Search results for: multimodal transformer autoencoder

Authors: Inas Hussein

Abstract:

Communication in the modern world is increasingly becoming multimodal due to globalization and the digital space we live in which have remarkably affected how people communicate. Accordingly, Multimodal Discourse Analysis (MDA) is an emerging paradigm in discourse studies with the underlying assumption that other semiotic resources such as images, colours, scientific symbolism, gestures, actions, music and sound, etc. combine with language in order to  communicate meaning. One of the effective multimodal media that combines both verbal and non-verbal elements to create meaning is political cartoons. Furthermore, since political and social issues are mirrored in political cartoons, these are regarded as potential objects of discourse analysis since they not only reflect the thoughts of the public but they also have the power to influence them. The aim of this paper is to analyze some selected cartoons on the recognition of Jerusalem as Israel's capital by the American President, Donald Trump, adopting a multimodal approach. More specifically, the present research examines how the various semiotic tools and resources utilized by the cartoonists function in projecting the intended meaning. Ten political cartoons, among a surge of editorial cartoons highlighted by the Anti-Defamation League (ADL) - an international Jewish non-governmental organization based in the United States - as publications in different Arabic-language newspapers in Egypt, Saudi Arabia, UAE, Oman, Iran and UK, were purposively selected for semiotic analysis. These editorial cartoons, all published during 6^th–18^th December 2017, invariably suggest one theme: Jewish and Israeli domination of the United States. The data were analyzed using the framework of Visual Social Semiotics. In accordance with this methodological framework, the selected visual compositions were analyzed in terms of three aspects of meaning: representational, interactive and compositional. In analyzing the selected cartoons, an interpretative approach is being adopted. This approach prioritizes depth to breadth and enables insightful analyses of the chosen cartoons. The findings of the study reveal that semiotic resources are key elements of political cartoons due to the inherent political communication they convey. It is proved that adequate interpretation of the three aspects of meaning is a prerequisite for understanding the intended meaning of political cartoons. It is recommended that further research should be conducted to provide more insightful analyses of political cartoons from a multimodal perspective.

Keywords: Multimodal Discourse Analysis (MDA), multimodal text, political cartoons, visual modality

Procedia PDF Downloads 204

Authors: Gia Surguladze, Nino Topuria, Lily Petriashvili, Giorgi Surguladze

Abstract:

Modeling of building processes of a multimodal freight transportation support information system is discussed based on modern CASE technologies. Functional efficiencies of ports in the eastern part of the Black Sea are analyzed taking into account their ecological, seasonal, resource usage parameters. By resources, we mean capacities of berths, cranes, automotive transport, as well as work crews and neighbouring airports. For the purpose of designing database of computer support system for Managerial (Logistics) function, using Object-Role Modeling (ORM) tool (NORMA – Natural ORM Architecture) is proposed, after which Entity Relationship Model (ERM) is generated in automated process. The software is developed based on Process-Oriented and Service-Oriented architecture, in Visual Studio.NET environment.

Keywords: seaport resources, business-processes, multimodal transportation, CASE technology, object-role model, entity relationship model, SOA

Procedia PDF Downloads 399

Authors: Prateek Saxena, Sanjeev Singh, Julius Roy

Abstract:

The security of the power distribution grid remains a paramount to the utility professionals while enhancing and making it more efficient. The most serious threat to the system can be maintaining the transformers, as the load is ever increasing with the addition of elements like electric vehicles. In this paper, intelligent transformer monitoring and grid management has been proposed. The engineering is done to use the evolving data from the smart meter for grid analytics and diagnostics for preventive maintenance. The two-tier architecture for hardware and software integration is coupled to form a robust system for the smart grid. The proposal also presents interoperable meter standards for easy integration. Distribution transformer analytics based on real-time data benefits utilities preventing outages, protects the revenue loss, improves the return on asset and reduces overall maintenance cost by predictive monitoring.

Keywords: electric vehicle charging, transformer monitoring, data analytics, intelligent grid

Procedia PDF Downloads 757

Authors: K. N. Dinesh Babu, M. V. Gopalan, G. R. Manjunatha, R. Ramaprabha, V. Rajini

Abstract:

In this paper, monitoring and control of tap changer mechanism of a transformer implementation in an intelligent electronic device (IED) is discussed. Its been a custom for decades to provide a separate panel for on load tap changer control for monitoring the tap position. However this facility cannot either record or transfer the information to remote control centers. As there is a technology shift towards the smart grid protection and control standards, the need for implementing remote control and monitoring has necessitated the implementation of this feature in numerical relays. This paper deals with the programming, settings and logic implementation which is applicable to both IEC 61850 compatible and non-compatible IEDs thereby eliminating the need for separate tap changer control equipment. The monitoring mechanism has been implemented in a 28MVA, 110 /6.9kV transformer with 16 tap position with GE make T60 IED at Ultratech cement limited Gulbarga, Karnataka and is in successful service.

