ability to generate visually appealing and realistic samples. Yet, we show that the current generator-discriminator architecture inherently limits the ability of GANs as a design concept generation (DCG) tool. Specifically, we conduct a DCG study on a large-scale dataset based on a GAN architecture to advance the understanding of the performance of these generative models in generating novel and diverse samples. Our findings, derived from a series of comprehensive and objective assessments, reveal that while the traditional GAN architecture can generate realistic samples, the generated and style-mixed samples closely resemble the training dataset, exhibiting significantly low creativity. We propose a new generic architecture for DCG with GANs (DCG-GAN) that enables GAN-based generative processes to be guided by geometric conditions and criteria such as novelty, diversity and desirability. We validate the performance of the DCG-GAN model through a rigorous quantitative assessment procedure and an extensive qualitative assessment involving 89 participants. We conclude by providing several future research directions and insights for the engineering design community to realize the untapped potential of GANs for DCG.

A General Architecture for Intersection-Based Decentralized Supervisory Control of Discrete Event Systems

A General Architecture for Intersection-Based Decentralized Supervisory Control of Discrete Event Systems

Tech-Business Analytics in Tertiary Industry Sector

Purpose: Tech-business analytics play a crucial role in the tertiary industry sector by enabling data-driven decision-making and providing analytical insights to enhance customer experiences, improve operational performance, and drive corporate growth. This sector, which includes industries such as healthcare, finance, education, and professional services, benefits from tech-business analytics through the analysis of vast data sets like patient records, financial transactions, student performance, and customer interactions. By identifying patterns, trends, and insights within these data sets, businesses can make more informed decisions. Design/Methodology/Approach: The tertiary industry sector employs an organized, iterative approach to tech-business analytics, which involves defining business issues, collecting and cleansing data, analysing this data, interpreting the results, and communicating these findings to stakeholders. Through this process, businesses are empowered to make informed decisions that support their growth and development. Findings/Result: The report talks about how Tech Business Analytics in the Tertiary Industry Sector will have controlled the expansion itself from its evolution to this point. Originality/Value: A description of how business analytics varies from tech business analytics in the tertiary industry sector. For technical purposes, there is also a generic architecture that examines 30 recently published Tech Business Analytics in Tertiary Industry sector research projects. Paper Type: Exploratory Analysis.

A Histological and Histochemical study of small intestine in adult stage of cats.

This study was performed to detect histological & histochemical facts of the bowel in adult cats.was approved on six mature cats the body masses it between 3 to 3.5 Kg, which occur in College of Veterinary Medicine, University of Baghdad. Killing euthanized by injection of an overdose of sodium pentobarbital intra-cardiac. Tissue sections were treated, sectioned and exposed to specific dyes to detected histological details.small intestine of adult cats is lined by simple columnar epithelium with goblet cells. The general architecture was like to other mammals with little individualities: the shape of the villi different from leaf-similar in the duodenum ,pointed in jejunum &leaf-formed in the ileum due to the type nutrition, the Brunner’s gland was current in the initial portion and caudal part duodenum and initial part of the jejunum act to neutralize the acidity of the food coming from the stomach with foods.The paneth cells have few numbers with rather epithelial cells in duodenum. Peyer's spot was current at the termination of jejunum and ileum. Goblet cells were able to secrete a high amount of slimy mucous into the lumen of the bowel ,.Carbohydrate profile small intestine shown the found of non-sulfated acidic mucin in the crypts and intestinal glands while the a neutral & sulfated acidic mucins found in the villi & superficial epithelium.

Facilitating Model Training With Automated Techniques

Automating artificial intelligence (AI) model training has emerged as a significant challenge in the field of automation. The complete pipeline from raw data to model deployment poses the need to define robust processes that ensure the efficiency of the services that expose the models. This paper introduces a generic architecture for automating data preparation, training of models, selection of models, and deployment of models as web services for third-party consumption using Microsoft Azure Machine Learning’s (AzureML) CI/CD tools. We conducted a practical experiment utilizing AzureML pipelines with predefined and custom modules, demonstrating its readiness for integration into any production application. We also successfully integrated this architecture into a real-world product designed for industrial forecasting. This practical implementation demonstrates the effectiveness and adaptability of our approach, indicating its potential to address diverse training needs. Keywords and phrases: Artificial Intelligence, Automation, Optimization.

