Search results for: Python

Authors: Nkechinyere Joy Olawuyi, Babajide Samuel Afolabi, Bola Ibitoye

Abstract:

This study collected a preprocessed dataset of chest radiographs and formulated a deep neural network model for detecting abnormalities. It also evaluated the performance of the formulated model and implemented a prototype of the formulated model. This was with the view to developing a deep neural network model to automatically classify abnormalities in chest radiographs. In order to achieve the overall purpose of this research, a large set of chest x-ray images were sourced for and collected from the CheXpert dataset, which is an online repository of annotated chest radiographs compiled by the Machine Learning Research Group, Stanford University. The chest radiographs were preprocessed into a format that can be fed into a deep neural network. The preprocessing techniques used were standardization and normalization. The classification problem was formulated as a multi-label binary classification model, which used convolutional neural network architecture to make a decision on whether an abnormality was present or not in the chest radiographs. The classification model was evaluated using specificity, sensitivity, and Area Under Curve (AUC) score as the parameter. A prototype of the classification model was implemented using Keras Open source deep learning framework in Python Programming Language. The AUC ROC curve of the model was able to classify Atelestasis, Support devices, Pleural effusion, Pneumonia, A normal CXR (no finding), Pneumothorax, and Consolidation. However, Lung opacity and Cardiomegaly had a probability of less than 0.5 and thus were classified as absent. Precision, recall, and F1 score values were 0.78; this implies that the number of False Positive and False Negative is the same, revealing some measure of label imbalance in the dataset. The study concluded that the developed model is sufficient to classify abnormalities present in chest radiographs into present or absent.

Keywords: transfer learning, convolutional neural network, radiograph, classification, multi-label

Procedia PDF Downloads 129

Authors: Y. Trantafyllou, C. Alexandraki

Abstract:

Music interpretation accounts to the way musicians shape their performance by deliberately deviating from composers’ intentions, which are commonly communicated via some form of music transcription, such as a music score. For transcribed and non-improvised music, music expression is manifested by introducing subtle deviations in tempo, dynamics and articulation during the evolution of performance. This paper presents an application, named FlameCens, which, given two recordings of the same piece of music, presumably performed by different musicians, allow visualising deviations in tempo and dynamics during playback. The application may also compare a certain performance to the music score of that piece (i.e. MIDI file), which may be thought of as an expression-neutral representation of that piece, hence depicting the expressive queues employed by certain performers. FlameCens uses the Dynamic Time Warping algorithm to compare two audio sequences, based on CENS (Chroma Energy distribution Normalized Statistics) audio features. Expressive deviations are illustrated in a moving flame, which is generated by an animation of particles. The length of the flame is mapped to deviations in dynamics, while the slope of the flame is mapped to tempo deviations so that faster tempo changes the slope to the right and slower tempo changes the slope to the left. Constant slope signifies no tempo deviation. The detected deviations in tempo and dynamics can be additionally recorded in a text file, which allows for offline investigation. Moreover, in the case of monophonic music, the color of particles is used to convey the pitch of the notes during performance. FlameCens has been implemented in Python and it is openly available via GitHub. The application has been experimentally validated for different music genres including classical, contemporary, jazz and popular music. These experiments revealed that FlameCens can be a valuable tool for music specialists (i.e. musicians or musicologists) to investigate the expressive performance strategies employed by different musicians, as well as for music audience to enhance their listening experience.

Keywords: audio synchronization, computational music analysis, expressive music performance, information visualization

Procedia PDF Downloads 131

Authors: Mohan Li

Abstract:

Nowadays, intelligent research technology is more and more important than traditional research methods in urban research work, and this proportion will greatly increase in the next few decades. Frequently such analyzing work cannot be carried without some software engineering knowledge. And here, domain models of urban research will be necessary when applying software engineering knowledge to urban work. In many urban plan practice projects, making rational models, feeding reliable data, and providing enough computation all make indispensable assistance in producing good urban planning. During the whole work process, domain models can optimize workflow design. At present, human beings have entered the era of big data. The amount of digital data generated by cities every day will increase at an exponential rate, and new data forms are constantly emerging. How to select a suitable data set from the massive amount of data, manage and process it has become an ability that more and more planners and urban researchers need to possess. This paper summarizes and makes predictions of the emergence of technologies and technological iterations that may affect urban research in the future, discover urban problems, and implement targeted sustainable urban strategies. They are summarized into seven major domain models. They are urban and rural regional domain model, urban ecological domain model, urban industry domain model, development dynamic domain model, urban social and cultural domain model, urban traffic domain model, and urban space domain model. These seven domain models can be used to guide the construction of systematic urban research topics and help researchers organize a series of intelligent analytical tools, such as Python, R, GIS, etc. These seven models make full use of quantitative spatial analysis, machine learning, and other technologies to achieve higher efficiency and accuracy in urban research, assisting people in making reasonable decisions.

