Статті в журналах з теми "Transformers network"
Щоб переглянути інші типи публікацій з цієї теми, перейдіть за посиланням: Transformers network.
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 статей у журналах для дослідження на тему "Transformers network".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте статті в журналах для різних дисциплін та оформлюйте правильно вашу бібліографію.
1
S., S., Thulasi Bikku, P. Muthukumar, K. Sandeep, Jampani Chandra Sekhar, and V. Krishna Pratap. "Enhanced Intrusion Detection Using Stacked FT-Transformer Architecture." Journal of Cybersecurity and Information Management 8, no. 2 (2024): 19–29. http://dx.doi.org/10.54216/jcim.130202.
Анотація:
The function of network intrusion detection systems (NIDS) in protecting networks from cyberattacks is crucial. Many of the more conventional techniques rely on signature-based approaches, which have a hard time distinguishing between various types of assaults. Using stacked FT-Transformer architecture, this research suggests a new way to identify intrusions in networks. When it comes to dealing with complicated tabular data, FT-Transformers—a variant of the Transformer model—have shown outstanding performance. Because of the inherent tabular nature of network traffic data, FT-Transformers are an attractive option for intrusion detection jobs. In this area, our study looks at how FT-Transformers outperform more conventional machine learning (ML) methods. Our working hypothesis is that, in comparison to single-layered ML models, FT-Transformers will achieve better detection accuracy due to their intrinsic capacity to grasp long-range correlations in network traffic data. We also test the FT-Transformer model on several network traffic datasets that include various protocols and attack kinds to see how well it performs and how generalizable it is. The purpose of this research is to shed light on how well and how versatile FT-Transformers perform for detecting intrusions in networks. We aim to prove that FT-Transformers can secure networks from ever-changing cyber threats by comparing their performance to that of classic ML models and by testing their generalizability.
2
Jianwen, Mo, Mo Lunlin, Yuan Hua, Lin Leping, and Chen Lingping. "CNN with Embedding Transformers for Person Reidentification." Mathematical Problems in Engineering 2023 (July 14, 2023): 1–12. http://dx.doi.org/10.1155/2023/4591991.
Анотація:
For person reidentification (ReID), most slicing methods (such as part-based convolutional baseline (PCB) and AlignedReID) introduce a lot of background devoid of pedestrian parts, resulting in the cross-aliasing of features in the deep network. Besides, the resulting component features are not perfectly aligned with each other, thus affecting model performance. We propose a convolutional neural network (CNN) with embedding transformers (CET) person ReID network architecture based on the respective advantages of CNN and transformer. In CET, first, the residual transformer (RT) structure is first embedded in the backbone network of CNN to obtain a feature extractor, named transformers in CNN. The feature aliasing phenomenon is improved by utilizing transformer’s advantage in grasping the relevance of global information. Second, a feature fuse with learnable vector structure for fusing the output vector is added to the output of the transformer at the end of the network. A two branches loss structure is designed to balance the two different fusion strategies. Finally, the self-attention mechanism in transformer is used for automatic part alignment of human body parts to solve the part alignment problem caused by inaccurate detection frames. The experimental results show that CET network architecture achieves better performance than PCB and some other block-slicing methods.
3
Voitov, O. N., I. I. Golub, L. V. Semenova, E. V. Karpova, and A. L. Buchinsky. "Effects of unbalanced loads in a low-voltage network on flow distribution in a medium-voltage network." iPolytech Journal 28, no. 2 (July 4, 2024): 247–60. http://dx.doi.org/10.21285/1814-3520-2024-2-247-260.
Анотація:
We address the problem of improving the calculation accuracy of power flow in a medium-voltage distribution network based on the measurements of smart meters installed on the secondary side of 6(10)/0.4 kV transformers. In order to account for the effect of unbalanced loads in the low-voltage network on power flow in the medium-voltage network, three-phase three-wire lines were reduced to a single-line option. This enabled the use of symmetric mode calculation programs for the asymmetric mode. The loads in the medium-voltage network were determined by adding power losses in transformer windings and core to the loads measured on the secondary side of transformers. The calculation of winding power losses using the methods of phase coordinates and symmetrical components involves determination of currents in the windings of each phase according to 48 sections of load capacity and voltage module measurements, performed by the smart meter during the day. The correctness of expressions for calculating power losses in transformer windings is confirmed by the equality of total losses in phase coordinates and symmetrical components. The negative sequence power losses in transformer windings were found to be close to zero, while zero sequence losses are significantly lower than the positive sequence losses for almost all transformers with a double star-zero winding connection scheme, regardless of the load factor and rated power. The conducted studies confirmed the possibility and effectiveness of using smart meter measurements for determining loads and calculating power flow in the medium-voltage network. This conclusion was illustrated using an actual distribution network with 26 transformers. Future research should aim to clarify the mathematical models of transformers in the joint calculation of medium- and low-voltage distribution networks.
4
Krupa, Tadeusz. "Elements of Theory of the Correct Operations of Logistics Transforming Networks." Foundations of Management 9, no. 1 (December 20, 2017): 347–60. http://dx.doi.org/10.1515/fman-2017-0026.
Анотація:
Abstract In this paper, transformer logistics networks are treated as flow models of discrete manufacturing systems (FMDMS). The purpose is to formalize FMDMS into logical formulas called transformer functions. Transformer logistics networks are able to handle buffers and their production orders in a way that ensures full monitoring of the logic technology stored in the memory of a transforming network (t-network). The structural and functional complexity of the t-network makes it impossible to carry out formal proof of its proper functioning for any new order placement in buffers and transformers. This is because with the growing capacity in buffers, the number of available states of tnetworks also increases, and as such, the number of transformers and buffers unable to effectively generate new production orders that protect the t-network is also increasing. The problem therefore becomes to maintain t-network equilibrium technology that guarantees the continuity of the logical operations and processes of resource transformation.
5
DAUHIALA, D., V. TIKHANOVICH, and K. BABAMURATOV. "LOSSES IN NETWORK AND SOUND TRANSFORMERS." HERALD OF POLOTSK STATE UNIVERSITY. Series С FUNDAMENTAL SCIENCES, no. 1 (April 18, 2023): 38–43. http://dx.doi.org/10.52928/2070-1624-2023-40-1-38-43.
Анотація:
The causes of losses in transformers are described. The main types of energy losses in the transformer are
revealed. The concepts of Q-factor and efficiency are considered. The estimation of the passage of audio frequency
signals is given, the principles and rules for constructing frequency characteristics are considered. Recommendations
for the design of sound and mains frequency transformers are given. Constructive measures to reduce losses are
considered. The given data can be used for the design of network and sound transformers.
6
Hanus, Oleksii, and Kostiantyn Starkov. "STUDY OF THE NATURE OF OVERVOLTAGES IN THE ELECTRICAL NETWORK ARISING FROM VOLTAGE TRANSFORMERS." Bulletin of the National Technical University "KhPI". Series: Energy: Reliability and Energy Efficiency, no. 1 (2) (July 2, 2021): 28–36. http://dx.doi.org/10.20998/2224-0349.2021.01.05.