Keywords: transformer protection, tap changer control, tap position monitoring, on load tap changer, intelligent electronic device (IED)

Procedia PDF Downloads 565

Authors: T. Bryan , V. Kepuska, I. Kostnaic

Abstract:

A method of data compression and denoising is introduced that is based on atomic decomposition of audio data using “basis vectors” that are learned from the audio data itself. The basis vectors are shown to have higher data compression and better signal-to-noise enhancement than the Gabor and gammatone “seed atoms” that were used to generate them. The basis vectors are the input weights of a Sparse AutoEncoder (SAE) that is trained using “envelope samples” of windowed segments of the audio data. The envelope samples are extracted from the audio data by performing atomic decomposition with Gabor or gammatone seed atoms. This process identifies segments of audio data that are locally coherent with the seed atoms. Envelope samples are extracted by identifying locally coherent audio data segments with Gabor or gammatone seed atoms, found by matching pursuit. The envelope samples are formed by taking the kronecker products of the atomic envelopes with the locally coherent data segments. Oracle signal-to-noise ratio (SNR) verses data compression curves are generated for the seed atoms as well as the basis vectors learned from Gabor and gammatone seed atoms. SNR data compression curves are generated for speech signals as well as early American music recordings. The basis vectors are shown to have higher denoising capability for data compression rates ranging from 90% to 99.84% for speech as well as music. Envelope samples are displayed as images by folding the time series into column vectors. This display method is used to compare of the output of the SAE with the envelope samples that produced them. The basis vectors are also displayed as images. Sparsity is shown to play an important role in producing the highest denoising basis vectors.

Keywords: sparse dictionary learning, autoencoder, sparse autoencoder, basis vectors, atomic decomposition, envelope sampling, envelope samples, Gabor, gammatone, matching pursuit

Procedia PDF Downloads 227

Authors: Thummalapalli CSM Gupta, Banti Sidhiwala

Abstract:

Mineral oils of naphthenic and paraffinic type have been traditionally been used as insulating liquids in the transformer applications to protect the solid insulation from moisture and ensures effective heat transfer/cooling. The performance of these type of oils have been proven in the field over many decades and the condition monitoring and diagnosis of transformer performance have been successfully monitored through oil properties and dissolved gas analysis methods successfully. Different type of gases representing various types of faults due to components or operating conditions effectively. While large amount of data base has been generated in the industry on dissolved gas analysis for mineral oil based transformer oils and various models for predicting the fault and analysis, oil specifications and standards have also been modified to include stray gassing limits which cover the low temperature faults and becomes an effective preventative maintenance tool that can benefit greatly to know the reasons for the breakdown of electrical insulating materials and related components. Natural esters have seen a rise in popularity in recent years due to their "green" credentials. Some of its benefits include biodegradability, a higher fire point, improvement in load capability of transformer and improved solid insulation life than mineral oils. However, the Stray gases evolution like hydrogen and hydrocarbons like methane (CH4) and ethane (C2H6) show very high values which are much higher than the limits of mineral oil standards. Though the standards for these type esters are yet to be evolved, the higher values of hydrocarbon gases that are available in the market is of concern which might be interpreted as a fault in transformer operation. The current paper focuses on developing a natural ester based transformer oil which shows very levels of stray gassing by standard test methods show much lower values compared to the products available currently and experimental results on various test conditions and the underlying mechanism explained.

Keywords: biodegadability, fire point, dissolved gassing analysis, stray gassing

Procedia PDF Downloads 65

Authors: Bianca Suárez Puerta

Abstract:

Multimodal discourse refers to a method of study the media continuum between reality, screens as a device, audience, author, and media as a production from the audience. For this study we used semantic differential, a method proposed in the sixties by Osgood, Suci and Tannenbaum, starts from the assumption that under each particular way of perceiving the world, in each singular idea, there is a common cultural meaning that organizes experiences. In relation to these shared symbolic dimension, this method has had significant results, as it focuses on breaking down the meaning of certain significant acts into series of statements that place the subjects in front of some concepts. In Colombia, in 2016, a tool was designed to measure the meaning of a multimodal production, specially the acts of sense of transmedia productions that managed to receive funds from the Ministry of ICT of Colombia, and also, to analyze predictable patterns that can be found in calls and funds aimed at the production of culture in Colombia, in the context of the peace agreement, as a request for expressions from a hegemonic place, seeking to impose a worldview.

Keywords: semantic differential, semiotics, transmedia, critical analysis of discourse

Procedia PDF Downloads 180

Authors: Anthony Okechukwu Ifediniru, Austin Ikechukwu Gbasouzor, Isidore Uche Uju

Abstract:

This article is centered on the design, analysis, construction, and test of a locally made arc welding machine that operates on 250vac with 8 amp output taps ranging from 60vac to 250vac at a fixed frequency, which is of benefit to urban areas; while considering its cost-effectiveness, strength, portability, and mobility. The welding machine uses a power supply to create an electric arc between an electrode and the metal at the welding point. A current selector coil needed for current selection is connected to the primary winding. Electric power is supplied to the primary winding of its transformer and is transferred to the secondary winding by induction. The voltage and current output of the secondary winding are connected to the output terminal, which is used to carry out welding work. The output current of the machine ranges from 110amps for low current welding to 250amps for high current welding. The machine uses a step-down transformer configuration for stepping down the voltage in order to obtain a high current level for effective welding. The welder can adjust the output current within a certain range. This allows the welder to properly set the output current for the type of welding that is being performed. The constructed arc welding machine was tested by connecting the work piece to it. Since there was no shock or spark from the transformer’s laminated core and was successfully used to join metals, it confirmed and validated the design.