Keywords: big data, domain model, urban planning, urban quantitative analysis, machine learning, workflow design

Procedia PDF Downloads 177

Authors: Durga Toshniwal, Lokesh Agrawal

Abstract:

As the web continues to play an increasing role in information exchange, and conducting daily activities, computer users have become the target of miscreants which infects hosts with malware or adware for financial gains. Unfortunately, even a single visit to compromised web site enables the attacker to detect vulnerabilities in the user’s applications and force the downloading of multitude of malware binaries. We provide an approach to effectively scan the so-called drive-by downloads on the Internet. Drive-by downloads are result of URLs that attempt to exploit their visitors and cause malware to be installed and run automatically. To scan the web for malicious pages, the first step is to use a crawler to collect URLs that live on the Internet, and then to apply fast prefiltering techniques to reduce the amount of pages that are needed to be examined by precise, but slower, analysis tools (such as honey clients or antivirus programs). Although the technique is effective, it requires a substantial amount of resources. A main reason is that the crawler encounters many pages on the web that are legitimate and needs to be filtered. In this paper, to characterize the nature of this rising threat, we present implementation of a web crawler on Python, an approach to search the web more efficiently for pages that are likely to be malicious, filtering benign pages and passing remaining pages to antivirus program for detection of malwares. Our approaches starts from an initial seed of known, malicious web pages. Using these seeds, our system generates search engines queries to identify other malicious pages that are similar to the ones in the initial seed. By doing so, it leverages the crawling infrastructure of search engines to retrieve URLs that are much more likely to be malicious than a random page on the web. The results shows that this guided approach is able to identify malicious web pages more efficiently when compared to random crawling-based approaches.

Keywords: web crawler, malwares, seeds, drive-by-downloads, security

Procedia PDF Downloads 230

Authors: Michael Moeti, Khuliso Sigama, Thapelo Samuel Matlala

Abstract:

One of the principal challenges that the world is confronted with is insecurity. The crime rate is increasing exponentially, and protecting our physical assets especially in the motorist industry, is becoming impossible when applying our own strength. The need to develop technological solutions that detect and report theft without any human interference is inevitable. This is critical, especially for vehicle owners, to ensure theft detection and speedy identification towards recovery efforts in cases where a vehicle is missing or attempted theft is taking place. The vehicle theft detection system uses Convolutional Neural Network (CNN) to recognize the driver's face captured using an installed mobile phone device. The location identification function uses a Global Positioning System (GPS) to determine the real-time location of the vehicle. Upon identification of the location, Global System for Mobile Communications (GSM) technology is used to report or notify the vehicle owner about the whereabouts of the vehicle. The installed mobile app was implemented by making use of python as it is undoubtedly the best choice in machine learning. It allows easy access to machine learning algorithms through its widely developed library ecosystem. The graphical user interface was developed by making use of JAVA as it is better suited for mobile development. Google's online database (Firebase) was used as a means of storage for the application. The system integration test was performed using a simple percentage analysis. Sixty (60) vehicle owners participated in this study as a sample, and questionnaires were used in order to establish the acceptability of the system developed. The result indicates the efficiency of the proposed system, and consequently, the paper proposes the use of the system can effectively monitor the vehicle at any given place, even if it is driven outside its normal jurisdiction. More so, the system can be used as a database to detect, locate and report missing vehicles to different security agencies.