Анотація:
A non-linear dynamic mathematical model of voltage transformer has been considered and overvoltages arising on the elements of voltage transformer equivalent circuit during transient processes have been investigated. The influence of voltage transformer secondary circuit capacitance on overvoltage multiplicity in the primary circuits and the duration of transients has been determined. The advantages of approximation of nonlinearity of voltage transformers by hyperbolic sine are used. Mathematical expressions determining the nature of changes in the forced and free components of the transient process in an electrical network with a voltage transformer have been obtained. It is shown that with the increase of the electric network capacitance the duration of the transition process damping increases and the frequency of the forced oscillations and the level of overvoltage decrease. It is proved that even small, in comparison with the primary nominal sinusoidal voltage, aperiodic components of the voltage transient process can lead to significant overvoltages during voltage transformer outages. It has been substantiated that both the secondary resistance and the switching torque influence the overvoltage multiplicity arising in the primary winding of voltage transformers. It is shown that the closed secondary winding worsens the disconnection process of non-linear inductance of voltage transformers. The values to which overvoltages increase in this case are determined. According to the results of calculations it is determined that with open secondary winding of voltage transformers the duration of transient process significantly increases. It has been found that the decrease of frequency of forced oscillations, which occurs in this case, is accompanied by an increase of currents in the primary winding of the voltage transformer, which is dangerous in terms of thermal stability of the winding insulation. It is shown, that closing the secondary winding of voltage transformers leads to significant reduction of transient damping time. It is suggested that this algorithm can be used to provide a rapid breakdown (suppression) of ferroresonant processes. The effectiveness of such a measure of stopping of ferroresonance processes as short-term shunting of secondary winding of voltage transformers has been investigated. The correlation of parameters of electric networks (capacity of busbar sections, nonlinearity of characteristics of voltage transformers, disconnection torque, etc.) at which ferroresonance process may occur and consideration of which may allow, in terms of prevention of ferroresonance processes, to identify substations (electric networks) that require more detailed research has been determined. The results of analytical studies were tested in the electric networks of JSC "Kharkivoblenergo" and used in the electricity distribution system for the selection of specific voltage transformers for certain configurations of electrical networks.
7
Lakehal, Abdelaziz, and Fouad Tachi. "Bayesian Duval Triangle Method for Fault Prediction and Assessment of Oil Immersed Transformers." Measurement and Control 50, no. 4 (May 2017): 103–9. http://dx.doi.org/10.1177/0020294017707461.
Анотація:
Dissolved gas analysis of transformer insulating oil is considered the best indicator of a transformer’s overall condition and is most widely used. In this study, a Bayesian network was developed to predict failures of electrical transformers. The Duval triangle method was used to develop the Bayesian model. The proposed prediction model represents a transformer fault prediction, possible faulty behaviors produced by this transformer (symptoms), along with results of possible dissolved gas analysis. The model essentially captures how possible faults of a transformer can manifest themselves by symptoms (gas proportions). Using our model, it is possible to produce a list of the most likely faults and a list of the most informative gas analysis. Also, the proposed approach helps to eliminate the uncertainty that could exist, regarding the fault nature due to gases trapped in the transformer, or faults that result in more simultaneous gas percentages. The model accurately provides transformer fault diagnosis and prediction ability by calculating the probability of released gases. Furthermore, it predicts failures based on their relationships in the Bayesian network. Finally, we show how the approach works for five distinct electrical transformers of a power plant, by describing the advantages of having available a Bayesian network model based on the Duval triangle method for the fault prediction tasks.
8
Azmi Murad Abd Aziz, Mohd Aizam Talib, Ahmad Farid Abidin, and Syed Abdul Mutalib Al Junid. "Development of Power Transformer Health Index Assessment Using Feedforward Neural Network." Journal of Advanced Research in Applied Sciences and Engineering Technology 30, no. 3 (May 15, 2023): 276–89. http://dx.doi.org/10.37934/araset.30.3.276289.
Анотація:
The role of a power transformer is to convert the electrical power level and send it to the consumer, making it an essential component of a power system. In addition, transformer asset management is essential for monitoring the functioning of transformers in the system to prevent failure and anticipating the health state of transformers, using a technique known as the health index (HI). However, the calculation and computation to determine the transformer HI based on a scoring and ranking technique is complex and required expert validation. Therefore, this paper presents a transformer HI prediction using a feedforward neural network (FFNN) to improve the existing complex scoring and ranking technique. Levenberg–Marquardt (LM), Bayesian Regularized (BR), and Scaled Conjugate Gradient (SCG) are the FFNN training techniques presented in this study to forecast the transformer HI. To validate the techniques, the HI values generated by different FFNN techniques were compared to the scoring and ranking system. Then, the performance of the proposed ANN was evaluated using the correlation coefficient and mean square error (MSE). As a result, the transformer HI was successfully predicted by employing three FFNN techniques, namely the LM, BR, and SCG techniques, which were able to determine whether the transformer's condition is very good, good, fair, or poor. In conclusion, the ANN suggested in this study has also been validated with the ranking and scoring approach, which provides high similarity score in comparison to the transformer health index.
9
Varga, Aleksandr A., and Galaktion V. Shvedov. "Estimating the Error of Calculating the Load Losses in 6—10 kV Distribution Electric Networks." Vestnik MEI, no. 5 (2021): 37–43. http://dx.doi.org/10.24160/1993-6982-2021-5-37-43.
Анотація:
The aim of the study is to estimate the level of load losses for actual (nonuniform) loading of transformer substations in comparison with the load loss level under the conditions of their being uniformly loaded for a typical 6 kV urban distribution electric network. The extent to which such networks are equipped with automated information and measurement systems for fiscal electricity metering is extremely poor. Therefore, not only the load curves of the 6--10/0.4 kV substation transformers, but also their load factors are unknown in the majority of cases. Under these conditions, in calculating the load losses in 6--10 kV distribution electric networks, it is assumed that all substation transformers are uniformly (equally) loaded. The load losses in these electrical networks are calculated with a certain error associated with the assumption according to which the substation transformers are supposed to be uniformly loaded. Under these conditions, estimation of this error becomes of issue. The article describes the modeling, calculation and analysis of technical losses of electricity under the conditions of nonuniformly loaded transformers of a typical urban distribution network consisting of four transformers and four cables interconnecting them. The modeling and calculation of power losses were carried out using the RAP-10-st computer program for several different groups of transformer loading factors. Within each group, different subgroups were produced by rearranging the group loading factors. Each subgroup modeled a nonuniform transformers loading mode in the studied network. For each of these modes, the power losses were calculated and studied with the use of the RAP-10-st computer program. A conclusion has been drawn from the obtained analysis results regarding the error in determining the load losses associated with the assumption about the uniform loading of the substation transformers in the network. The obtained results may prompt electric grid companies to increase the extent of fitting their networks with automated information and measurement systems for fiscal electricity metering to improve the accuracy of determining the load losses.
10
Nadhirah, Nurul Fatin, Hana Abdull Halim, Nurhakimah Mohd Mukhtar, and Samila Mat Zali. "Varying the energisation condition to mitigate sympathetic inrush current." International Journal of Electrical and Computer Engineering (IJECE) 13, no. 6 (December 1, 2023): 5975. http://dx.doi.org/10.11591/ijece.v13i6.pp5975-5985.
Анотація:
Transformers are generally easy to access and can contribute significantly to entire power system. When a transformer is turned on for the first time, it produces a magnetising inrush current which acts as a starting current. Energisation of transformer has a substantial impact on inrush current and transformer that are connected in parallel. Sympathetic inrush current is a phenomenon that appears when a transformer is switched-on in network whereas the other transformers that was earlier energised. Besides, when sympathetic inrush phenomena occur, the peak and period fluctuate significantly. In this paper, the transformers will be energised in three different ways and each condition will be explored in depth. The operation time of the transformer’s energisation whether it is energised simultaneously or at different times are tested and analysed in terms of their characteristics. It is performed using power system computer aided design (PSCAD) software, starting with a develop model of the energisation and then generate the outcomes. The results of the simulation demonstrate that energising the transformer in different ways can give different effect on the sympathetic inrush current, as well as the variables that affect it and methods for reducing it.
11
Lee, Choongman, Gyu-Jung Cho, and Joorak Kim. "Development of Scott Transformer Model in Electromagnetic Transients Programs for Real-Time Simulations." Applied Sciences 11, no. 12 (June 21, 2021): 5752. http://dx.doi.org/10.3390/app11125752.