Keywords: AC current, arc welding machine, DC current, transformer, welds

Procedia PDF Downloads 153

Authors: Pranjal Johri, Misbah Ul-Islam

Abstract:

Power Transformers are one of the most critical and failure prone entities in an electrical power system. It is an established practice that each design of a power transformer has to undergo numerous type tests for design validation and routine tests are performed on each and every power transformer before dispatch from manufacturer’s works. Different countries follow different standards for testing the transformers. Most common and widely followed standard for Power Transformers is IEC 60076 series. Though these standards put up a strict testing requirements for power transformers, however, few aspects of transformer characteristics and guaranteed parameters can be ensured by some additional tests. Based on certain observations during routine test of a transformer and analyzing the data of a large fleet of transformers, three propositions have been discussed and put forward to be included in test schedules and standards. The observations in the routine test raised questions on design flux density of transformer. In order to ensure that flux density in any part of the core & yoke does not exceed 1.9 tesla at 1.1 pu as well, following propositions need to be followed during testing:  From the data studied, it was evident that generally NLC at 1.1 pu is apporx. 3 times of No Load Current at 1 pu voltage.  During testing the power factor at 1.1 pu excitation, it must be comparable to calculated values from the Cold Rolled Grain Oriented steel material curves, including building factor.  A limit of 3 % to be extended for higher than rated voltages on difference in Vavg and Vrms, during no load testing.  Extended over excitation test to be done in case above propositions are observed to be violated during testing.

Keywords: power transfoemrs, no load current, DGA, power factor

Procedia PDF Downloads 66

Authors: Prashanth Pillai, Purnaprajna Mangsuli

Abstract:

Oil exploration and production companies invest considerable time and efforts to extract essential well attributes (like well status, surface, and target coordinates, wellbore depths, event timelines, etc.) from unstructured data sources like technical reports, which are often non-standardized, multimodal, and highly domain-specific by nature. It is also important to consider the context when extracting attribute values from reports that contain information on multiple wells/wellbores. Moreover, semantically similar information may often be depicted in different data syntax representations across multiple pages and document sources. We propose a hierarchical multi-source fact extraction workflow based on a deep learning framework to extract essential well attributes at scale. An information retrieval module based on the transformer architecture was used to rank relevant pages in a document source utilizing the page image embeddings and semantic text embeddings. A question answering framework utilizingLayoutLM transformer was used to extract attribute-value pairs incorporating the text semantics and layout information from top relevant pages in a document. To better handle context while dealing with multi-well reports, we incorporate a dynamic query generation module to resolve ambiguities. The extracted attribute information from various pages and documents are standardized to a common representation using a parser module to facilitate information comparison and aggregation. Finally, we use a probabilistic approach to fuse information extracted from multiple sources into a coherent well record. The applicability of the proposed approach and related performance was studied on several real-life well technical reports.

Keywords: natural language processing, deep learning, transformers, information retrieval

Procedia PDF Downloads 167

Authors: Sharifa Alshahrani

Abstract:

Comics is a multimodal text wherein meaning is created by taking in all modes of expression at once. It uses two different semiotic modes, the verbal and the visual modes, together to make meaning and these different semiotic modes can be socially and culturally shaped to give meaning. Therefore, comics translation cannot treat comics as a monomodal text by translating only the verbal mode inside or outside the speech balloons as the cultural differences are encoded in the visual mode as well. Due to the development of the internet and editing software, comics translation is not anymore confined to the publishing houses and official translation as scanlation, or the fan translation took the initiative in translating comics for being emotionally attracted to the culture and genre. Scanlation is carried out by volunteering fans who translate out of passion. However, quality is one of the debatable issues relating to scanlation and fan translation. This study will investigate how the dynamic multimodal relationship in comics is exploited and interpreted in the translation by exploring the translation strategies and procedures adopted by the publishing houses and scanlation in interpreting comics into Arabic using three analytical frameworks; cultural references model, multimodal relation model and translation strategies and procedures models.

Keywords: comics, multimodality, translation, scanlation

Procedia PDF Downloads 186

Authors: Mariam Waheed Mekheimar

Abstract:

Nascent research is conducted to the advancement of discourse analysis to include different modes as images, sound, and text. The focus of this study will be to elucidate how images are embedded with texts in an audio-visual medium as cinema to send political messages; it also seeks to broaden our understanding of politics beyond a relatively narrow conceptualization of the 'political' through studying non-traditional discourses as the cinematic discourse. The aim herein is to develop a systematic approach to film analysis to capture political meanings in films. The method adopted in this research is Multimodal Discourse Analysis (MDA) focusing on embedding visuals with texts. As today's era is the era of images and that necessitates analyzing images. Drawing on the writings of O'Halloran, Kress and Van Leuween, John Bateman and Janina Wildfeuer, different modalities will be studied to understand how those modes interact in the cinematic discourse. 'People at the top movie' is selected as an example to unravel the political meanings throughout film tackling the cinematic representation of the notion of social justice.