Keywords: CNN, location identification, tracking, GPS, GSM

Procedia PDF Downloads 172

Authors: Reza Miresmaeili, Asghar Heydari Astaraee, Fereshteh Dolati

Abstract:

In the present study, an analytical model based on dislocation density model was developed to simulate grain refinement in surface mechanical attrition treatment (SMAT). The correlation between SMAT time and development in plastic strain on one hand, and dislocation density evolution, on the other hand, was established to simulate the grain refinement in SMAT. A dislocation density-based constitutive material law was implemented using VUHARD subroutine. A random sequence of shots is taken into consideration for multiple impacts model using Python programming language by utilizing a random function. The simulation technique was to model each impact in a separate run and then transferring the results of each run as initial conditions for the next run (impact). The developed Finite Element (FE) model of multiple impacts describes the coverage evolution in SMAT. Simulations were run to coverage levels as high as 4500%. It is shown that the coverage implemented in the FE model is equal to the experimental coverage. It is depicted that numerical SMAT coverage parameter is adequately conforming to the well-known Avrami model. Comparison between numerical results and experimental measurements for residual stresses and depth of deformation layers confirms the performance of the established FE model for surface engineering evaluations in SMA treatment. X-ray diffraction (XRD) studies of grain refinement, including resultant grain size and dislocation density, were conducted to validate the established model. The full width at half-maximum in XRD profiles can be used to measure the grain size. Numerical results and experimental measurements of grain refinement illustrate good agreement and show the capability of established FE model to predict the gradient microstructure in SMA treatment.

Keywords: dislocation density, grain refinement, severe plastic deformation, simulation, surface mechanical attrition treatment

Procedia PDF Downloads 138

Authors: WonSun Chen, James Boylan, Erwin Muharemovic, Denny Meyer

Abstract:

In Australia, drivers aged between 18 and 24 remained at high risk of road fatality over the last decade. Despite the successful implementation of the Graduated Licensing System (GLS) that supports young drivers in their early phases of driving, the road fatality statistics for these drivers remains high. In response to these statistics, studies conducted in Australia prior to the start of the COVID-19 pandemic have demonstrated the benefits of using telematics devices for improving driving behaviour, However, the impact of COVID-19 lockdown on young drivers’ driving behaviour has emerged as a global concern. Therefore, this naturalistic study aimed to evaluate and compare the driving behaviour(such as acceleration, braking, speeding, etc.) of young drivers with the adoption of in-vehicle telematics devices. Forty-two drivers aged between 18 and 30 and residing in the Australian state of Victoria participated in this study during the period of May to October 2022. All participants drove with the telematics devices during the first 30-day. At the start of the second 30-day, twenty-one participants were randomised to an intervention group where they were provided with an additional telematics ray device that provided visual feedback to the drivers, especially when they committed to aggressive driving behaviour. The remaining twenty-one participants remined their driving journeys without the extra telematics ray device (control group). Such trustworthy data enabled the assessment of changes in the driving behaviour of these young drivers using a machine learning approach in Python. Results are expected to show participants from the intervention group will show improvements in their driving behaviour compared to those from the control group.Furthermore, the telematics data enable the assessment and quantification of such improvements in driving behaviour. The findings from this study are anticipated to shed some light in guiding the development of customised campaigns and interventions to further address the high road fatality among young drivers in Australia.

Keywords: driving behaviour, naturalistic study, telematics data, young drivers

Procedia PDF Downloads 124

Authors: Eric York, James Tadio, Silas Owusu Antwi

Abstract:

Laundry wastewater contains a diverse range of chemical pollutants that can have detrimental effects on human health and the environment. In this study, simulation studies by Spyder Python software v 3.2 to assess the efficacy of biochar in removing PO₄³⁻ from wastewater were conducted. Through modeling and simulation, the mechanisms involved in the adsorption process of phosphate by biochar were studied by altering variables which is specific to the phosphate from common laundry phosphate detergents, such as the aqueous solubility, initial concentration, and temperature using the Dudinin Approach (DA). Results showed that the concentration equilibrate at near the highest concentrations for Sugar beet-120 mgL⁻¹, Tailing-85 mgL⁻¹, CaO- rich-50 mgL⁻¹, Eggshell and rice straw-48 mgL⁻¹, Undaria Pinnatifida Roots-190 mgL⁻¹, Ca-Alginate Granular Beads -240 mgL⁻¹, Laminaria Japonica Powder -900 mgL⁻¹, Pinesaw dust-57 mgL⁻¹, Ricehull-190 mgL⁻¹, sesame straw- 470 mgL⁻¹, Sugar Bagasse-380 mgL⁻¹, Miscanthus Giganteus-240 mgL⁻¹, Wood Bc-130 mgL⁻¹, Pine-25 mgL⁻¹, Sawdust-6.8 mgL⁻¹, Sewage Sludge-, Rice husk-12 mgL⁻¹, Corncob-117 mgL⁻¹, Maize straw- 1800 mgL⁻¹ while Peanut -Eucalyptus polybractea-, Crawfish equilibrated at near concentration. CO₂ activated Thalia, sewage sludge biochar, Broussonetia Papyrifera Leaves equilibrated just at the lower concentration. Only Soyer bean Stover exhibited a sharp rise and fall peak in mid-concentration at 2 mgL⁻¹ volume. The modelling results were consistent with experimental findings from the literature, ensuring the accuracy, repeatability, and reliability of the simulation study. The simulation study provided insights into adsorption for PO₄³⁻ from wastewater by biochar using concentration per volume that can be adsorbed ideally under the given conditions. Studies showed that applying the principle experimentally in real wastewater with all its complexity is warranted and not far-fetched.

