Transformation Of Units Research Articles

Crime Type Identification Using High-Order Deep Residual Network with Multiple Attention Algorithm

ABSTRACT Crime type identification is crucial for improving public safety through more accurate prevention and efficient responses. However, practical applications often suffer from a significant lack of effective samples features, making it difficult to focus on the most informative aspects during identification. This study addresses these challenges by proposing a novel crime type identification method that leverages a deep neural network enhanced with multiple attention mechanisms. The approach includes a tailored data processing method involving target encoding to convert categorical data into numerical form, L 2 normalizer to standardize data and ensure balanced feature contribution, and variance threshold feature selection to remove low-variance features. Additionally, a High-Order Deep Residual Network with Multiple Attention (HO-ResNet-MA) is developed, featuring an optimized Huta68 block (Huta-6(8)-MA ResBlock) with an enhanced Contextual Transformer (CoT) unit for local attention and a queue-and-exclusion layer for global attention. To validate the effectiveness of the proposed method, homicide reports data and Chicago crimes data are processed and fed into the crime type identification model, resulting in accuracies of over 84.1% and 99.5%, respectively. This study makes contributions to the field of crime analysis by validating the practical applicability of these approaches, and enhancing the efficiency of public safety workers.

A Transformer-based Semantic-Aware U-Net Model for Improving in Endometrial Cancer Detection

The rise in Endometrial Cancer (EC) necessitates early diagnosis using medical imaging. Deep learning models have been developed to address misdiagnosis, but their accuracy is limited by small-scale datasets. To address this issue, a new Deep EC Prediction (DeepECP) model, combining Feature Entanglement Generative Adversarial Network (FE-GAN) and multimodal Convolutional Neural Network with Long Short-Term Memory (CNN-LSTM), has been designed to generate synthetic MRI images and detect EC stages. However, precise identification and categorization of various EC stages require the segmentation of MRI scans. While the U-Net model is frequently used for this task, it has constraints in capturing long-range relationships. Other models, like Transformers with self-attention mechanisms may not have strong low-level localization capabilities. Hence, this paper introduces a Transformer-based Semantic-Aware U-Net (TSA-UNet) model with DeepECP for segmenting and classifying EC stages in MRI scans. The TSA-UNet model combines Transformer and U-Net structures, consisting of encoder, fusion, and decoder units. The encoder is a CNN-Transformer that captures multilevel bitemporal features and encodes tokenized sequences to get high-level features. The Transformer unit uses Semantic Guidance Attention (SGA) to enhance feature discrimination and reduce semantic gaps. An Adaptive Cross-Fusion (ACF) unit merges bitemporal features, and a Cascaded Upsampling Decoder (CUD) combines low-level CNN and high-level Transformer characteristics to localize and segment EC stages in MRI scans. The segmented images are then inputted to the DeepECP model for classification. Finally, the experimental outcomes reveal that the TSA-UNet-DeepECP attains higher efficiency than conventional models for segmentation and classification of EC from MRI images.

Power electronics for green hydrogen generation with focus on methods, topologies, and comparative analysis.