Keywords: Egyptian cinema, multimodal discourse analysis, people at the top, social justice

Procedia PDF Downloads 386

Authors: Nishtha Madaan, Prateek Chaudhury, Nishant Kumar, Srikanta Bedathur

Abstract:

In modern NLP applications, word embeddings are a crucial backbone that can be readily shared across a number of tasks. However, as the text distributions change and word semantics evolve over time, the downstream applications using the embeddings can suffer if the word representations do not conform to the data drift. Thus, maintaining word embeddings to be consistent with the underlying data distribution is a key problem. In this work, we tackle this problem and propose TransDrift, a transformer-based prediction model for word embeddings. Leveraging the flexibility of the transformer, our model accurately learns the dynamics of the embedding drift and predicts future embedding. In experiments, we compare with existing methods and show that our model makes significantly more accurate predictions of the word embedding than the baselines. Crucially, by applying the predicted embeddings as a backbone for downstream classification tasks, we show that our embeddings lead to superior performance compared to the previous methods.

Keywords: NLP applications, transformers, Word2vec, drift, word embeddings

Procedia PDF Downloads 58

Authors: Yagya Raj Pandeya, Joonwhoan Lee

Abstract:

The dynamic behavior of music and video makes it difficult to evaluate musical instrument playing in a video by computer system. Any television or film video clip with music information are rich sources for analyzing musical instruments using modern machine learning technologies. In this research, we integrate the audio and video information sources using convolutional neural network (CNN) and pass network learned features through recurrent neural network (RNN) to preserve the dynamic behaviors of audio and video. We use different pre-trained CNN for music and video feature extraction and then fine tune each model. The music network use 2D convolutional network and video network use 3D convolution (C3D). Finally, we concatenate each music and video feature by preserving the time varying features. The long short term memory (LSTM) network is used for long-term dynamic feature characterization and then use late fusion with generalized mean. The proposed network performs better performance to recognize the musical instrument using audio-video multimodal neural network.

Keywords: multimodal, 3D convolution, music-video feature extraction, generalized mean

Procedia PDF Downloads 188

Authors: D. Subedi, D. Sharma

Abstract:

Current transformers (CTs) are used to transform large primary currents to a small secondary current. Since most standard equipment’s are not designed to handle large primary currents the CTs have an important part in any electrical system for the purpose of Metering and Protection both of which are integral in Power system. Now a days due to advancement in solid state technology, the operation times of the protective relays have come to a few cycles from few seconds. Thus, in such a scenario it becomes important to study the transient response of the current transformers as it will play a vital role in the operating of the protective devices. This paper shows the steady state and transient behavior of current transformers and how it changes with change in connected burden. The transient and steady state response will be captured using the data acquisition software LabVIEW. Analysis is done on the real time data gathered using LabVIEW. Variation of current transformer characteristics with changes in burden will be discussed.

Keywords: accuracy, accuracy limiting factor, burden, current transformer, instrument security factor

Procedia PDF Downloads 317

Authors: Bjorn-Ole Meyer, Dominik Marti, Peter E. Andersen

Abstract:

A multimodal imaging system, combining spectrally resolved multiphoton microscopy (MPM) and optical coherence microscopy (OCM) is demonstrated. MPM and OCM are commonly integrated into multimodal imaging platforms to combine functional and morphological information. The MPM signals, such as two-photon fluorescence emission (TPFE) and signals created by second harmonic generation (SHG) are biomarkers which exhibit information on functional biological features such as the ratio of pyridine nucleotide (NAD(P)H) and flavin adenine dinucleotide (FAD) in the classification of cancerous tissue. While the spectrally resolved imaging allows for the study of biomarkers, using a spectrometer as a detector limits the imaging speed of the system significantly. To overcome those limitations, an OCM setup was added to the system, which allows for fast acquisition of structural information. Thus, after rapid imaging of larger specimens, navigation within the sample is possible. Subsequently, distinct features can be selected for further investigation using MPM. Additionally, by probing a different contrast, complementary information is obtained, and different biomarkers can be investigated. OCM images of tissue and cell samples are obtained, and distinctive features are evaluated using MPM to illustrate the benefits of the system.