Анотація:
This paper presents a Scott transformer model to be applied in electromagnetic transients (EMT) programs, particularly in the absence of a detailed Scott transformer model for performing real-time simulations (RTS). Regarding a Scott transformer, a common topology for converting a three-phase network into two single-phase networks, the transformer model in EMT programs is essential to simulate large-scale electric railway systems. A code-based model has been developed to simulate the transformer in RTS directly and contain the transformer’s actual impedance characteristics. By establishing a mathematical foundation with the current injection method, we presented a matrix representation in conjunction with a network solution of EMT programs. The proposed model can handle more practical parameters of Scott transformers with a relatively low computational load. Thus, it supports the flexible computation of real-time simulators with a finite number of processor units. The accuracy of the model is verified by simulating it and comparing the simulation results with an industrial transformer’s certified performance. Furthermore, a case study involving a comparison of the results with the field measurement data of an actual Korean railway system demonstrated the efficacy of the model.
12
Lara, Hector, and Esteban Inga. "Efficient Strategies for Scalable Electrical Distribution Network Planning Considering Geopositioning." Electronics 11, no. 19 (September 28, 2022): 3096. http://dx.doi.org/10.3390/electronics11193096.
Анотація:
This article presents a heuristic model to find the optimal route or layout of a subway electrical distribution network, obtaining full coverage of users in different scenarios and respecting technical criteria such as maximum distance to avoid voltage drop and capacity. In this way, the location of the transformer substations is achieved through an analysis of candidate sites. The medium voltage network will connect each transformer to a minimum spanning tree (MST), reducing the cost of materials associated with constructing the electrical grid. This work considers the latitude and longitude of each house and electrical count. Georeferenced scenario information is taken from the OpenStreetMap platform to provide an authentic context for distance and location calculations in the deployment of the power grid. The heuristic model offers to decrease time in solving the electrical network layout. As input variables, different powers of the "multi-transformer" transformers are considered to minimize the number of transformers and solve the power supply, reducing the transformers’ oversizing and minimizing the transformers’ idle capacity. The experimentation showed that none exceeded the limit allowed in an urban area of 3.5%.
13
Ilyas, Iriandi, and Muhamad Taufan Agassy. "ANALISIS KEGAGALAN CURRENT TRANSFORMER (CT) TIPE DUA BELITAN SEKUNDER DENGAN INTI MAGNETIK TERPISAH PADA SISTEM PROTEKSI DAN PEMBATAS DAYA." SAINSTECH: JURNAL PENELITIAN DAN PENGKAJIAN SAINS DAN TEKNOLOGI 32, no. 1 (March 29, 2022): 41–48. http://dx.doi.org/10.37277/stch.v32i1.1240.
Анотація:
ABSTRACT
Current transformers function to transform large currents into small currents used for electrical energy measurement instruments, power limitation and protection of electric power networks, current transformers have many types, the use of current transformers in actual facts such as electrical power requested by consumers to producers of electricity providers at 53 kVA to 197 kVA using a single ratio current transformer type and> 197 kVA using a secondary twisting transformer type with a separate magnetic core In this type of current transformers with different ratios have different saturation values also due to high interference current values, so they are not able to provide an accurate secondary current value according to the class and the ability of current transformers against measuring instruments, power limitation and protection of electric power networksBy analyzing the failure of these types can be a reference or reference in the use of current transformers by selecting the appropriate type, ratio and accuracy class on the electricity network
14
Alharthi, Musleh, and Ausif Mahmood. "Enhanced Linear and Vision Transformer-Based Architectures for Time Series Forecasting." Big Data and Cognitive Computing 8, no. 5 (May 16, 2024): 48. http://dx.doi.org/10.3390/bdcc8050048.
Анотація:
Time series forecasting has been a challenging area in the field of Artificial Intelligence. Various approaches such as linear neural networks, recurrent linear neural networks, Convolutional Neural Networks, and recently transformers have been attempted for the time series forecasting domain. Although transformer-based architectures have been outstanding in the Natural Language Processing domain, especially in autoregressive language modeling, the initial attempts to use transformers in the time series arena have met mixed success. A recent important work indicating simple linear networks outperform transformer-based designs. We investigate this paradox in detail comparing the linear neural network- and transformer-based designs, providing insights into why a certain approach may be better for a particular type of problem. We also improve upon the recently proposed simple linear neural network-based architecture by using dual pipelines with batch normalization and reversible instance normalization. Our enhanced architecture outperforms all existing architectures for time series forecasting on a majority of the popular benchmarks.
15
Shao, Ran, Xiao-Jun Bi, and Zheng Chen. "A novel hybrid transformer-CNN architecture for environmental microorganism classification." PLOS ONE 17, no. 11 (November 11, 2022): e0277557. http://dx.doi.org/10.1371/journal.pone.0277557.
Анотація:
The success of vision transformers (ViTs) has given rise to their application in classification tasks of small environmental microorganism (EM) datasets. However, due to the lack of multi-scale feature maps and local feature extraction capabilities, the pure transformer architecture cannot achieve good results on small EM datasets. In this work, a novel hybrid model is proposed by combining the transformer with a convolution neural network (CNN). Compared to traditional ViTs and CNNs, the proposed model achieves state-of-the-art performance when trained on small EM datasets. This is accomplished in two ways. 1) Instead of the original fixed-size feature maps of the transformer-based designs, a hierarchical structure is adopted to obtain multi-scale feature maps. 2) Two new blocks are introduced to the transformer’s two core sections, namely the convolutional parameter sharing multi-head attention block and the local feed-forward network block. The ways allow the model to extract more local features compared to traditional transformers. In particular, for classification on the sixth version of the EM dataset (EMDS-6), the proposed model outperforms the baseline Xception by 6.7 percentage points, while being 60 times smaller in parameter size. In addition, the proposed model also generalizes well on the WHOI dataset (accuracy of 99%) and constitutes a fresh approach to the use of transformers for visual classification tasks based on small EM datasets.
16
Moon, Ji-Hwan, Gyuho Choi, Yu-Hwan Kim, and Won-Yeol Kim. "PCTC-Net: A Crack Segmentation Network with Parallel Dual Encoder Network Fusing Pre-Conv-Based Transformers and Convolutional Neural Networks." Sensors 24, no. 5 (February 24, 2024): 1467. http://dx.doi.org/10.3390/s24051467.
Анотація:
Cracks are common defects that occur on the surfaces of objects and structures. Crack detection is a critical maintenance task that traditionally requires manual labor. Large-scale manual inspections are expensive. Research has been conducted to replace expensive human labor with cheaper computing resources. Recently, crack segmentation based on convolutional neural networks (CNNs) and transformers has been actively investigated for local and global information. However, the transformer is data-intensive owing to its weak inductive bias. Existing labeled datasets for crack segmentation are relatively small. Additionally, a limited amount of fine-grained crack data is available. To address this data-intensive problem, we propose a parallel dual encoder network fusing Pre-Conv-based Transformers and convolutional neural networks (PCTC-Net). The Pre-Conv module automatically optimizes each color channel with a small spatial kernel before the input of the transformer. The proposed model, PCTC-Net, was tested with the DeepCrack, Crack500, and Crackseg9k datasets. The experimental results showed that our model achieved higher generalization performance, stability, and F1 scores than the SOTA model DTrC-Net.
17
Zhou, Mingjie, Jing Xu, Chaojian Xing, Yankai Li, and Shuxin Liu. "Research for transformer operation state prediction method based on BO-CNN-GRU." Journal of Physics: Conference Series 2770, no. 1 (May 1, 2024): 012013. http://dx.doi.org/10.1088/1742-6596/2770/1/012013.