This research article meticulously examines advanced power electronic converters crucial for optimizing electrolyzer perfor- mance in hydrogen production systems. It conducts a thorough review of mature electrolyzer types, detailing their specifications, electric models, manufacturers, and scalability. To meet the high current and stable DC voltage demands of industrial electrolyzers, the study delves into a broad spectrum of AC-DC and DC-DC converter topologies. It explores cutting-edge solutions like 12-pulse and 20-pulse rectifiers, advanced higher multi-pulse rectifiers utilizing conventional and auto-connected transformer units, Multi-Step Auto-connected Transformers (MSAT), polygon autotransformers, and auxiliary filters and circuits such as pulse multiplication circuits and active power filters. These innovations significantly reduce harmonic distortions and enhance power quality, addressing challenges inherent in conventional 6-pulse diode bridge rectifiers. The research also focuses on the efficiency and power factor correction capabilities of Active Front End (AFE) converters and the 3L-DNPC rectifier. Additionally, it investigates various DC-DC converters, including the Continuous Input Current Non-Isolated Bidirec- tional Interleaved Buck-Boost DC-DC Converter, the Interleaved Buck Converter (IBC) with extended duty cycles, the 3-level buck-boost converter with a coupled inductor, Quadratic converters designed for fault tolerance, the three-level interleaved buck converter, among others. Converter designs like the galvanically isolated half-bridge converter with controlled reverse-blocking switches for achieving zero voltage switching (ZVS), the Push-Pull isolated DC-DC converter prioritizing current stability, and the Isolated Full-Bridge Boost Converter (IFBBC) adept at handling fluctuating power outputs from renewable sources are also explored. Moreover, the study scrutinizes a range of converter configurations such as the Two-stage ZVT boost converter with LCL-Type SRC, multiphase interleaved DC-DC stage, and three-port isolated DC/DC converter for efficient integration with multiple energy sources concurrently. Discussing feature trends in power-to-hydrogen systems, the research reviews multi-cell modular rectifiers and multi-stage rectifiers. Overall, the study underscores the critical role of advanced converter topologies in enhancing efficiency, reliability, and power quality in electrolyzer systems, thereby contributing significantly to the progression of sustainable energy technologies.

Demonstration of an In-Flight Entertainment System Using Power-over-Fiber

The use of optical fibers is increasing in modern aircraft because this helps solve challenges of size, weight, communication, and reliability in new generation aircraft. This study describes a video and power transmission system using optical fibers (PoF) for in-flight entertainment (IFE) system application. We present the benefits and the limitations of this application, and we perform two practical experiments to demonstrate their performance. We used off-the-shelf devices in the experiments, such as one 15-Watt semiconductor laser operating at 808 nm, GaAs photovoltaic converters, optical transmitters and receivers, and video monitors. The power and video signals were transmitted using two 50-m length multimode fibers. In addition, we proposed and tested two types of energy transformation units (ETUs), which are responsible for supplying electrical energy to the IFE video monitor and the optical fiber receiver.

Applying a novel generalized model of sedimentation to the design of settlers and thickeners: Integration and extension of the classical procedures

A new generalized procedure previously introduced and justified by the authors, based on the simultaneous integration of the macroscopic and microscopic differential mass balances by introducing the concept of separation units as an extension of the transfer unit concept was applied here to sedimentation. This similarity leads to a broader analogy with basic transfer and transformation unit operations, at this step focused on the formulation and contrast of a generalized form of the so-called design or sizing equation. This model is comparatively applied under restricted flow Conditions-Vertical Flow Settlers (CSTS) - with the classical methods and extended to the design of horizontal flow settlers (PFS). That opens the door to the sizing of settlers and thickeners in the presence of deviations from these ideal flow models and with other operating configurations. Such a contribution to a desirable generalization and homologation is according to Chemical Engineering methodology specifically in the frequently empirical or casuistic treatment of unit operations with particulate solids.

An efficient deep learning based scheme for adaptive auto-reclosing in power transmission lines

As a large percentage of faults in transmission lines are single-line-to-ground (SLG) and transient in nature, the use of adaptive single-phase auto-reclosing (ASPAR) schemes is essential. In this paper, a new scheme for ASPAR is proposed to improve the stability and reliability of power transmission lines. The proposed approach distinguishes between the transient and permanent faults by using the feature of Stationary Wavelet Transform (SWT) and Gate Recurrent Unit (GRU) deep artificial neural network. Also, by utilizing another GRU network, the proposed scheme predicts the extinction time of the secondary arc (SA) one power cycle before the complete extinction. This can increase the speed of the faulty phase reclosing in the case of a transient fault. Simulation results carried out on a 400 kV power transmission line in the EMTP-RV and MATLAB software environments illustrate that the F1 score of the proposed ASPAR in classifying the faults is 99.55 %, and its average error in predicting the extinction time of the SA based on the root mean square error (RMSE) criterion is 0.0557. Also, the superiority and high-quality performance of the proposed protection scheme are demonstrated by comparing the results with several state-of-the-art methods.