Keywords: optical coherence microscopy, multiphoton microscopy, multimodal imaging, two-photon fluorescence emission

Procedia PDF Downloads 483

Authors: Moustafa Ibrahim Moustafa

Abstract:

The purpose of this study was to determine the immediate and long term effects of a multimodal program, with the addition of cervical sagittal curve restoration and forward head correction, on severity of dizziness, disability, frequency of dizziness, and severity of cervical pain. 72 patients with cervicogenic dizziness, definite hypolordotic cervical spine, and forward head posture were randomized to experimental or a control group. Both groups received the multimodal program, additionally, the study group received the Denneroll cervical traction. All outcome measures were measured at three intervals. The general linear model indicated a significant group × time effects in favor of experimental group on measures of anterior head translation (F=329.4 P < .0005), cervical lordosis (F=293.7 P < .0005), severity of dizziness (F=262.1 P < .0005), disability (F=248.9 P < .0005), frequency of dizziness (F=53.9 P < .0005), and severity of cervical pain (F=350.1 P < .0005). The addition of Dennroll cervical traction to a multimodal program can positively affect dizziness management outcomes.

Keywords: randomized controlled trial, traction, dizziness, cervical

Procedia PDF Downloads 279

Authors: O. P. Rahi, Manoj Kumar

Abstract:

Power transformers are essential and expensive equipments in electrical power system. Dissolved gas analysis (DGA) is one of the most useful techniques to detect incipient faults in power transformers. However, the identification of the faulted location by conventional method is not always an easy task due to variability of gas data and operational variables. In this paper, an extension theory based power transformer fault diagnosis method is presented. Extension theory tries to solve contradictions and incompatibility problems. This paper first briefly introduces the basic concept of matter element theory, establishes the matter element models for three-ratio method, and then briefly discusses extension set theory. Detailed analysis is carried out on the extended relation function (ERF) adopted in this paper for transformer fault diagnosis. The detailed diagnosing steps are offered. Simulation proves that the proposed method can overcome the drawbacks of the conventional three-ratio method, such as no matching and failure to diagnose multi-fault. It enhances diagnosing accuracy.

Keywords: DGA, extension theory, ERF, fault diagnosis power transformers, fault diagnosis, fuzzy logic

Procedia PDF Downloads 377

Authors: Loreine Makki, Marc Anthony Mannah, Christophe Batard, Nicolas Ginot, Julien Weckbrodt

Abstract:

Planar transformers are extensively utilized in high-frequency, high power density power electronic converters. The breakthrough of wide-bandgap technology compelled power electronic system miniaturization while inducing pivotal effects on system modeling and manufacturing within the power electronics industry. A significant consideration to simulate and model the unanticipated parasitic parameters emerges with the requirement to mitigate electromagnetic disturbances. This paper will present an equivalent circuit model of a shielded pulse planar transformer quantifying leakage inductance and resistance in addition to the interwinding capacitance of the primary and secondary windings. ANSYS Q3D Extractor was utilized to model and simulate the transformer, intending to study the immunity of the simulated equivalent model to high dv/dt occurrences. A convenient correlation between simulation and experimental results is presented.

Keywords: Planar transformers, wide-band gap, equivalent circuit model, shielded, ANSYS Q3D Extractor, dv/dt

Procedia PDF Downloads 178

Authors: Raheel Shakoor Siddiqui, Shahbaz Malik, Manikandar Srinivas Cheruvu, Sanjay Narayana Murthy, Livio DiMascio

Abstract:

Background: elderly traumatic hip fracture patients frequently present to trauma services globally. Rising low energy falls amongst an osteoporotic aging population is the commonest cause for injury. Hip fractures in this population are a major cause for severe pain, morbidity and mortality. The term hip fracture is interchangeable with neck of femur fracture, fractured neck of femur or proximal femur fracture. Hip fracture pain management protocols and guidelines suggest conventional analgesia, nerve block and opioid based treatment as rescue analgesia. There is a current global opioid crisis with overuse, abuse and dependence. Adverse opioid related complications in vulnerable elderly patients further adds to morbidity and mortality. Systematic reviews in literature have evidenced superiority of multimodal analgesia in osteoarthritic primary joint replacements compared to opioids however, this has not yet been conducted for elderly traumatic hip fracture patients. Aims: The primary aim of this systematic review is to provide standardised evidence following Cochrane and PRISMA guidance in determining advantages of perioperative multimodal analgesia over conventional opioid based treatments in elderly traumatic hip fractures. Methods: 5 databases were searched from January 2000-2023 which identified 8 randomised controlled trials and 446 total participants. These trials met defined PICOS eligibility criteria of patient mean age ≥ 65 years presenting with a unilateral traumatic fractured neck of femur for operative intervention. Analgesic intervention with perioperative multimodal analgesia has been compared to conventional opioid based analgesia. Outcomes of interest include, primarily, the change in postoperative opioid consumption within a 0-30 postoperative period and secondarily, the change in postoperative adverse events and complications. A qualitative synthesis has been performed due to clinical heterogenicity and variance amongst trials. Results: GRADE evidence of moderate quality supports perioperative multimodal analgesia leads to a reduction in postoperative opioid consumption however, low quality evidence supports a reduction of adverse effects and complications. Conclusion: Perioperative multimodal analgesia whether used preoperative, intraoperative and/or postoperative leads to a reduction in postoperative opioid consumption for elderly traumatic hip fracture patients. This review recommends the use of perioperative multimodal analgesia as part of hip fracture pain protocols however, caution and clinical judgement should be used as the risk of adverse effects may not be lower.