Анотація:
Abstract In the realm of electrical engineering, this study introduces a novel approach for forecasting transformer operational conditions by leveraging BO-CNN-GRU (Bayesian Optimized Convolutional Neural Network Gated Loop Unit). The initial step involves an in-depth analysis of the key parameters that significantly impact the operational performance of the transformer. Then, the comprehensive weights of each characteristic parameter of the transformer are obtained by the G1 method, entropy weight method, and CRITIC method, and the comprehensive state data of the transformer is obtained. Finally, BO (Bayesian optimization) and CNN (convolutional neural network) are used to optimize the GRU neural network to form a comprehensive prediction model of the future operation state for the transformer based on BO-CNN-GRU. Employing transformer operation status data, we develop a unified neural network model to forecast the forthcoming operation statuses of transformers. This model can more accurately and faster predict the future status of transformers. By analyzing specific cases, it has been determined that the predictive model’s average accuracy in forecasting transformer operational status one month ahead is at an impressive 98.44%, which can accurately predict the future state changes of transformers.
18
Tan, Chang, Jianxun Hong, Zihao Wu, Qiuyuan Huang, and Cheng Wang. "Parasitic Parameter Prediction for Planar Transformers Based on Neural Network." Journal of Physics: Conference Series 2584, no. 1 (September 1, 2023): 012083. http://dx.doi.org/10.1088/1742-6596/2584/1/012083.
Анотація:
Abstract Parasitic parameters such as leakage inductance and distributed capacitance of planar transformers have a direct impact on the performance and efficiency of transformers. Traditional methods for parasitic parameter prediction are commonly based on empirical formulas or simulation software, but they have problems of high computational complexity, time-consuming and low accuracy. In this paper, a method for predicting parasitic parameters of planar transformers based on a multilayer perceptron (MLP) under a specific winding structure is proposed, which can improve the efficiency of transformer design. The experiments demonstrate that the model can effectively predict the leakage inductance, distributed capacitance, and AC loss of planar transformers.
19
Krishna, A. Prudhvi, P. Srinivasa Varma, R. B. R. Prakash, and V. Kiran Babu. "Prioritization of network transformers in electrical distribution system by considering social welfare index." Indonesian Journal of Electrical Engineering and Computer Science 16, no. 1 (October 1, 2019): 25. http://dx.doi.org/10.11591/ijeecs.v16.i1.pp25-32.
Анотація:
<span lang="EN-US">To supply a meshed distribution system, network transformers are required. When few transformers are not in service, they must be repaired or replaced. A method is proposed for prioritizing the transformers considering the critical loads. Repair or replacement of transformers can be done by giving priority based on risk reduction. By addressing the possibility of network collapse due to failure of the feeder and impacted customers, risk can be predicted where the loads are extremely used at feeders section, network transformers and secondary mains. To select the transformer that needs to be replaced quickly and economically, an algorithm is proposed and it was tested on IEEE test system using GridLAB-D, MATLAB softwares. An index is proposed to give priority to emergency needs like hospitals and water pumping stations. Replacement or repair can be done by prioritizing network transformers incorporating social welfare index. </span>
20
Fiennes, J., and C. R. de Souza. "The Complex Transformer as a Network-Model Element." International Journal of Electrical Engineering & Education 40, no. 1 (January 2003): 27–35. http://dx.doi.org/10.7227/ijeee.40.1.3.
21
Goran Jerbić. "APPLICATION OF PHASE SHIFTING TRANSFORMERS IN THE CROATIAN POWER SUPPLY SYSTEM." Journal of Energy - Energija 56, no. 2 (November 16, 2022): 216–31. http://dx.doi.org/10.37798/2007562353.
Анотація:
The phase shifting transformers with on load tap changer are increasingly found in modern power supply systems, especially under the present conditions of the full opening of the electricity market. The construction of the Žerjavinec TS 400/220/110 kV for the first time introduces into the Croatian transmission system a 400/220/(10,5) kV 400 MVA phase shifting network transformer with on load tap changer. The present article highlights some specific aspects of phase shifting transformers in the light of their application in the Croatian system. For a more efficient use of the advantages of phase shifting transformers, the transformer of this type at Žerjavinec would need to be provided with a complementary transformer (pair), either within the Croatian transmission network or within the neighbouring systems.
22
Kumari, Rekha, Gurpreet Kaur, Aditya Rawat, Harshit Chauhan, Kartik Singh Negi, and Rishi Mishra. "ANALYSIS OF TRANSFORMER-DEEP NEURAL NETWORK USING DEEP LEARNING." International Journal of Engineering Applied Sciences and Technology 8, no. 2 (June 1, 2023): 313–19. http://dx.doi.org/10.33564/ijeast.2023.v08i02.048.
Анотація:
Transformers were first used for natural language processing (NLP) tasks, but they quickly spread to other deep learning fields, including computer vision. They assess the interdependence of pairs. Attention is a part that enables to dynamically highlight relevant features of the input data (words in the case of text strings, parts of images in the case of visual Transformers). The cost grows continually with the number of tokens. The most common Trans- former Architecture for image classification uses only the Transformer Encoder to transform the various input tokens. However, the decoder component of the traditional Transformer Architecture is also used in a variety of other applications. In this section, we first introduce the Attention Mechanism (Section 1), followed by the Basic Transformer Block, which includes the Vision Transformer (Section 2).
23
N. S., Okorie. "Optimization of Energy Efficiency for Electric Power Distribution System Losses." International Journal for Research in Applied Science and Engineering Technology 9, no. 12 (December 31, 2021): 934–40. http://dx.doi.org/10.22214/ijraset.2021.39074.
Анотація:
Abstract: This study evaluated the existing electric power network of Mile 2 Diobu zone, Port Harcourt distribution network which consists of four (4) 11kV distribution feeders namely; Ojoto, Nsukka, Udi and Silverbird. This work considered Ojoto and Nsukka Street distribution network for improved power quality. The three (3) 33/11kv injection substations are fed from 165 MVA transmission station (PH Town) at Amadi junction by Nzimiro. Collection and analysis of data collected from the injection substations that supply electricity to mile 2 Diobu, Port Harcourt was the first consideration. The distribution network was modeled in Electrical Transient Analyzer Program (ETAP) using Newton-Raphson Load Flow equations. The simulation result of the existing condition network shows that the network has low voltage profile problem on Nsukka network and overloading of distribution transformers on Ojoto networks. The following optimization techniques are applied: up-gradation of distribution transformers, and transformer load tap changer to improve the distribution network for Mile 2 Diobu, Port Harcourt electrical power network. The simulation result of the improved distribution network for Mile 2 Diobu, Port Harcourt power network shows that the voltage profile Nsukka network has improved within the statutory limit which is between 95.0 -105.0% and the loading of the distribution transformers on Ojoto and Nsukka networks are all below 70% required capacity. Keywords: Optimization, Energy Efficiency Distribution
24
Petrescu, L., E. Cazacu, V. Ioniţă, and Maria-Cătălina Petrescu. "An Experimental Device for Measuring the Single-Phase Transformers Inrush Current." Scientific Bulletin of Electrical Engineering Faculty 19, no. 1 (April 1, 2019): 18–22. http://dx.doi.org/10.1515/sbeef-2019-0004.
Анотація:
AbstractElectrical transformers are essential parts of power supply networks and it is important that their life-time to be preserved. The inrush current of this devices could determine malfunctioning of the transformers or even others component of the network. For this reason, determining the inrush current for single-phase transformers is an important issue in power quality analysis of electrical grids. In this paper we presented an experimental device (hardware set-up and software program) that can measure this in rush current features for small transformers (up to 10 kVA). Also, the device affords the users to measure inrush current knowing the geometry of the transformer, the dimensions and the magnetic characteristic of the core.
25
Jamali, Ali, and Masoud Mahdianpari. "Swin Transformer and Deep Convolutional Neural Networks for Coastal Wetland Classification Using Sentinel-1, Sentinel-2, and LiDAR Data." Remote Sensing 14, no. 2 (January 13, 2022): 359. http://dx.doi.org/10.3390/rs14020359.