Structural pathway for nucleation and growth of topologically close-packed phase from parent hexagonal crystal

The solid diffusive phase transformation involving the nucleation and growth of new phase particles is universal and frequently employed but has not yet been fully understood at the atomic level. Here, our first-principles calculations reveal a structural formation pathway of a series of topologically close-packed (TCP) phases within the hexagonally close-packed (hcp) matrix. The results show that the nucleation follows a nonclassical nucleation process, and apart from atomic diffusion, the entire hcp→TCP transformation only involves shuffle-based displacements, with a specific 3-layer unstable hcp-ordering serving as the basic structural transformation unit (BSTU). The thickening of plate-like TCP phases relies on the formation of these unstable hcp-orderings at their coherent TCP/matrix interface to nucleate a ledge, but the ledge lacks the dislocation characteristics which is commonly considered in the conventional view. Additionally, the atomic structure of the critical nucleus for the Mg2Ca and MgZn2 Laves phases was predicted in terms of Classical Nucleation Theory (CNT). The formation of polytypes and off-stoichiometry in TCP precipitates is found to be related to the nonclassical nucleation behavior. These insights contribute to a deeper understanding of solid diffusive phase transformations at the atomic level and provide a foundation for investigating other technologically significant transformations.

Evaluation of Transformer Protection System Coordination at ULTG Jeneponto

Power transformers are essential components in electrical power systems as they transform electrical quantities. In practical terms, electrical power systems may experience disturbances such as short-circuit faults, either between phases or to ground. One type of relay used to accommodate these faults is the Over Current Relay (OCR) for phase-to-phase faults and the Ground Fault Relay (GFR) for ground faults. This research evaluates the coordination settings of transformer protection systems at PT PLN (Persero) ULTG Jeneponto by analyzing the characteristics plot of each transformer unit. The evaluation results reveal intersecting characteristics on the high voltage (HV) and medium voltage (MV) sides of Transformers #3 Tallasa and #3 Bulukumba. By implementing recommended settings for these two transformer units, anomalies can be addressed, and the protection system can coordinate effectively.

Lightweight convolutional neural networks with context broadcast transformer for real-time semantic segmentation

With the increasing application of embedded mobile devices in various fields, lightweight real-time semantic segmentation systems have attracted more and more attention. Many current methods have successfully reduced the model's parameters, but they have led to low model accuracy, diminishing their practical value. In recent years, the Transformer architecture has achieved good results in many tasks, effectively capturing long-range dependencies and enhancing accuracy. However, the Transformer is not adept at extracting local features, and the model's computational cost is generally too high, hindering real-time inference implementation. We propose a lightweight semantic segmentation network called LCBFormer-Net, which embeds Transformer units between asymmetric encoders and decoders to fully leverage their advantages. On the encoder side, we design the Lightweight Multi-Fusion Unit (LMFU) and Partition Grouping Shuffle Channel Attention (PGSCA). The former fully utilizes input features, merging information multiple times through multiple branches and employing depthwise convolutions with dilation rate to further obtain sufficient features. The latter includes a lightweight grouped channel attention, better guide feature extraction. The Lightweight Context Broadcast Transformer (LCB Transformer) is the Transformer unit we designed, with a lightweight structure that significantly reduces GPU memory consumption. It also improves self-attention and feed-forward networks, enhancing the model's robustness. The decoder includes the Multi-scale Semantic Information Attention Fusion (MSIAF) module, guiding the fusion of features at three different scales and employing a hybrid attention mechanism with both channel and spatial attention to guide feature extraction. LCBFormer-Net achieves good segmentation results with a parameter count of 0.88 M on multiple challenging datasets with diverse scenes.