Keywords: trauma, orthopaedics, hip, fracture, neck of femur fracture, analgesia, multimodal analgesia, opioid

Procedia PDF Downloads 68

Authors: Pitigalage Chamath Chandira Peiris

Abstract:

A widely researched domain in the field of image processing in recent times has been single image super-resolution, which tries to restore a high-resolution image from a single low-resolution image. Many more single image super-resolution efforts have been completed utilizing equally traditional and deep learning methodologies, as well as a variety of other methodologies. Deep learning-based super-resolution methods, in particular, have received significant interest. As of now, the most advanced image restoration approaches are based on convolutional neural networks; nevertheless, only a few efforts have been performed using Transformers, which have demonstrated excellent performance on high-level vision tasks. The effectiveness of CNN-based algorithms in image super-resolution has been impressive. However, these methods cannot completely capture the non-local features of the data. Enhancer is a simple yet powerful Transformer-based approach for enhancing the resolution of images. A method for single image super-resolution was developed in this study, which utilized an efficient and effective transformer design. This proposed architecture makes use of a locally enhanced window transformer block to alleviate the enormous computational load associated with non-overlapping window-based self-attention. Additionally, it incorporates depth-wise convolution in the feed-forward network to enhance its ability to capture local context. This study is assessed by comparing the results obtained for popular datasets to those obtained by other techniques in the domain.

Keywords: single image super resolution, computer vision, vision transformers, image restoration

Procedia PDF Downloads 75

Authors: Esther Chiramal, Kelvin Soh Boon Kai

Abstract:

Explainable AI is a strong strategy implemented to understand complex black-box model predictions in a human-interpretable language. It provides the evidence required to execute the use of trustworthy and reliable AI systems. On the other hand, however, it also opens the door to locating possible vulnerabilities in an AI model. Traditional adversarial text attack uses word substitution, data augmentation techniques, and gradient-based attacks on powerful pre-trained Bidirectional Encoder Representations from Transformers (BERT) variants to generate adversarial sentences. These attacks are generally white-box in nature and not practical as they can be easily detected by humans e.g., Changing the word from “Poor” to “Rich”. We proposed a simple yet effective Grey-box cum Black-box approach that does not require the knowledge of the model while using a set of surrogate Transformer/BERT models to perform the attack using Explainable AI techniques. As Transformers are the current state-of-the-art models for almost all Natural Language Processing (NLP) tasks, an attack generated from BERT1 is transferable to BERT2. This transferability is made possible due to the attention mechanism in the transformer that allows the model to capture long-range dependencies in a sequence. Using the power of BERT generalisation via attention, we attempt to exploit how transformers learn by attacking a few surrogate transformer variants which are all based on a different architecture. We demonstrate that this approach is highly effective to generate semantically good sentences by changing as little as one word that is not detectable by humans while still fooling other BERT models.

Keywords: BERT, explainable AI, Grey-box text attack, transformer

Procedia PDF Downloads 112

Authors: Rasha Ali Dheyab

Abstract:

In today's increasingly interconnected and technologically-driven world, communication has evolved beyond traditional verbal exchanges. This paper delves into the fascinating realm of multimodal communication, a dynamic field at the intersection of linguistics, gesture studies, and technology. The study of how humans convey meaning through a combination of spoken language, gestures, facial expressions, and digital platforms has gained prominence as our modes of interaction continue to diversify. This exploration begins by examining the foundational theories in linguistics and gesture studies, tracing their historical development and mutual influences. It further investigates the role of nonverbal cues, such as gestures and facial expressions, in augmenting and sometimes even altering the meanings conveyed by spoken language. Additionally, the paper delves into the modern technological landscape, where emojis, GIFs, and other digital symbols have emerged as new linguistic tools, reshaping the ways in which we communicate and express emotions. The interaction between traditional and digital modes of communication is a central focus of this study. The paper investigates how technology has not only introduced new modes of expression but has also influenced the adaptation of existing linguistic and gestural patterns in online discourse. The emergence of virtual reality and augmented reality environments introduces yet another layer of complexity to multimodal communication, offering new avenues for studying how humans navigate and negotiate meaning in immersive digital spaces. Through a combination of literature review, case studies, and theoretical analysis, this paper seeks to shed light on the intricate interplay between language, gesture, and technology in the realm of multimodal communication. By understanding how these diverse modes of expression intersect and interact, we gain valuable insights into the ever-evolving nature of human communication and its implications for fields ranging from linguistics and psychology to human-computer interaction and digital anthropology.