Keywords: simulation studies, phosphate removal, biochar, adsorption, wastewater treatment

Procedia PDF Downloads 142

Authors: A. M. John-Otumu, M. M. Rahman, O. C. Nwokonkwo, M. C. Onuoha

Abstract:

Online social media networks have long served as a primary arena for group conversations, gossip, text-based information sharing and distribution. The use of natural language processing techniques for text classification and unbiased decision-making has not been far-fetched. Proper classification of this textual information in a given context has also been very difficult. As a result, we decided to conduct a systematic review of previous literature on sentiment classification and AI-based techniques that have been used in order to gain a better understanding of the process of designing and developing a robust and more accurate sentiment classifier that can correctly classify social media textual information of a given context between hate speech and inverted compliments with a high level of accuracy by assessing different artificial intelligence techniques. We evaluated over 250 articles from digital sources like ScienceDirect, ACM, Google Scholar, and IEEE Xplore and whittled down the number of research to 31. Findings revealed that Deep learning approaches such as CNN, RNN, BERT, and LSTM outperformed various machine learning techniques in terms of performance accuracy. A large dataset is also necessary for developing a robust sentiment classifier and can be obtained from places like Twitter, movie reviews, Kaggle, SST, and SemEval Task4. Hybrid Deep Learning techniques like CNN+LSTM, CNN+GRU, CNN+BERT outperformed single Deep Learning techniques and machine learning techniques. Python programming language outperformed Java programming language in terms of sentiment analyzer development due to its simplicity and AI-based library functionalities. Based on some of the important findings from this study, we made a recommendation for future research.

Keywords: artificial intelligence, natural language processing, sentiment analysis, social network, text

Procedia PDF Downloads 116

Authors: Masomeh Jamshid Nejad

Abstract:

Nowadays, statistical literacy is no longer a necessary skill but an essential skill with broad applications across diverse fields, especially in operational decision areas such as business management, finance, and economics. As such, learning and deep understanding of statistical concepts are essential in the context of business studies. One of the crucial topics in statistical theory and its application is the normal distribution, often called a bell-shaped curve. To interpret data and conduct hypothesis tests, comprehending the properties of normal distribution (the mean and standard deviation) is essential for business students. This requires undergraduate students in the field of economics and business management to visualize and work with data following a normal distribution. Since technology is interconnected with education these days, it is important to teach statistics topics in the context of Python, R-studio, and Microsoft Excel to undergraduate students. This research endeavours to shed light on the effect of Excel-based instruction on learners’ knowledge of statistics, specifically the central concept of normal distribution. As such, two groups of undergraduate students (from the Business Management program) were compared in this research study. One group underwent Excel-based instruction and another group relied only on traditional teaching methods. We analyzed experiential data and BBA participants’ responses to statistic-related questions focusing on the normal distribution, including its key attributes, such as the mean and standard deviation. The results of our study indicate that exposing students to Excel-based learning supports learners in comprehending statistical concepts more effectively compared with the other group of learners (teaching with the traditional method). In addition, students in the context of Excel-based instruction showed ability in picturing and interpreting data concentrated on normal distribution.