Even‐Order Harmonic Distortion Observations During Multiple Geomagnetic Disturbances: Investigation From New Zealand

AbstractLarge geomagnetic storms are a space weather hazard to power transmission networks due to the effects of Geomagnetically Induced Currents (GICs). GIC can negatively impact power transmission systems through the generation of even‐order current and voltage harmonics due to half‐cycle transformer saturation. This study investigates a decade of even‐order voltage total harmonic distortion (hereon referred to as Even‐Order Total Harmonic Distortion (ETHD)) observations provided by Transpower New Zealand Ltd., the national system operator. We make use of ETHD measurements at 139 locations throughout New Zealand, monitored at 377 separate circuit breakers, focusing on 10 large geomagnetic disturbances during the period 2013–2023. Analysis identified 5 key substations, which appeared to act as sources of ETHD. The majority of these substations include single phase transformer banks, and evidence of significant GIC magnitudes. The ETHD from the source substations was found to propagate into the surrounding network, with the percentage distortion typically decaying away over distances of 150–200 km locally, that is, at a rate of −0.0043 %km−1. During the study period some significant changes occurred in the power network, that is, removal of the Halfway Bush (HWB) single phase bank transformer T4 in November 2017, and decommissioning of the New Plymouth substation in December 2019. Decommissioning of these two assets resulted in less ETHD occurring in the surrounding regions during subsequent geomagnetic storms. However, ETHD still increased at HWB with increasing levels of GIC, indicating that three phase transformer units were still susceptible to saturation, albeit with about 1/3 of the ETHD percentage exhibited by single phase transformers.

Analysis of existing approaches to monitoring and diagnostics of the active part of power transformers

Subject of research: monitoring and diagnostics of the condition of the active part of power transformers. Purpose of research: identification of promising trends in the development of methods for monitoring and diagnosing the state of the active part of power transformers. Object of research: power transformers. Methods of research: critical system analysis. Main results of research: the problem of monitoring and diagnosing the condition of the active part of power transformers, including windings, is considered. It is noted that a significant number (more than four million units) of distribution transformers are in operation in the Russian Federation. The causes of emergency failures of the specified class of power transformers are analyzed. One of the main causes of failures is highlighted – a winding circuit in the transformer winding. Attention is focused on the problem of monitoring the state of the active part of the transformer and on the diagnosis of the development of short circuits in the transformer winding. The presence of several groups of methods for monitoring and diagnosing winding circuits of transformer windings is shown. The economic attractiveness of monitoring and diagnostics based on the reaction to artificial electromagnetic influences, and monitoring and diagnostics based on the reaction to natural electromagnetic influences is shown.

Study of Problem and Effect Mitigation Technique of Unbalance Capacitance in 500 kV Single Phase Transformers

This article presents the study of the effect of unbalanced stray capacitance in single-phase 500 kV power transformers and establishes the mitigation technique. The power transformers were operated in Wang Noi Substation for transmitting the electrical power. There are 3 units of single-phase power transformer. Previously, a transformer of phase B failed after being serviced for several years. Thus, the new power transformer was ordered instead of the failed one. However, once the new system had been operated, the hotspots at the surge arrester of tertiary winding (22 kV) were found and the unbalance of phase voltages was observed, resulting in overvoltage as a root cause of the problem. Once a thorough investigation was done, it was found that there was an effect of unbalanced stray capacitance between tertiary winding and ground core since the new power transformer has some different designs compared with the remaining. Hence, the mitigation technique was performed by adding the capacitor parallel to the tertiary winding until the voltage balance was achieved. Simulation with the PSCAD program was performed to determine the optimum value for additional capacitors, i.e., voltage balance, using an iterative method. In this article, the capacitors used for additional were capacitor voltage transformers, which have been installed. As a result of the troubleshooting, the voltage at the 22 kV bus was more balanced as well as not causing overvoltage, and the hotspots at the surge arrester of tertiary winding were not detected resulting in gaining the stability and reliability of using these transformers.