Effects of Expansionary Fiscal Policy in a Commodity-Exporting Economy: Evidence from Mongolia

Abstract This study contributes to the ongoing debate on the consequences of expansionary fiscal policy by evaluating the macroeconomic effects of various fiscal policy options in a small open economy using a dynamic stochastic general equilibrium (DSGE) model. In addition, the study emphasizes the importance of studying Mongolia, which has unique characteristics and exhibits significant research gaps regarding its fiscal policy. The general architecture of the selected DSGE model includes different types of firms and households, commodity sectors, natural resource funds, and abundant fiscal tools regarding both expenditure and revenue. Employing numerous types of fiscal policy shocks, this study reveals that an exogenous increase in government investment yields the most significant long-term economic benefits, boosting potential output by 0.3%. Among the policy options, government transfers are the least effective in promoting economic output, and existing transfer policies in Mongolia appear to exert only a modest impact on growth, instead primarily contributing to the redistribution of resources. Finally, the estimated output multipliers (except transfers) are greater than one, implying that fiscal policy instruments may be an effective tool for managing the economy in Mongolia. JEL classification numbers: E17, E62, H24, H54, H55. Keywords: Fiscal policy, Fiscal policy multiplier, Small open economy, Dynamic Stochastic General Equilibrium model, Natural Resource Sector.

A comprehensive application architecture for unlocking the potential of student feedback

This article proposes a novel solution aimed at streamlining the student feedback, enabling them to assess various aspects and activities within their institution. Different from the classical solution of collecting students feedback using a set of questions grouped in one or many forms, we designed a solution capable to empower students of all age groups to enhance their higher education experiences by openly expressing opinions about their enrolled faculties, providing valuable guidance to prospective college students. Our proposal introduces a generic application architecture designed to integrate the presentation of the main opportunities and offers from many universities with the student reviews on all aspects of these universities. Thus, we avoid a number of factors that reduce the quantity and objectivity of student feedback. Important elements in our architecture refer to reviewing and rating system as well as the search system which facilitate the use of the existing evaluations on different aspects of an university. We devised a generic architecture for a client-server application, specifying a minimum set of required functionalities. For validation, we designed and implemented a client-server application based on the generic architecture we outlined. The application offers several key advantages, including robust information security measures, and diverse communication channels for authenticated users within the platform. The application’s architecture includes an encryption system based on multiple encryption techniques. Additionally, an algorithm for search optimization within the platform have been developed, further enhancing its utility.

Efficient IoT Traffic Inference: From Multi-view Classification to Progressive Monitoring

Machine learning-based techniques have proven to be effective in Internet-of-Things (IoT) network behavioral inference. Existing works developed data-driven models based on features from network packets and/or flows, but mainly in a static and ad-hoc manner, without adequately quantifying their gains versus costs. In this article, we develop a generic architecture that comprises two distinct inference modules in tandem, which begins with IoT network behavior classification followed by continuous monitoring. In contrast to prior relevant works, our generic architecture flexibly accounts for various traffic features, modeling algorithms, and inference strategies. We argue quantitative metrics are required to systematically compare and efficiently select various traffic features for IoT traffic inference. This article 1 makes three contributions: (1) For IoT behavior classification, we identify four metrics, namely, cost, accuracy, availability, and frequency, that allow us to characterize and quantify the efficacy of seven sets of packet-based and flow-based traffic features, each resulting in a specialized model. By experimenting with traffic traces of 25 IoT devices collected from our testbed, we demonstrate that specialized-view models can be superior to a single combined-view model trained on a plurality of features by accuracy and cost. We also develop an optimization problem that selects the best set of specialized models for a multi-view classification. (2) For monitoring the expected IoT behaviors, we develop a progressive system consisting of one-class clustering models (per IoT class) at three levels of granularity. We develop an outlier detection technique on top of the convex hull algorithm to form custom-shape boundaries for the one-class models. We show how progression helps with computing costs and the explainability of detecting anomalies. (3) We evaluate the efficacy of our optimally selected classifiers versus the superset of specialized classifiers by applying them to our IoT traffic traces. We demonstrate how the optimal set can reduce the processing cost by a factor of six with insignificant impacts on the classification accuracy. Also, we apply our monitoring models to a public IoT dataset of benign and attack traces and show they yield an average true-positive rate of 94% and a false-positive rate of 5%. Finally, we publicly release our data (training and testing instances of classification and monitoring tasks) and code for convex hull-based one-class models.

Architecture as fluid technology. The housing blocks by Corporación de la Vivienda of Chile

ABSTRACT In a general sense, architecture and technology articulate two modes of evaluation of objects and built environments; however, are there conditions that allow us to consider architectural works as technologies? This paper explores this possibility by analyzing bibliographic, documentary, and planimetric information on housing blocks developed by the state agency, Corporación de la Vivienda (CORVI), between 1953 and 1974. Using Process Tracing as a method of analysis, milestones were identified that showed the ways in which the design and production of this typology were influenced by a regime of values based on the idea of rationalization, and that the heterogeneity of its cases showed flexibility and adaptation to the ways of calculating and realizing this value. The conclusions address the causality observed between value and typology in this case, and suggest that this behavior expresses a technological condition of architecture.