Анотація:
The use of machine learning algorithms to classify complex landscapes has been revolutionized by the introduction of deep learning techniques, particularly in remote sensing. Convolutional neural networks (CNNs) have shown great success in the classification of complex high-dimensional remote sensing imagery, specifically in wetland classification. On the other hand, the state-of-the-art natural language processing (NLP) algorithms are transformers. Although the transformers have been studied for a few remote sensing applications, the integration of deep CNNs and transformers has not been studied, particularly in wetland mapping. As such, in this study, we explore the potential and possible limitations to be overcome regarding the use of a multi-model deep learning network with the integration of a modified version of the well-known deep CNN network of VGG-16, a 3D CNN network, and Swin transformer for complex coastal wetland classification. Moreover, we discuss the potential and limitation of the proposed multi-model technique over several solo models, including a random forest (RF), support vector machine (SVM), VGG-16, 3D CNN, and Swin transformer in the pilot site of Saint John city located in New Brunswick, Canada. In terms of F-1 score, the multi-model network obtained values of 0.87, 0.88, 0.89, 0.91, 0.93, 0.93, and 0.93 for the recognition of shrub wetland, fen, bog, aquatic bed, coastal marsh, forested wetland, and freshwater marsh, respectively. The results suggest that the multi-model network is superior to other solo classifiers from 3.36% to 33.35% in terms of average accuracy. Results achieved in this study suggest the high potential for integrating and using CNN networks with the cutting-edge transformers for the classification of complex landscapes in remote sensing.
26
Jiao, Jinyue, Zhiqiang Gong, and Ping Zhong. "Dual-Branch Fourier-Mixing Transformer Network for Hyperspectral Target Detection." Remote Sensing 15, no. 19 (September 24, 2023): 4675. http://dx.doi.org/10.3390/rs15194675.
Анотація:
In recent years, transformers have shown great potential in hyperspectral image processing and have also been gradually applied in hyperspectral target detection (HTD). Nonetheless, applying a typical transformer to HTD remains challenging. The heavy computation burden of the multi-head self-attention (MSA) in transformers limits its efficient HTD, while the limited ability to extract local spectral features can reduce the discrimination of the learned spectral features. To further explore the potential of transformers for HTD, for balance of representation ability and computational efficiency, we propose a dual-branch Fourier-mixing transformer network for hyperspectral target detection (DBFTTD). First, this work explores a dual-branch Fourier-mixing transformer network. The transformer-style network replaces the MSA sublayer in the transformer with a Fourier-mixing sublayer, which shows advantages in improving computational efficiency and learning valuable spectral information effectively for HTD. Second, this work proposes learnable filter ensembles in the Fourier domain that are inspired by ensemble learning to improve detection performance. Third, a simple but efficient dropout strategy is proposed for data augmentation. Sufficient and balanced training samples are constructed for training the dual-branch network, and training samples for balanced learning can further improve detection performance. Experiments on four data sets indicate that our proposed detector is superior to the state-of-the-art detectors.
27
Cremasco, Andrea, Wei Wu, Andreas Blaszczyk, and Bogdan Cranganu-Cretu. "Network modelling of dry-type transformer cooling systems." COMPEL - The international journal for computation and mathematics in electrical and electronic engineering 37, no. 3 (May 8, 2018): 1039–53. http://dx.doi.org/10.1108/compel-12-2016-0534.
Анотація:
Purpose The application of dry-type transformers is growing in the market because the technology is non-flammable, safer and environmentally friendly. However, the unit dimensions are normally larger and material costs become higher, as no oil is present for dielectric insulation or cooling. At designing stage, a transformer thermal model used for predicting temperature rise is fundamental and the modelling of cooling system is particularly important. This paper aims to describe a thermal model used to compute dry transformers with different cooling system configurations. Design/methodology/approach The paper introduces a fast-calculating thermal and pressure network model for dry-transformer cooling systems, preliminarily verified by analytical methods and advanced CFD simulations, and finally validated with experimental results. Findings This paper provides an overview of the network model of dry-transformer cooling system, describing its topology and its main variants including natural or forced ventilation, with or without cooling duct in the core, enclosure with roof and floor ventilation openings and air barriers. Finally, it presents a formulation for the new heat exchanger element. Originality/value The network approach presented in this paper allows to model efficiently the cooling system of dry-type transformers. This model is based on physical principles rather than empirical assessments that are valid only for specific transformer technologies. In comparison with CFD simulation approach, the network model runs much faster and the accuracies still fall in acceptable range; therefore, one is able to utilize this method in optimization procedures included in transformer design systems.
28
Shen, Zhan, Bingxin Xu, Chenglei Liu, Cungang Hu, Bi Liu, Zhike Xu, Long Jin, and Wu Chen. "The Modeling and Simplification of a Thermal Model of a Planar Transformer Based on Internal Power Loss." Sustainability 14, no. 19 (September 21, 2022): 11915. http://dx.doi.org/10.3390/su141911915.
Анотація:
With the development of high-performance wide-band-gap devices and increasing converter frequency, planar transformers are widely used in high-frequency and high-power-density power conversions. Due to the skin effect and proximity effect, accurate thermal analysis and a simplified thermal model of planar transformers are needed for quick thermal verification as well as system design. This paper proposes two thermal simplification models based on the planar transformer’s thermal impedance network. The internal power loss and thermal coupling between each component are first analyzed. Then, based on thermal radiation theory, the simplified thermal model of the planar transformer is presented. It only requires the input of the total power loss of the planar transformer to calculate the temperature rise, and it does not need the power loss of each component. Finally, the simulation and experimental verification are carried out on a MHz prototype.
29
Xu, Honghua, Yong Li, Lei Zhu, and Ziqiang Xu. "Condition assessment of transformers in wind farm based on modified one-dim residual neural network." Journal of Physics: Conference Series 2378, no. 1 (December 1, 2022): 012078. http://dx.doi.org/10.1088/1742-6596/2378/1/012078.
Анотація:
Abstract The working environment of transformers in the wind farm is more complex than others, which brings the difference in condition assessment. Moreover, many condition assessment methods based on characteristics or machine learning have difficulty in recognition in cases of multiple transformers, conditions and measuring points. To assess conditions, this paper establishes a condition classification model of the transformer with a modified one-dim residual neural network and uses vibration signal, current and voltage as inputs. The built network mode has faster convergence speed and classification accuracy in transformer condition assessment and is more suitable for transformer condition assessment than the original one.
30
Bae, Jinwoo, Sungho Moon, and Sunghoon Im. "Deep Digging into the Generalization of Self-Supervised Monocular Depth Estimation." Proceedings of the AAAI Conference on Artificial Intelligence 37, no. 1 (June 26, 2023): 187–96. http://dx.doi.org/10.1609/aaai.v37i1.25090.
Анотація:
Self-supervised monocular depth estimation has been widely studied recently. Most of the work has focused on improving performance on benchmark datasets, such as KITTI, but has offered a few experiments on generalization performance. In this paper, we investigate the backbone networks (e.g., CNNs, Transformers, and CNN-Transformer hybrid models) toward the generalization of monocular depth estimation. We first evaluate state-of-the-art models on diverse public datasets, which have never been seen during the network training. Next, we investigate the effects of texture-biased and shape-biased representations using the various texture-shifted datasets that we generated. We observe that Transformers exhibit a strong shape bias and CNNs do a strong texture-bias. We also find that shape-biased models show better generalization performance for monocular depth estimation compared to texture-biased models. Based on these observations, we newly design a CNN-Transformer hybrid network with a multi-level adaptive feature fusion module, called MonoFormer. The design intuition behind MonoFormer is to increase shape bias by employing Transformers while compensating for the weak locality bias of Transformers by adaptively fusing multi-level representations. Extensive experiments show that the proposed method achieves state-of-the-art performance with various public datasets. Our method also shows the best generalization ability among the competitive methods.
31
Rysev, Pavel V., Mikhail S. Peshko, and Alexander O. Shepelev. "Intelligent voltage regulation system in the distribution electrical network based on fuzzy logic." Yugra State University Bulletin 18, no. 3 (October 8, 2022): 107–17. http://dx.doi.org/10.18822/byusu202203107-117.