Long-term Load Forecasting for Optimal Power System Planning and Decision-Making

Forecasting the future load growth of an area based on its load demand is often a proactive measure to ensure a steady electricity power supply to that area. The study focused on long-term load forecasting for power system planning, specifically examining the electric load demand from consumers on distribution transformers within Port Harcourt City, located in Rivers State, Nigeria. The study encompassed a comprehensive review of both statistical and artificial intelligence-based approaches. Historical load data for distribution transformer readings spanning 2008 to 2017 were acquired from the Port Harcourt Electricity Distribution Company (PHEDC) and subjected to analysis using the curve-fitting technique. For the period between 2015 and 2030, a yearly load forecast simulation was conducted using the Fourier Series model, implemented with MATLAB software. This simulation aimed to provide insights into future load demand, facilitating careful and informed decision-making in the investment, operation, and maintenance of power system equipment. The effectiveness of the forecasting investigation was assessed using the Root Mean Square Error (RMSE), confirming the efficiency, reliability, and validity of the employed model. The study's forecasted results are presented as a valuable guide and practical tool for policymakers and the utility company (PHEDC) to enhance proper planning and decision-making processes. Considering the observed trend in the results, it is suggested that installing additional transformer units in the region would be necessary to alleviate the loads on existing overloaded transformer units within the power system network.

GSSTU: Generative Spatial Self-Attention Transformer Unit for Enhanced Video Prediction.

Future frame prediction is a challenging task in computer vision with practical applications in areas such as video generation, autonomous driving, and robotics. Traditional recurrent neural networks have limited effectiveness in capturing long-range dependencies between frames, and combining convolutional neural networks (CNNs) with recurrent networks has limitations in modeling complex dependencies. Generative adversarial networks have shown promising results, but they are computationally expensive and suffer from instability during training. In this article, we propose a novel approach for future frame prediction that combines the encoding capabilities of 3-D CNNs with the sequence modeling capabilities of Transformers. We also propose a spatial self-attention mechanism and a novel neighborhood pixel intensity loss to preserve structural information and local intensity, respectively. Our approach outperforms existing methods in terms of structural similarity (SSIM), peak signal-to-noise ratio (PSNR), and learned perceptual image patch similarity (LPIPS) scores on five public datasets. More precisely, our model exhibited an average improvement of 4.64%, 18.5%, and 42% concerning SSIM, PSNR, and LPIPS for the second most proficient method correspondingly, across all datasets. The results demonstrate the effectiveness of our proposed method in generating high-quality predictions of future frames.

Digital transformation and integration of artificial intelligence in financial institutions

Purpose Integrating artificial intelligence (AI) into various industries, including the financial sector, has transformed them. This paper aims to examine the influence of integrating AI, including machine learning, process automation, predictive analytics and chatbots, on financial institutions and explores its various aspects and areas. The study aims to determine the impact of AI integration on financial services, products and customer experience. Design/methodology/approach The research study uses quantitative and qualitative methods, as well as secondary data analysis. It investigates four AI subfields: machine learning, process automation, predictive analytics and chatbots. Findings The research findings indicate that integrating AI, particularly in machine learning and chatbot subfields, holds promise and high strategic potential for financial institutions. These subfields can contribute significantly to enhancing financial services and customer experience. However, the significance of predictive analytics integration and process automation is relatively lower. Although these subfields retain their usefulness, they might necessitate alternative workflows and tools that incorporate human involvement. Overall, AI integration minimizes human interactions and errors in financial institutions. Originality/value The research study contributes original insights by exploring the specific subfields of AI within the financial industry and assessing their strategic significance. It provides recommendations for financial institutions to adopt AI integration partially in multiple phases, measure and evaluate the impact of the transformation and structure internal units and expertise to strategize adoption and change.