Keywords: multimodal communication, linguistics ., gesture studies., emojis., verbal communication., digital

Procedia PDF Downloads 50

Authors: Yi Meng, Yun Gao

Abstract:

Having spent significant periods studying and working in North America and Europe, we, as two Chinese art educators, have been profoundly shaped by both Eastern and Western cultures. Consequently, our ambition is to enrich students' learning experiences by delving into and merging both cultural perspectives for innovative, creative expressions. This exposition draws on our action research study on students' visual literacy practices in a visual literacy course at a prominent Chinese university. The central premise was to explore innovative art forms by cross-utilizing various aspects of diverse cultures. By examining distinct cultural elements, we encouraged students to break away from familiar approaches and forge new paths in their creative endeavors. In implementing our curriculum, we utilized a multimodal pedagogy that deviated from the predominant print-based presentations typically employed in our classroom settings. This pedagogical approach effectively encouraged students to critically analyze the artifact, imbue it with their understanding and perspectives, and then produce an original piece. This approach also motivated students to leverage the semiotic potential of various communicative modes to address diverse cultural issues through their multimodal designs. To demonstrate the potential for cultural amalgamation, we utilized the artwork of Hong Kong-based artist Tik Ka. His works epitomize the fusion of Chinese traditions with Western pop culture, which served as a visual and conceptual reference point for students. Seeing how these distinct cultural elements could coexist and enrich each other in Tik Ka's work was inspiring and motivating for the students. Taken together, these pedagogical strategies helped create a dialogical space where students could actively experience, analyze, and negotiate complex modes of expression. This environment fostered active learning, encouraging students to apply their knowledge, question their assumptions, and reconsider their perspectives. Overall, such a unique approach to visual literacy education has the potential to reshape students' understanding of both cultures. By encouraging them to critically engage with their multimodal designs, we promoted an in-depth, nuanced appreciation of these diverse cultural heritages. The students no longer just interpreted and replicated images—they actively contributed to a dynamic and ongoing conversation between cultures.

Keywords: multimodal pedagogy, creative expressions, visual literacy education, multimodal designs

Procedia PDF Downloads 43

Authors: Ashleigh Cicconi

Abstract:

Teachers may not have access to necessary and effective strategies for managing stress, trauma, and emotional exhaustion, which can lead to burnout. This practice-based research focused on the exploration of teacher well-being through participation in a wellness program in order to mitigate high stress levels and feelings of burnout. The purpose of this qualitative research was to explore how a multimodal, trauma-informed yoga and arts-based mindfulness program impacted stress levels and overall well-being for teachers in a school setting. The case study approach was used to investigate participant perceptions of interactions between multimodal accessibility, a trauma-informed wellness program, and teacher well-being. A sample size of 10 teachers employed full-time at a public high school in the Mid-Atlantic region were recruited via email correspondence to participate in the eight-week wellness program. Data were triangulated across semi-structured interviews, journal entries, and focus group guided questions, and transcripts were uploaded into the NVivo software application for thematic analysis. Data showed perceptions of improvements in overall well-being from participation in the wellness program and that utilizing trauma-informed practices may be an effective coping skill for stress. The multimodal design of the program was perceived to positively impact participation and accessibility to wellness strategies. Findings from this study suggest that the inclusion of trauma-informed practices within a wellness program may be effective for managing stress and trauma experienced by teachers, thereby aiding in improvement in overall well-being. Findings also suggest that multimodality may be effective for increasing participation in and accessibility to wellness strategies.

Keywords: trauma informed practices, wellness program, teacher wellbeing, accessible program, multimodal

Procedia PDF Downloads 26

Authors: L. Ashok Kumar, N. Sujith Kumar

Abstract:

Most of the PV systems are designed with transformer for safety purpose with galvanic isolation. However, the transformer is big, heavy and expensive. Also, it reduces the overall frequency of the conversion stage. Generally PV inverter with transformer is having efficiency around 92%–94% only. To overcome these problems, transformerless PV system is introduced. It is smaller, lighter, cheaper and higher in efficiency. However, dangerous leakage current will flow between PV array and the grid due to the stray capacitance. There are different types of configurations available for transformerless inverters like H5, H6, HERIC, oH5, and Dual paralleled buck inverter. But each configuration is suffering from its own disadvantages like high conduction losses, shoot-through issues of switches, dead-time requirements at zero crossing instants of grid voltage to avoid grid shoot-through faults and MOSFET reverse recovery issues. The main objective of the proposed transformerless inverter is to address two key issues: One key issue for a transformerless inverter is that it is necessary to achieve high efficiency compared to other existing inverter topologies. Another key issue is that the inverter configuration should not have any shoot-through issues for higher reliability.

Keywords: grid-connected, photovoltaic (PV) systems, transformerless inverter, stray capacitance, common-mode, leakage current, pulse width modulation (PWM)

Procedia PDF Downloads 471

Authors: Pranav Gunhal., Krish Jhurani

Abstract:

This study tackles the significant computational challenge of predicting excited state dynamics in organic photovoltaic (OPV) materials—a pivotal factor in the performance of solar energy solutions. Time-dependent density functional theory (TDDFT), though effective, is computationally prohibitive for larger and more complex molecules. As a solution, the research explores the application of transformer neural networks, a type of artificial intelligence (AI) model known for its superior performance in natural language processing, to predict excited state dynamics in OPV materials. The methodology involves a two-fold process. First, the transformer model is trained on an extensive dataset comprising over 10,000 TDDFT calculations of excited state dynamics from a diverse set of OPV materials. Each training example includes a molecular structure and the corresponding TDDFT-calculated excited state lifetimes and key electronic transitions. Second, the trained model is tested on a separate set of molecules, and its predictions are rigorously compared to independent TDDFT calculations. The results indicate a remarkable degree of predictive accuracy. Specifically, for a test set of 1,000 OPV materials, the transformer model predicted excited state lifetimes with a mean absolute error of 0.15 picoseconds, a negligible deviation from TDDFT-calculated values. The model also correctly identified key electronic transitions contributing to the excited state dynamics in 92% of the test cases, signifying a substantial concordance with the results obtained via conventional quantum chemistry calculations. The practical integration of the transformer model with existing quantum chemistry software was also realized, demonstrating its potential as a powerful tool in the arsenal of materials scientists and chemists. The implementation of this AI model is estimated to reduce the computational cost of predicting excited state dynamics by two orders of magnitude compared to conventional TDDFT calculations. The successful utilization of transformer neural networks to accurately predict excited state dynamics provides an efficient computational pathway for the accelerated discovery and design of new OPV materials, potentially catalyzing advancements in the realm of sustainable energy solutions.

Keywords: transformer neural networks, organic photovoltaic materials, excited state dynamics, time-dependent density functional theory, predictive modeling

Procedia PDF Downloads 80

Authors: Ahmet Y. Arabul, Ibrahim Senol, Fatma Keskin Arabul, Mustafa G. Aydeniz, Yasemin Oner, Gokhan Kalkan

Abstract:

In the standards of IEC 60076-2 and IEC 60076-7, three different hot-spot temperature estimation methods are suggested. In this study, the algorithms which used in hot-spot temperature calculations are analyzed by comparing the algorithms with the results of an experimental set-up made by a Transformer Monitoring System (TMS) in use. In tested system, TMS uses only top oil temperature and load ratio for hot-spot temperature calculation. And also, it uses some constants from standards which are on agreed statements tables. During the tests, it came out that hot-spot temperature calculation method is just making a simple calculation and not uses significant all other variables that could affect the hot-spot temperature.

Keywords: Hot-spot temperature, monitoring system, power transformer, smart grid

Procedia PDF Downloads 548

Authors: Pankaj Kumar Jha, Mahendra Singh Hada, Brijendra Singh, Sandeep Yadav

Abstract:

Protective relays are commonly connected to the secondary windings of instrument transformers, i.e., current transformers (CTs) and/or capacitive voltage transformers (CVTs). The purpose of CT and CVT is to provide galvanic isolation from high voltages and reduce primary currents and voltages to a nominal quantity recognized by the protective relays. Selecting the correct instrument transformers for an application is imperative: failing to do so may compromise the relay’s performance, as the output of the instrument transformer may no longer be an accurately scaled representation of the primary quantity. Having an accurately rated instrument transformer is of no use if these devices are not properly connected. The performance of the protective relay is reliant on its programmed settings and on the current and voltage inputs from the instrument transformers secondary. This paper will help in understanding the fundamental concepts of the connections of Instrument Transformers to the protection relays and the effect of incorrect connection on the performance of protective relays. Multiple case studies of protection system mal-operations due to incorrect connections of instrument transformers will be discussed in detail in this paper. Apart from the connection issue of instrument transformers to protective relays, this paper will also discuss the effect of multiple earthing of CTs and CVTs secondary on the performance of the protection system. Case studies presented in this paper will help the readers to analyse the problem through real-world challenges in complex power system networks. This paper will also help the protection engineer in better analysis of disturbance records. CT and CVT connection errors can lead to undesired operations of protection systems. However, many of these operations can be avoided by adhering to industry standards and implementing tried-and-true field testing and commissioning practices. Understanding the effect of missing neutral of CVT, multiple earthing of CVT secondary, and multiple grounding of CT star points on the performance of the protection system through real-world case studies will help the protection engineer in better commissioning the protection system and maintenance of the protection system.

Keywords: bus reactor, current transformer, capacitive voltage transformer, distance protection, differential protection, directional earth fault, disturbance report, instrument transformer, ICT, REF protection, shunt reactor, voltage selection relay, VT fuse failure

Procedia PDF Downloads 51

Authors: George Eduful, Ebenezer A. Jackson, Kingsford A. Atanga

Abstract:

This paper investigates the impacts of cable length and capacity of transformer on ferroresonant overvoltage in cable-fed transformers. The study was conducted by simulation using the EMTP RV. Results show that ferroresonance can cause dangerous overvoltages ranging from 2 to 5 per unit. These overvoltages impose stress on insulations of transformers and cables and subsequently result in system failures. Undertaking Basic Multiple Regression Analysis (BMR) on the results obtained, a statistical model was obtained in terms of cable length and transformer capacity. The model is useful for ferroresonant prediction and control in cable-fed transformers.

Keywords: ferroresonance, cable-fed transformers, EMTP RV, regression analysis

Procedia PDF Downloads 498