Анотація:
The paper considers the issues of voltage regulation in the distribution network. The possibility of using fuzzy logic implemented in a fuzzy controller for voltage control is shown.
The purpose of the study: to increase the resource of switches of voltage regulators of power transformers when controlling voltage and reactive power modes of distribution electrical networks.
Subject of research: the introduction of a fuzzy controller in the voltage level control system in the distribution electrical network.
Methods and objects of research: the study was conducted using numerical simulation, fuzzy logic methods. The objects of the study were the distribution electrical network as well as power transformers equipped with voltage regulators, adjustable compensating devices working together under the control of a fuzzy controller.
Results of research: based on the conducted experiments, it was shown that the joint use of voltage regulation means (transformers, compensating devices) in the distribution electrical network when controlling a fuzzy controller allows to increase the resource and extend the service life of transformer switching devices.
32
Jasika, Ranko, Katarina Maksić, and Jovan Mrvić. "Investigation of ferroresonance phenomena in isolated 6 kV power network." Zbornik radova Elektrotehnicki institut Nikola Tesla, no. 33 (2023): 1–13. http://dx.doi.org/10.5937/zeint33-47844.
Анотація:
Ferroresonance is a complex nonlinear electrical phenomenon that can cause significant dielectric and thermal stresses in power equipment, leading to potential faults, equipment damage, power supply interruptions, etc. Three-phase networks with isolated neutral point and single-phase isolated inductive voltage transformers are susceptible to ferroresonance, caused by the interaction of the transformer's nonlinear inductances and zero sequence capacitance of power network. The paper analyzes the possibility of ferroresonance occurrence in a real 6 kV power network, with an isolated neutral point, using the electromagnetic transients program ATP/EMTP. A detailed network model is developed, taking into account the nonlinearity of voltage and power transformers, which is utilized to analyze various configurations, switching operations, and faults in the network that could trigger ferroresonance. It is shown that, under certain circumstances, ferroresonance can occur, with faults clearing being the most common initiators of ferroresonance. The efficiency of different measures to suppress ferroresonance is also analyzed. The impact of the resistance values of voltage transformer's open delta winding dumping resistor on the suppression of ferroresonance occurrence, is considered.
33
Jaiswal, Sushma, Harikumar Pallthadka, Rajesh P. Chinchewadi, and Tarun Jaiswal. "Optimized Image Captioning: Hybrid Transformers Vision Transformers and Convolutional Neural Networks: Enhanced with Beam Search." International Journal of Intelligent Systems and Applications 16, no. 2 (April 8, 2024): 53–61. http://dx.doi.org/10.5815/ijisa.2024.02.05.
Анотація:
Deep learning has improved image captioning. Transformer, a neural network architecture built for natural language processing, excels at image captioning and other computer vision applications. This paper reviews Transformer-based image captioning methods in detail. Convolutional neural networks (CNNs) extracted image features and RNNs or LSTM networks generated captions in traditional image captioning. This method often has information bottlenecks and trouble capturing long-range dependencies. Transformer architecture revolutionized natural language processing with its attention strategy and parallel processing. Researchers used Transformers' language success to solve image captioning problems. Transformer-based image captioning systems outperform previous methods in accuracy and efficiency by integrating visual and textual information into a single model. This paper discusses how the Transformer architecture's self-attention mechanisms and positional encodings are adapted for image captioning. Vision Transformers (ViTs) and CNN-Transformer hybrid models are discussed. We also discuss pre-training, fine-tuning, and reinforcement learning to improve caption quality. Transformer-based image captioning difficulties, trends, and future approaches are also examined. Multimodal fusion, visual-text alignment, and caption interpretability are challenges. We expect research to address these issues and apply Transformer-based image captioning to medical imaging and distant sensing. This paper covers how Transformer-based approaches have changed image captioning and their potential to revolutionize multimodal interpretation and generation, advancing artificial intelligence and human-computer interactions.
34
Braña, L., A. Costa, and R. Lopes. "Development of a power transformer model for high-frequency transient phenomena." Renewable Energy and Power Quality Journal 19 (September 2021): 217–21. http://dx.doi.org/10.24084/repqj19.260.
Анотація:
In recent years, the proliferation of distributed renewable energy sources and the application of new rules for the exploitation of electrical networks imposed by the markets have dictated increasingly demanding operating conditions for electric power transformers, creating new challenges in their exploration and conservation. Transformers that, in addition to the transmission lines, are certainly the most important and critical element of any electrical energy system. Adequate models are necessary to accurately describe transformer behavior and internal response when submitted to different external requests imposed by the network, particularly during transient phenomena, as well as, to properly assess system vulnerabilities and network optimization. This effort is being carried out today by several research groups in the world, namely from Cigré and IEEE. In this work, a transformer model to be integrated into a timedomain equivalent circuit is developed and discussed. Results obtained with this model are compared with measurements obtained by the Cigré JWG A2/C4.52 in a power transformer used as a reference for the working group.
35
T, Thoyyibah, Wasis Haryono, Achmad Udin Zailani, Yan Mitha Djaksana, Neny Rosmawarni, and Nunik Destria Arianti. "Transformers in Machine Learning: Literature Review." Jurnal Penelitian Pendidikan IPA 9, no. 9 (September 25, 2023): 604–10. http://dx.doi.org/10.29303/jppipa.v9i9.5040.
Анотація:
In this study, the researcher presents an approach regarding methods in Transformer Machine Learning. Initially, transformers are neural network architectures that are considered as inputs. Transformers are widely used in various studies with various objects. The transformer is one of the deep learning architectures that can be modified. Transformers are also mechanisms that study contextual relationships between words. Transformers are used for text compression in readings. Transformers are used to recognize chemical images with an accuracy rate of 96%. Transformers are used to detect a person's emotions. Transformer to detect emotions in social media conversations, for example, on Facebook with happy, sad, and angry categories. Figure 1 illustrates the encoder and decoder process through the input process and produces output. the purpose of this study is to only review literature from various journals that discuss transformers. This explanation is also done by presenting the subject or dataset, data analysis method, year, and accuracy achieved. By using the methods presented, researchers can conclude results in search of the highest accuracy and opportunities for further research.
36
Jurisic, Bruno, Marijan Perković, Ivan Novko, Luka Kovačić, Igor Žiger, and Tomislav Župan. "Proposal of Testing Procedure for Resonance and Ferroresonance Inception Possibility in Instrument Transformers." Journal of Energy - Energija 73, no. 2 (June 4, 2024): 21–24. http://dx.doi.org/10.37798/2024732520.
Анотація:
This paper deals with the possibility of ferroresonance occurence in the interaction between circuit breakers and inductive instrument transformers. Existing standards lack guidance on testing for ferroresonant behaviour. The paper proposes a standardized testing procedure and presents measurements on a full-scale system. EMTP simulations complement the measurements for a broader network topology analysis, i.e. circuit breaker capacitance combinations. EMTP simulations are validated for a 170 kV voltage transformer and a combined instrument transformer, showing accuracy within 10%. The paper also extends the EMTP modelling application to a 420 kV voltage power transformer during design phase, ensuring it doesn't experience ferroresonance. This study offers a practical approach for testing and simulating ferroresonance in inductive instrument transformers, contributing to the safe operation of power networks.
37
Hao, Zixin. "Comparative Analysis of Transformer Integration in U-net Networks for Enhanced Medical Image Segmentation." Highlights in Science, Engineering and Technology 94 (April 26, 2024): 333–40. http://dx.doi.org/10.54097/z4b39y45.
