Search results for: transformer fine-tuning

Authors: Giovanni Duran Polo

Abstract:

Demonstrate the importance of the stages such as Information, Diagnosis, Management, and Intervention is fundamental to have a historical, live, and quality inhabited center. One of the major actions to take is to promote the concept of the management of a historic center with harmonious development. For that, concerned actors should strengthen the concept that said historic center may be the neighborhood of all and for all. The centers of historical cities, presented as any other urban area, social, environmental issues etc; yet they get added value that have no other city neighborhoods. The equity component, either by the urban plan, or environmental quality offered properties of architectural, landscape or some land uses are the differentiating element, while the tool that makes them attractive face pressure exerted by new housing developments or shopping centers. That's why through the experience of working in historical centers, they are declared the actions in heritage areas. This paper will show how the encounter with each of these places are trying to take the phases of information, to gather all the data needed to be closer to the territory with specific data, diagnosis; which allowed the actors to see what state they were, felt how the heart is related to the rest of the city, show what problems affected the situation and what potential it had to compete in a global market. Also, to discuss the importance of the organization, as it is legal and normative basis for it have an order and a concept, when you know what can and what cannot, in an area where the citizen has many myth or history, when he wanted to intervene in protected buildings. It is also appropriate to show how it could develop the intervention phase, where the shares on the tangible elements and intervention for the protection of the heritage property are executed. The management is the final phase which will carry out all that was raised on paper, it's time to orient, explain, persuade, promote, and encourage citizens to take care of the heritage. It is profitable and also an obligation and it is not an insurmountable burden. It has to be said this is the time to pull all the cards to make the historical center and heritage becoming more alive today. It is the moment to make it more inhabited and to transformer it into a quality place, so citizens will cherish and understand the importance of such a place. Inhabited historical centers, endowments and equipment required, with trade quality, with constant cultural offer, with well-preserved buildings and tidy, modern and safe public spaces are always attractive for tourism, but first of all, the place should be conceived for citizens, otherwise everything will be doomed to failure.

Keywords: development, diagnosis, heritage historic center, intervention, management, patrimony

Procedia PDF Downloads 395

Authors: Nkwachukwu Chukwuchekwa, Joy Ulumma Chukwuchekwa

Abstract:

Magnetic characteristics of grain-oriented electrical steel (GOES) are usually measured at high flux densities suitable for its typical applications in power transformers. There are limited magnetic data at low flux densities which are relevant for the characterization of GOES for applications in metering instrument transformers and low frequency magnetic shielding in magnetic resonance imaging medical scanners. Magnetic properties such as coercivity, B-H loop, AC relative permeability and specific power loss of conventional grain oriented (CGO) and high permeability grain oriented (HGO) electrical steels were measured and compared at high and low flux densities at power magnetising frequency. 40 strips comprising 20 CGO and 20 HGO, 305 mm x 30 mm x 0.27 mm from a supplier were tested. The HGO and CGO strips had average grain sizes of 9 mm and 4 mm respectively. Each strip was singly magnetised under sinusoidal peak flux density from 8.0 mT to 1.5 T at a magnetising frequency of 50 Hz. The novel single sheet tester comprises a personal computer in which LabVIEW version 8.5 from National Instruments (NI) was installed, a NI 4461 data acquisition (DAQ) card, an impedance matching transformer, to match the 600  minimum load impedance of the DAQ card with the 5 to 20  low impedance of the magnetising circuit, and a 4.7 Ω shunt resistor. A double vertical yoke made of GOES which is 290 mm long and 32 mm wide is used. A 500-turn secondary winding, about 80 mm in length, was wound around a plastic former, 270 mm x 40 mm, housing the sample, while a 100-turn primary winding, covering the entire length of the plastic former was wound over the secondary winding. A standard Epstein strip to be tested is placed between the yokes. The magnetising voltage was generated by the LabVIEW program through a voltage output from the DAQ card. The voltage drop across the shunt resistor and the secondary voltage were acquired by the card for calculation of magnetic field strength and flux density respectively. A feedback control system implemented in LabVIEW was used to control the flux density and to make the induced secondary voltage waveforms sinusoidal to have repeatable and comparable measurements. The low noise NI4461 card with 24 bit resolution and a sampling rate of 204.8 KHz and 92 KHz bandwidth were chosen to take the measurements to minimize the influence of thermal noise. In order to reduce environmental noise, the yokes, sample and search coil carrier were placed in a noise shielding chamber. HGO was found to have better magnetic properties at both high and low magnetisation regimes. This is because of the higher grain size of HGO and higher grain-grain misorientation of CGO. HGO is better CGO in both low and high magnetic field applications.