Keywords: statistics, excel-based instruction, data visualization, pedagogy

Procedia PDF Downloads 55

Authors: Muhammad Saqib Manzoor, Camille Dickson-Deane, Prashan Karunaratne

Abstract:

Cobb-Douglas production (utility) function is a fundamental function widely used in economics teaching and research. The key reason is the function's characteristics to describe the actual production using inputs like labour and capital. The characteristics of the function like returns to scale, marginal, and diminishing marginal productivities are covered in the introductory units in both microeconomics and macroeconomics with a 2-dimensional static visualisation of the function. However, less insight is provided regarding three-dimensional surface, changes in the curvature properties due to returns to scale, the linkage of the short-run production function with its long-run counterpart and marginal productivities, the level curves, and the constraint optimisation. Since (freshman) learners have diverse prior knowledge and cognitive skills, the existing “one size fits all” approach is not very helpful. The aim of this study is to bridge this gap by introducing technological intervention with interactive animations of the three-dimensional surface and sequential unveiling of the characteristics mentioned above using Python software. A small classroom intervention has helped students enhance their analytical and visualisation skills towards active and authentic learning of this topic. However, to authenticate the strength of our approach, a quasi-Delphi study will be conducted to ask domain-specific experts, “What value to the learning process in economics is there using a 2-dimensional static visualisation compared to using a 3-dimensional dynamic visualisation?’ Here three perspectives of the intervention were reviewed by a panel comprising of novice students, experienced students, novice instructors, and experienced instructors in an effort to determine the learnings from each type of visualisations within a specific domain of knowledge. The value of this approach is key to suggesting different pedagogical methods which can enhance learning outcomes.

Keywords: cobb-douglas production function, quasi-Delphi method, effective teaching and learning, 3D-visualisations

Procedia PDF Downloads 146

Authors: Guillaume Richard, Sarra Zaied

Abstract:

Marine waste quantities increase more and more, 5 million tons of plastic waste enter the ocean every year. Their spatiotemporal distribution is never homogeneous and depends mainly on the hydrodynamic characteristics of the environment, as well as the size and location of the waste. As part of optimizing collect of marine plastic wastes, it is important to measure and monitor their evolution over time. In this context, diverse studies have been dedicated to describing waste behavior in order to identify its accumulation in ocean areas. None of the existing tools which track objects at sea had the objective of tracking down a slick of waste. Moreover, the applications related to marine waste are in the minority compared to rescue applications or oil slicks tracking applications. These approaches are able to accurately simulate an object's behavior over time but not during the collection mission of a waste sheet. This paper presents numerical modeling of a boat’s wake impact on the floating marine waste behavior during a collection mission. The aim is to predict the trajectory of a marine waste slick to optimize its collection using meteorological data of ocean currents, wind, and possibly waves. We have made the choice to use Ocean Parcels which is a Python library suitable for trajectoring particles in the ocean. The modeling results showed the important role of advection and diffusion processes in the spatiotemporal distribution of floating plastic litter. The performance of the proposed method was evaluated on real data collected from the Copernicus Marine Environment Monitoring Service (CMEMS). The results of the evaluation in Cape of Good Hope (South Africa) prove that the proposed approach can effectively predict the position and velocity of marine litter during collection, which allowed for optimizing time and more than $90\%$ of the amount of collected waste.

Keywords: marine litter, advection-diffusion equation, sea current, numerical model

Procedia PDF Downloads 88

Authors: Navneet Kaur, S. D. Tiwari

Abstract:

Superparamagnetism is an interesting phenomenon and observed in small particles of magnetic materials. It arises due to a reduction in particle size. In the superparamagnetic state, as the thermal energy overcomes magnetic anisotropy energy, the magnetic moment vector of particles flip their magnetization direction between states of minimum energy. Superparamagnetic nanoparticles have been attracting the researchers due to many applications such as information storage, magnetic resonance imaging, biomedical applications, and sensors. For information storage, thermal fluctuations lead to loss of data. So that nanoparticles should have high blocking temperature. And to achieve this, nanoparticles should have a higher magnetic moment and magnetic anisotropy constant. In this work, the magnetic anisotropy constant of the antiferromagnetic nanoparticles system is determined. Magnetic studies on nanoparticles of NiO (nickel oxide) are reported well. This antiferromagnetic nanoparticle system has high blocking temperature and magnetic anisotropy constant of order 105 J/m3. The magnetic study of NiO nanoparticles in the superparamagnetic region is presented. NiO particles of two different sizes, i.e., 6 and 8 nm, are synthesized using the chemical route. These particles are characterized by an x-ray diffractometer, transmission electron microscope, and superconducting quantum interference device magnetometry. The magnetization vs. applied magnetic field and temperature data for both samples confirm their superparamagnetic nature. The blocking temperature for 6 and 8 nm particles is found to be 200 and 172 K, respectively. Magnetization vs. applied magnetic field data of NiO is fitted to an appropriate magnetic expression using a non-linear least square fit method. The role of particle size distribution and magnetic anisotropy is taken in to account in magnetization expression. The source code is written in Python programming language. This fitting provides us the magnetic anisotropy constant for NiO and other magnetic fit parameters. The particle size distribution estimated matches well with the transmission electron micrograph. The value of magnetic anisotropy constants for 6 and 8 nm particles is found to be 1.42 X 105 and 1.20 X 105 J/m3, respectively. The obtained magnetic fit parameters are verified using the Neel model. It is concluded that the effect of magnetic anisotropy should not be ignored while studying the magnetization process of nanoparticles.