Анотація:
Transformer is popular in Natural Language Processing (NLP) and is a cornerstone of large models. Transformer has been used by researchers to address the limitations of Convolutional Neural Networks (CNNs) in medical picture segmentation models. Through an extensive literature review and case studies, this paper comparatively analyzes the performance of different models in this field, summarizes different methods of integrating transformers into U-net, and points out existing gaps and challenges. Research has found that the Transformer model can significantly improve the accuracy and efficiency of medical image analysis. The paper discusses the advantages, disadvantages, innovations, performance, and complexity of various models in detail, and shows how to enhance performance by integrating the Transformer structure into the U-net network. In particular, the paper also analyzes the advantages of Transformers that are most suitable for integration into the encoder part and highlights the balance that needs to be made between improving performance and computational cost. The conclusion shows that although there is no perfect model, optimal performance and efficiency can be achieved by selecting different combinations of Transformer and U-net according to the actual situation. It can be seen from the networks’ performance that the mixed use of a U-shaped convolutional network and Transformer module has good development prospects and high research significance.
38
Zhou, Li, Tongqin Shi, Songquan Huang, Fangchao Ke, Zhenxi Huang, Zhaoyang Zhang, and Jinzheng Liang. "Convolutional neural network for real-time main transformer detection." Journal of Physics: Conference Series 2229, no. 1 (March 1, 2022): 012021. http://dx.doi.org/10.1088/1742-6596/2229/1/012021.
Анотація:
Abstract For substation constructions, the main transformer is the dominant electrical equipment, and its arrival and operation affect the progress of project directly. In the context of smart grid construction, in order to improve the efficiency of real-time main transformer detection, this paper proposes an identification and detection method based on the SSD algorithm. The SSD algorithm is able to extract the target device (such as main transformer) accurately and the Lenet algorithm module can analyse the features contained in the image. To improve the accuracy of the detection method, the image migration algorithm of VGG-Net is used to expand the negative samples of main transformers to improve the generalisation of the algorithm. Finally, the image set collected in the real substation projects is used for validation, and result shows that the method identifies main transformers more accurately, with high effectiveness and feasibility.
39
Liu, Yijie. "High-frequency transformers optimized design for power electronic transformers." Applied and Computational Engineering 10, no. 1 (September 25, 2023): 196–202. http://dx.doi.org/10.54254/2755-2721/10/20230174.
Анотація:
Because of its small size, high frequency transformers are widely used to maximize energy transfer. However, the leakage inductance and distributed capacitance of high frequency transformer can not only cause resonance, but also lead to transient changes of voltage and current in high frequency, which can lead to voltage spike, so that the switch tube is damaged. For transformers with the same output power, high-frequency transformers are much smaller and have lower calorific value than low-frequency transformers. Therefore, at present, many consumer electronics and network product power adapters are switching power supplies, and the internal high-frequency transformer is the most important component of switching power supplies. The basic principle is to turn the input alternating current into DC first, and then turn it into high frequency through a transistor or FET, etc., through a high-frequency transformer to change voltage, and then rectify the output again, plus other control parts, and stabilize the output DC voltage. In this thesis, we choose a more rational and cost effective winding structure, choose a more appropriate core material based on the comparison of different core materials, research on the insulation and cooling properties of transformer so as to improve the insulation properties of the transformer, make it safer and more efficient. The study has important significance to decrease the power loss of high frequency transformer and decrease the size of high frequency transformer.
40
Duan, Shiyao, Jiaojiao Li, Rui Song, Yunsong Li, and Qian Du. "Unmixing-Guided Convolutional Transformer for Spectral Reconstruction." Remote Sensing 15, no. 10 (May 18, 2023): 2619. http://dx.doi.org/10.3390/rs15102619.
Анотація:
Deep learning networks based on CNNs or transformers have made progress in spectral reconstruction (SR). However, many methods focus solely on feature extraction, overlooking the interpretability of network design. Additionally, models exclusively based on CNNs or transformers may lose other prior information, sacrificing reconstruction accuracy and robustness. In this paper, we propose a novel Unmixing-Guided Convolutional Transformer Network (UGCT) for interpretable SR. Specifically, transformer and ResBlock components are embedded in Paralleled-Residual Multi-Head Self-Attention (PMSA) to facilitate fine feature extraction guided by the excellent priors of local and non-local information from CNNs and transformers. Furthermore, the Spectral–Spatial Aggregation Module (S2AM) combines the advantages of geometric invariance and global receptive fields to enhance the reconstruction performance. Finally, we exploit a hyperspectral unmixing (HU) mechanism-driven framework at the end of the model, incorporating detailed features from the spectral library using LMM and employing precise endmember features to achieve a more refined interpretation of mixed pixels in HSI at sub-pixel scales. Experimental results demonstrate the superiority of our proposed UGCT, especially in the grss_d f c_2018 dataset, in which UGCT attains an RMSE of 0.0866, outperforming other comparative methods.
41
Klochikhin, Evgeny A., and Yurii P. Neugodnikov. "Evaluation of Technical and Economical Eff ectiveness of the Use of New Transformers for Railways Traction Substations." Innotrans, no. 4 (2023): 62–66. http://dx.doi.org/10.20291/2311-164x-2023-4-62-66.
Анотація:
Evaluation of technical and economical effectiveness of the use of new power transformers for railways traction substations is considered. Comparative analyses of transformers cost, no-load losses and short circuit expenses is conducted. Also reduction of losses in the traction network under non-traction load is evaluated. The use of new TDNZhD and TRDNZhD transformers leads to a great total economic effect in comparison to the standard TDTNZh transformer.
42
Odinaev, Ismoil, Andrey Pazderin, Murodbek Safaraliev, Firuz Kamalov, Mihail Senyuk, and Pavel Y. Gubin. "Detection of Current Transformer Saturation Based on Machine Learning." Mathematics 12, no. 3 (January 25, 2024): 389. http://dx.doi.org/10.3390/math12030389.
Анотація:
One of the tasks in the operation of electric power systems is the correct functioning of the protection system and emergency automation algorithms. Instrument voltage and current transformers, operating in accordance with the laws of electromagnetism, are most often used for information support of the protection system and emergency automation algorithms. Magnetic core saturation of the specified current transformers can occur during faults. As a result, the correct functioning of the protection system and emergency automation algorithms is compromised. The consequences of current transformers saturation are mostly reflected in the main protections of network elements operating on a differential principle. This work aims to consider the analysis of current transformer saturation detection methods. The problem of identifying current transformer saturation is reduced to binary classification, and methods for solving the problem based on artificial neural networks, support vector machine, and decision tree algorithms are proposed. Computational experiments were performed, and their results were analyzed with imbalanced (dominance of the number of current transformer saturation modes over the number of modes with its normal operation) and balanced classes 0 (no current transformer saturation) and 1 (current transformer saturation).
43
Majeed, Issah Babatunde, and Nnamdi I. Nwulu. "Impact of Reverse Power Flow on Distributed Transformers in a Solar-Photovoltaic-Integrated Low-Voltage Network." Energies 15, no. 23 (December 6, 2022): 9238. http://dx.doi.org/10.3390/en15239238.
Анотація:
Modern low-voltage distribution systems necessitate solar photovoltaic (PV) penetration. One of the primary concerns with this grid-connected PV system is overloading due to reverse power flow, which degrades the life of distribution transformers. This study investigates transformer overload issues due to reverse power flow in a low-voltage network with high PV penetration. A simulation model of a real urban electricity company in Ghana is investigated against various PV penetration levels by load flows with ETAP software. The impact of reverse power flow on the radial network transformer loadings is examined for high PV penetrations. Using the least squares method, simulation results are modelled in Excel software. Transformer backflow limitations are determined by correlating operating loads with PV penetration. At high PV penetration, the models predict reverse power flow into the transformer. Interpolations from the correlation models show transformer backflow operating limits of 78.04 kVA and 24.77% at the threshold of reverse power flow. These limits correspond to a maximum PV penetration limit of 88.30%. In low-voltage networks with high PV penetration; therefore, planners should consider transformer overload limits caused by reverse power flow, which degrades transformer life. This helps select control schemes near substation transformers to limit reverse power flow.