Keywords: flux density, electrical steel, LabVIEW, magnetization

Procedia PDF Downloads 291

Authors: Rowan P. Martnishn

Abstract:

During the execution of a National Endowment of the Arts grant, approximately 55 interviews were collected from professionals across various fields. These interviews were used to create deliverables – historical connections for creations that began as art and evolved entirely into computing technology. With dozens of hours’ worth of transcripts to be analyzed by qualitative coders, a quantitative methodology was created to sift through the documents. The initial step was to both clean and format all the data. First, a basic spelling and grammar check was applied, as well as a Python script for normalized formatting which used an open-source grammatical formatter to make the data as coherent as possible. 10 documents were randomly selected to manually review, where words often incorrectly translated during the transcription were recorded and replaced throughout all other documents. Then, to remove all banter and side comments, the transcripts were spliced into paragraphs (separated by change in speaker) and all paragraphs with less than 300 characters were removed. Secondly, a keyword extractor, a form of natural language processing where significant words in a document are selected, was run on each paragraph for all interviews. Every proper noun was put into a data structure corresponding to that respective interview. From there, a Bidirectional and Auto-Regressive Transformer (B.A.R.T.) summary model was then applied to each paragraph that included any of the proper nouns selected from the interview. At this stage the information to review had been sent from about 60 hours’ worth of data to 20. The data was further processed through light, manual observation – any summaries which proved to fit the criteria of the proposed deliverable were selected, as well their locations within the document. This narrowed that data down to about 5 hours’ worth of processing. The qualitative researchers were then able to find 8 more connections in addition to our previous 4, exceeding our minimum quota of 3 to satisfy the grant. Major findings of the study and subsequent curation of this methodology raised a conceptual finding crucial to working with qualitative data of this magnitude. In the use of artificial intelligence there is a general trade off in a model between breadth of knowledge and specificity. If the model has too much knowledge, the user risks leaving out important data (too general). If the tool is too specific, it has not seen enough data to be useful. Thus, this methodology proposes a solution to this tradeoff. The data is never altered outside of grammatical and spelling checks. Instead, the important information is marked, creating an indicator of where the significant data is without compromising the purity of it. Secondly, the data is chunked into smaller paragraphs, giving specificity, and then cross-referenced with the keywords (allowing generalization over the whole document). This way, no data is harmed, and qualitative experts can go over the raw data instead of using highly manipulated results. Given the success in deliverable creation as well as the circumvention of this tradeoff, this methodology should stand as a model for synthesizing qualitative data while maintaining its original form.

Keywords: B.A.R.T.model, keyword extractor, natural language processing, qualitative coding

Procedia PDF Downloads 28

Authors: Jia Li, Xing Wei, Yuchen Hong, Yang Lu

Abstract:

Detecting vehicle behavior has always been the focus of intelligent transportation, but with the explosive growth of the number of vehicles and the complexity of the road environment, the vehicle behavior videos captured by traditional surveillance have been unable to satisfy the study of vehicle behavior. The traditional method of manually labeling vehicle behavior is too time-consuming and labor-intensive, but the existing object detection and tracking algorithms have poor practicability and low behavioral location detection rate. This paper proposes a vehicle behavior detection algorithm based on the dual-stream convolution network and the multi-dimensional video dynamic detection network. In the videos, the straight-line behavior of the vehicle will default to the background behavior. The Changing lanes, turning and turning around are set as target behaviors. The purpose of this model is to automatically mark the target behavior of the vehicle from the untrimmed videos. First, the target behavior proposals in the long video are extracted through the dual-stream convolution network. The model uses a dual-stream convolutional network to generate a one-dimensional action score waveform, and then extract segments with scores above a given threshold M into preliminary vehicle behavior proposals. Second, the preliminary proposals are pruned and identified using the multi-dimensional video dynamic detection network. Referring to the hierarchical reinforcement learning, the multi-dimensional network includes a Timer module and a Spacer module, where the Timer module mines time information in the video stream and the Spacer module extracts spatial information in the video frame. The Timer and Spacer module are implemented by Long Short-Term Memory (LSTM) and start from an all-zero hidden state. The Timer module uses the Transformer mechanism to extract timing information from the video stream and extract features by linear mapping and other methods. Finally, the model fuses time information and spatial information and obtains the location and category of the behavior through the softmax layer. This paper uses recall and precision to measure the performance of the model. Extensive experiments show that based on the dataset of this paper, the proposed model has obvious advantages compared with the existing state-of-the-art behavior detection algorithms. When the Time Intersection over Union (TIoU) threshold is 0.5, the Average-Precision (MP) reaches 36.3% (the MP of baselines is 21.5%). In summary, this paper proposes a vehicle behavior detection model based on multi-dimensional dynamic detection network. This paper introduces spatial information and temporal information to extract vehicle behaviors in long videos. Experiments show that the proposed algorithm is advanced and accurate in-vehicle timing behavior detection. In the future, the focus will be on simultaneously detecting the timing behavior of multiple vehicles in complex traffic scenes (such as a busy street) while ensuring accuracy.

Keywords: vehicle behavior detection, convolutional neural network, long short-term memory, deep learning

Procedia PDF Downloads 130

Authors: George Zhou, Yunchan Chen, Candace Chien

Abstract:

Kidney replacement therapy is the current standard of care for end-stage renal diseases. In-center or home hemodialysis remains an integral component of the therapeutic regimen. Arteriovenous fistulas (AVF) make up the vascular circuit through which blood is filtered and returned. Naturally, AVF patency determines whether adequate clearance and filtration can be achieved and directly influences clinical outcomes. Our aim was to build a deep learning model for automated AVF stenosis screening based on the sound of blood flow through the AVF. A total of 311 patients with AVF were enrolled in this study. Blood flow sounds were collected using a digital stethoscope. For each patient, blood flow sounds were collected at 6 different locations along the patient’s AVF. The 6 locations are artery, anastomosis, distal vein, middle vein, proximal vein, and venous arch. A total of 1866 sounds were collected. The blood flow sounds are labeled as “patent” (normal) or “stenotic” (abnormal). The labels are validated from concurrent ultrasound. Our dataset included 1527 “patent” and 339 “stenotic” sounds. We show that blood flow sounds vary significantly along the AVF. For example, the blood flow sound is loudest at the anastomosis site and softest at the cephalic arch. Contextualizing the sound with location metadata significantly improves classification performance. How to encode and incorporate categorical metadata is an active area of research1. Herein, we study ordinal (i.e., integer) encoding schemes. The numerical representation is concatenated to the flattened feature vector. We train a vision transformer (ViT) on spectrogram image representations of the sound and demonstrate that using scalar multiples of our integer encodings improves classification performance. Models are evaluated using a 10-fold cross-validation procedure. The baseline performance of our ViT without any location metadata achieves an AuROC and AuPRC of 0.68 ± 0.05 and 0.28 ± 0.09, respectively. Using the following encodings of Artery:0; Arch: 1; Proximal: 2; Middle: 3; Distal 4: Anastomosis: 5, the ViT achieves an AuROC and AuPRC of 0.69 ± 0.06 and 0.30 ± 0.10, respectively. Using the following encodings of Artery:0; Arch: 10; Proximal: 20; Middle: 30; Distal 40: Anastomosis: 50, the ViT achieves an AuROC and AuPRC of 0.74 ± 0.06 and 0.38 ± 0.10, respectively. Using the following encodings of Artery:0; Arch: 100; Proximal: 200; Middle: 300; Distal 400: Anastomosis: 500, the ViT achieves an AuROC and AuPRC of 0.78 ± 0.06 and 0.43 ± 0.11. respectively. Interestingly, we see that using increasing scalar multiples of our integer encoding scheme (i.e., encoding “venous arch” as 1,10,100) results in progressively improved performance. In theory, the integer values do not matter since we are optimizing the same loss function; the model can learn to increase or decrease the weights associated with location encodings and converge on the same solution. However, in the setting of limited data and computation resources, increasing the importance at initialization either leads to faster convergence or helps the model escape a local minimum.