Keywords: anisotropy, superparamagnetic, nanoparticle, magnetization

Procedia PDF Downloads 134

Authors: Natasha Botha, Gary Corderely, Helen M. Inglis

Abstract:

With the move to cleaner energy applications the transport industry is working towards on-board hydrogen, or compressed natural gas-fuelled vehicles. A popular method for storage is to use composite overwrapped pressure vessels (COPV) because of their high strength to weight ratios. The proper design of these COPVs are according to international standards; this study aims to provide a preliminary design for a 350 Bar Type IV COPV (i.e. a polymer liner with a composite overwrap). Netting analysis, a popular analytical approach, is used as a first step to generate an initial design concept for the composite winding. This design is further improved upon by following the American Society of Mechanical Engineers (ASME) Boiler and Pressure Vessel standards, Section X: Fibre-reinforced composite pressure vessels. A design program based on these two approaches is developed using Python. A numerical model of a burst test simulation is developed based on the two approaches and compared. The results indicate that the netting analysis provides a good preliminary design, while the ASME-based design is more robust and accurate as it includes a better approximation of the material behaviour. Netting analysis is an easy method to follow when considering an initial concept design for the composite winding when not all the material characteristics are known. Once these characteristics have been fully defined with experimental testing, an ASME-based design should always be followed to ensure that all designs conform to international standards and practices. Future work entails more detailed numerical testing of the design for improvement, this will include the boss design. Once finalised prototype manufacturing and experimental testing will be conducted, and the results used to improve on the COPV design.

Keywords: composite overwrapped pressure vessel, netting analysis, design, American Society of Mechanical Engineers section x, fiber-reinforced, hydrogen storage

Procedia PDF Downloads 247

Authors: Virginie Hergault, Siham Chebbah, Bertrand Frere

Abstract:

Following in-house objectives, Central laboratory of Paris police Prefecture conducted a general review on models and Computational Fluid Dynamics (CFD) codes used to simulate pollutant dispersion in the atmosphere. Starting from that review and considering main features of Large Eddy Simulation, Central Laboratory Of Paris Police Prefecture (LCPP) postulates that the Fire Dynamics Simulator (FDS) model, from National Institute of Standards and Technology (NIST), should be well suited for air pollutant dispersion modeling. This paper focuses on the implementation and the evaluation of FDS in the frame of the European COST ES1006 Action. This action aimed at quantifying the performance of modeling approaches. In this paper, the CUTE dataset carried out in the city of Hamburg, and its mock-up has been used. We have performed a comparison of FDS results with wind tunnel measurements from CUTE trials on the one hand, and, on the other, with the models results involved in the COST Action. The most time-consuming part of creating input data for simulations is the transfer of obstacle geometry information to the format required by SDS. Thus, we have developed Python codes to convert automatically building and topographic data to the FDS input file. In order to evaluate the predictions of FDS with observations, statistical performance measures have been used. These metrics include the fractional bias (FB), the normalized mean square error (NMSE) and the fraction of predictions within a factor of two of observations (FAC2). As well as the CFD models tested in the COST Action, FDS results demonstrate a good agreement with measured concentrations. Furthermore, the metrics assessment indicate that FB and NMSE meet the tolerance acceptable.