44
Wei, Jiangshu, Jinrong Chen, Yuchao Wang, Hao Luo, and Wujie Li. "Improved deep learning image classification algorithm based on Swin Transformer V2." PeerJ Computer Science 9 (October 30, 2023): e1665. http://dx.doi.org/10.7717/peerj-cs.1665.
Анотація:
While convolutional operation effectively extracts local features, their limited receptive fields make it challenging to capture global dependencies. Transformer, on the other hand, excels at global modeling and effectively captures global dependencies. However, the self-attention mechanism used in Transformers lacks a local mechanism for information exchange within specific regions. This article attempts to leverage the strengths of both Transformers and convolutional neural networks (CNNs) to enhance the Swin Transformer V2 model. By incorporating both convolutional operation and self-attention mechanism, the enhanced model combines the local information-capturing capability of CNNs and the long-range dependency-capturing ability of Transformers. The improved model enhances the extraction of local information through the introduction of the Swin Transformer Stem, inverted residual feed-forward network, and Dual-Branch Downsampling structure. Subsequently, it models global dependencies using the improved self-attention mechanism. Additionally, downsampling is applied to the attention mechanism’s Q and K to reduce computational and memory overhead. Under identical training conditions, the proposed method significantly improves classification accuracy on multiple image classification datasets, showcasing more robust generalization capabilities.
45
Liu, Jie, Songren Mao, and Liangrui Pan. "Attention-Based Two-Branch Hybrid Fusion Network for Medical Image Segmentation." Applied Sciences 14, no. 10 (May 10, 2024): 4073. http://dx.doi.org/10.3390/app14104073.
Анотація:
Accurate segmentation of medical images is vital for disease detection and treatment. Convolutional Neural Networks (CNN) and Transformer models are widely used in medical image segmentation due to their exceptional capabilities in image recognition and segmentation. However, CNNs often lack an understanding of the global context and may lose spatial details of the target, while Transformers struggle with local information processing, leading to reduced geometric detail of the target. To address these issues, this research presents a Global-Local Fusion network model (GLFUnet) based on the U-Net framework and attention mechanisms. The model employs a dual-branch network that utilizes ConvNeXt and Swin Transformer to simultaneously extract multi-level features from pathological images. It enhances ConvNeXt’s local feature extraction with spatial and global attention up-sampling modules, while improving Swin Transformer’s global context dependency with channel attention. The Attention Feature Fusion module and skip connections efficiently merge local detailed and global coarse features from CNN and Transformer branches at various scales. The fused features are then progressively restored to the original image resolution for pixel-level prediction. Comprehensive experiments on datasets of stomach and liver cancer demonstrate GLFUnet’s superior performance and adaptability in medical image segmentation, holding promise for clinical analysis and disease diagnosis.
46
Zheng, Yuping, and Weiwei Jiang. "Evaluation of Vision Transformers for Traffic Sign Classification." Wireless Communications and Mobile Computing 2022 (June 4, 2022): 1–14. http://dx.doi.org/10.1155/2022/3041117.
Анотація:
Traffic sign recognition is one of the most important tasks in autonomous driving. Camera-based computer vision techniques have been proposed for this task, and various convolutional neural network structures are used and validated with multiple open datasets. Recently, novel Transformer-based models have been proposed for various computer vision tasks and have achieved state-of-the-art performance, outperforming convolutional neural networks in several tasks. In this study, our goal is to investigate whether the success of Vision Transformers can be replicated within the traffic sign recognition area. Based on existing resources, we first extract and contribute three open traffic sign classification datasets. Based on these datasets, we experiment with seven convolutional neural networks and five Vision Transformers. We find that Transformers are not as competitive as convolutional neural networks for the traffic sign classification task. Specifically, there are performance gaps of up to 12.81%, 2.01%, and 4.37% existing for the German, Indian, and Chinese traffic sign datasets, respectively. Furthermore, we propose some suggestions to improve the performance of Transformers.
47
Nozdrenkov, V. S., I. M. Diahovchenko, M. V. Petrovskyi, and V. V. Volokhin. "Fuzzy model of compensation for aging factors of distribution transformers." Electrical Engineering and Power Engineering, no. 2 (June 27, 2024): 7–17. http://dx.doi.org/10.15588/1607-6761-2024-2-1.
Анотація:
Purpose. This paper aims to evaluate the negative factors that affect the aging of power distribution transformers, develop a fuzzy control model for their compensation, and study the results of applying the proposed model to different modes of the electrical power network. Methodology. The mathematical method of fuzzy logic was used to implement the control system of the power grid operating modes. Findings. The article presents a structure based on fuzzy logic for compensating depreciation factors of distribution transformers. A tuning algorithm and measures were developed to optimize the transformer's load level and power factor. The developed model analyzes the parameters and factors affecting the normal operation of the transformer and warns of dangerous factors that threaten reliability and may lead to a malfunction. In addition, the efficiency of PV generating stations, shunt capacitor banks, and energy storage systems installed on the secondary voltage side to preserve the service life of distribution transformers was analyzed and discussed. Originality. The paper further develops the fuzzy logic models used to optimize the operation of the power grid and compensate for the aging factors of power distribution transformers Practical value. The results obtained in the paper can be used to build an optimal system for controlling the operation modes of the electric power grid, which reduces the factors that accelerate the aging of power distribution transformers.
48
Alyunov, A. N., O. S. Vyatkina, I. G. Akhmetova, R. D. Pentiuc, and K. E. Sakipov. "Issues on optimization of operating modes of power transformers." E3S Web of Conferences 124 (2019): 02015. http://dx.doi.org/10.1051/e3sconf/201912402015.
Анотація:
The article presents measure to optimize the operating modes of power transformers in order to minimize losses of electrical energy. The influence of actual voltage and service life of power transformers on electric power losses is shown. It was proposed to determine the economic capacity of power transformers taking into account the indicated factors, as well as taking into account the time of transformer switching on into the electric network and the form of the load schedule.
49
Kaur, Amrinder, Yadwinder Singh Brar, and Leena G. "Fault detection in power transformers using random neural networks." International Journal of Electrical and Computer Engineering (IJECE) 9, no. 1 (February 1, 2019): 78. http://dx.doi.org/10.11591/ijece.v9i1.pp78-84.
Анотація:
This paper discuss the application of artificial neural network-based algorithms to identify different types of faults in a power transformer, particularly using DGA (Dissolved Gas Analysis) test. The analysis of Random Neural Network (RNN) using Levenberg-Marquardt (LM) and Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithms has been done using the data of dissolved gases of power transformers collected from Punjab State Transmission Corporation Ltd.(PSTCL), Ludhiana, India. Sorting of the preprocessed data have been done using dimensionality reduction technique, i.e., principal component analysis. The sorted data is used as inputs to the Random Neural Networks (RNN) classifier. It has been seen from the results obtained that BFGS has better performance for the diagnosis of fault in transformer as compared to LM.
50
Lesniewska, Elzbieta. "Influence of the Selection of the Core Shape and Winding Arrangement on the Accuracy of Current Transformers with Through-Going Primary Cable." Energies 14, no. 7 (March 31, 2021): 1932. http://dx.doi.org/10.3390/en14071932.
Анотація:
The current transformers with split-core are used for installation in places where it is impossible to install classic current transformers. Moreover, this design allows for any measurement location change, and even switching one current transformer into several different shapes of bars or cables. Power network operators, striving for more accurate current measurements, require producers to provide current transformers with a special accuracy class 0.2S. Therefore, manufacturers and designers try to meet the market requirements and, similarly to non-demountable current transformers, i.e., with a toroidal core, design current transformers with split-core class 0.2S. To meet the high metrological requirements, 3D analyses of electromagnetic fields were performed, taking into account physical phenomena and not approximate analytical models. Two types of cores and four different arrangements of the secondary windings of the measuring current transformers were considered. The magnetic field distributions, current error, and phase displacement diagrams of all current transformer models were analyzed, and the model of the transformer structure with the best accuracy was selected. Computations were conducted based on the finite element numerical method, and the results were compared with the real model tests.