Keywords: arteriovenous fistula, blood flow sounds, metadata encoding, deep learning

Procedia PDF Downloads 87

Authors: P. Ramesh, Aby Joseph, Arya G. Lal, U. S. Aji

Abstract:

Integration of distributed green renewable energy sources in addition with battery energy storage is an inevitable requirement in a smart grid environment. To achieve this, an Intelligent Green Energy Management System (i-GEMS) needs to be incorporated to ensure coordinated operation between supply and load demand based on the hierarchy of Renewable Energy Sources (RES), battery energy storage and distribution grid. A bi-directional energy harvester is an integral component facilitating Intelligent Green Energy Management System (i-GEMS) and it is required to meet the technical challenges mentioned as follows: (1) capability for bi-directional mode of operation (buck/boost) (2) reduction of circuit parasitic to suppress voltage spikes (3) converter startup problem (4) high frequency magnetics (5) higher power density (6) mode transition issues during battery charging and discharging. This paper is focused to address the above mentioned issues and targeted to design, develop and implement a bi-directional energy harvester with galvanic isolation. In this work, the hardware architecture for bi-directional energy harvester rated 3.5 kW is developed with Isolated Full Bridge Boost Converter (IFBBC) as well as Dual Active Bridge (DAB) Converter configuration using modular power electronics hardware which is identical for both solar PV array and battery energy storage. In IFBBC converter, the current fed full bridge circuit is enabled and voltage fed full bridge circuit is disabled through Pulse Width Modulation (PWM) pulses for boost mode of operation and vice-versa for buck mode of operation. In DAB converter, all the switches are in active state so as to adjust the phase shift angle between primary full bridge and secondary full bridge which in turn decides the power flow directions depending on modes (boost/buck) of operation. Here, the control algorithm is developed to ensure the regulation of the common DC link voltage and maximum power extraction from the renewable energy sources depending on the selected mode (buck/boost) of operation. The circuit analysis and simulation study are conducted using PSIM 9.0 in three scenarios which are - 1.IFBBC with passive clamp, 2. IFBBC with active clamp, 3. DAB converter. In this work, a common hardware prototype for bi-directional energy harvester with 3.5 kW rating is built for IFBBC and DAB converter configurations. The power circuit is equipped with right choice of MOSFETs, gate drivers with galvanic isolation, high frequency transformer, filter capacitors, and filter boost inductor. The experiment was conducted for IFBBC converter with passive clamp under boost mode and the prototype confirmed the simulation results showing the measured efficiency as 88% at 2.5 kW output power. The digital controller hardware platform is developed using floating point microcontroller TMS320F2806x from Texas Instruments. The firmware governing the operation of the bi-directional energy harvester is written in C language and developed using code composer studio. The comprehensive analyses of the power circuit design, control strategy for battery charging/discharging under buck/boost modes and comparative performance evaluation using simulation and experimental results will be presented.

Keywords: bi-directional energy harvester, dual active bridge, isolated full bridge boost converter, intelligent green energy management system, maximum power point tracking, renewable energy sources

Procedia PDF Downloads 140

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 42