Computational Complexity Reduction of HEVC by Early Termination of Transform Unit Partitioning Using Raster Scan Approach

Computational Complexity Reduction of HEVC by Early Termination of Transform Unit Partitioning Using Raster Scan Approach

A multi-embedding domain video steganography algorithm based on TU partitioning and intra prediction mode

With High Efficiency Video Coding (HEVC) becoming a popular video coding standard in the world, combing the HEVC with data hiding methods is a complex task. In this paper, a multi-embedded video steganography algorithm based on Intra Prediction Mode (IPM) and Transform Unit (TU) partition is proposed. By analyzing the coding characteristics of the TU partition and the IPM, we select embedded regions that are relatively independent as much as possible. Then, suitable distortion functions are designed for different embedding domains. Using the minimum distortion framework of the STC model to embed messages. The experimental results indicate that the proposed method achieves a maximum embedding capacity 1.5 times larger than that of traditional single domain embedding methods, while maintaining good visual quality.

Future-shaping themes: Digital age foresight in Turkish university strategic plans

Background: Universities affect their communities and ecosystems both directly and indirectly. The strategic plans of all Turkish universities with Higher Education Council accreditation were thoroughly examined within this context. An ecosystem where digital technologies are used more efficiently and extensively is necessary for the ideal of a more sustainable environment and a world with happy people. Purpose: This study's main aim was to discover how universities in Türkiye predict the digital age via the emphasis they place on digital themes in their future plans. Study design/methodology/approach: The content analysis technique utilised within the scope of document analysis was chosen as the best method to satisfy this study's objectives. Findings/conclusions: The study found five main themes regarding the future of higher education based on codes obtained from university strategic plans. Digital transformation, digital campus, digital education, digital infrastructure, and future-focused units were these themes. We analysed the content of these five areas in connection with four types of universities around the country to see if there was a significant difference in attitude towards the digital age. Only future-focused units significantly distinguished these four types of universities, it was found. Limitations/future research: The research primarily examined the digital age themes that are evident in the strategic plans of universities. Academic studies for future research may involve an examination of the content of strategic plans in relation with different themes and categorizations. The collected results might be subjected to comparisons in order to assess the effectiveness of strategic plans.

RANCANGAN SISTEM KELISTRIKAN DATA CENTER BERSTANDAR TIER 3 PADA PERBANKAN

A data center is a facility used to place electronic systems and related components for the purposes of placing, storing and processing data. The data center is expected to be able to provide services as optimally as possible even in the event of a disturbance. Data center development must consider long-term development plans in terms of services, hardware and infrastructure. The majority of data centers in Indonesia are still standard tier 2. The majority of data centers with tier 3 standards have qualified specifications for companies that require high availability with good planning of the appropriate electrical system.The results of the research that has been done is that in designing a tier 3 standard data center, it must have full backups in the form of generators and UPS to meet the No Shutdown requirements. The use of estimated power designed in this study is 289433 Watts using a main power source from PLN using two transformer units backed up by four generator units and one UPS unit to back up critical loads.

RRTrN: A lightweight and effective backbone for scene text recognition

Models based on extremely deep convolutional networks and attention mechanisms now have powerful feature extraction abilities, improving text recognition performance in natural scenes. However, the very deep network model built with stacked layers brings massive parameters, which limits the application of the text recognition algorithm on storage-constrained devices. In this paper, we propose a lightweight and effective backbone called the Recursive Residual Transformer Network (RRTrN) for scene text recognition. Specifically, by leveraging recursive learning and a combination of convolutional layers and a transformer unit, RRTrN achieves powerful feature extraction while significantly reducing the number of parameters. This reduction in parameters promotes the deployment of our text recognition algorithm on storage-constrained devices, making it more accessible for practical applications. Furthermore, a recursive distillation strategy is presented to balance the recursive learning inference time and performance, enhancing the practicality and efficiency of RRTrN. Extensive experiments on mainstream benchmarks and popular models verify the generalization of RRTrN and achieve state-of-the-art recognition performance on five datasets. Notably, the classical STR model based on RRTrN can achieve a 3 percentage point increase in recognition accuracy or reduce the number of parameters by 80%.