Keywords: numerical simulations, atmospheric dispersion, cost ES1006 action, CFD model, cute experiments, wind tunnel data, numerical results

Procedia PDF Downloads 136

Authors: Baladitya Swaika, Rahul Khatry

Abstract:

Computer Adaptive Tests (CATs) is one of the most efficient ways for testing the cognitive abilities of students. CATs are based on Item Response Theory (IRT) which is based on item selection and ability estimation using statistical methods of maximum information selection/selection from posterior and maximum-likelihood (ML)/maximum a posteriori (MAP) estimators respectively. This study aims at combining both classical and Bayesian approaches to IRT to create a dataset which is then fed to a neural network which automates the process of ability estimation and then comparing it to traditional CAT models designed using IRT. This study uses python as the base coding language, pymc for statistical modelling of the IRT and scikit-learn for neural network implementations. On creation of the model and on comparison, it is found that the Neural Network based model performs 7-10% worse than the IRT model for score estimations. Although performing poorly, compared to the IRT model, the neural network model can be beneficially used in back-ends for reducing time complexity as the IRT model would have to re-calculate the ability every-time it gets a request whereas the prediction from a neural network could be done in a single step for an existing trained Regressor. This study also proposes a new kind of framework whereby the neural network model could be used to incorporate feature sets, other than the normal IRT feature set and use a neural network’s capacity of learning unknown functions to give rise to better CAT models. Categorical features like test type, etc. could be learnt and incorporated in IRT functions with the help of techniques like logistic regression and can be used to learn functions and expressed as models which may not be trivial to be expressed via equations. This kind of a framework, when implemented would be highly advantageous in psychometrics and cognitive assessments. This study gives a brief overview as to how neural networks can be used in adaptive testing, not only by reducing time-complexity but also by being able to incorporate newer and better datasets which would eventually lead to higher quality testing.

Keywords: computer adaptive tests, item response theory, machine learning, neural networks

Procedia PDF Downloads 176

Authors: Fidelia A. Orji, Julita Vassileva

Abstract:

This research aims to develop machine learning models for students' academic performance and study strategy prediction, which could be generalized to all courses in higher education. Key learning attributes (intrinsic, extrinsic, autonomy, relatedness, competence, and self-esteem) used in building the models are chosen based on prior studies, which revealed that the attributes are essential in students’ learning process. Previous studies revealed the individual effects of each of these attributes on students’ learning progress. However, few studies have investigated the combined effect of the attributes in predicting student study strategy and academic performance to reduce the dropout rate. To bridge this gap, we used Scikit-learn in python to build five machine learning models (Decision Tree, K-Nearest Neighbour, Random Forest, Linear/Logistic Regression, and Support Vector Machine) for both regression and classification tasks to perform our analysis. The models were trained, evaluated, and tested for accuracy using 924 university dentistry students' data collected by Chilean authors through quantitative research design. A comparative analysis of the models revealed that the tree-based models such as the random forest (with prediction accuracy of 94.9%) and decision tree show the best results compared to the linear, support vector, and k-nearest neighbours. The models built in this research can be used in predicting student performance and study strategy so that appropriate interventions could be implemented to improve student learning progress. Thus, incorporating strategies that could improve diverse student learning attributes in the design of online educational systems may increase the likelihood of students continuing with their learning tasks as required. Moreover, the results show that the attributes could be modelled together and used to adapt/personalize the learning process.

Keywords: classification models, learning strategy, predictive modeling, regression models, student academic performance, student motivation, supervised machine learning

Procedia PDF Downloads 130

Authors: Adolfo Martucci, Iulian Mihai

Abstract:

By means of the expansion through a Conical Nozzle and the low pressure inside the Test Chamber, a large hypersonic stable flow takes place for a duration of up to 30 minutes. Downstream the Test Chamber, the diffuser has the function of reducing the flow velocity to subsonic values, and as a consequence, the temperature increases again. In order to cool down the flow, a heat exchanger is present at the end of the diffuser. The Vacuum System generates the necessary vacuum conditions for the correct hypersonic flow generation, and the DeNOx system, which follows the Vacuum System, reduces the nitrogen oxide concentrations created inside the plasma flow behind the limits imposed by Italian law. This very large, powerful, and complex facility allows researchers and engineers to reproduce entire re-entry trajectories of space vehicles into the atmosphere. One of the most important parameters for a hypersonic flowfield representative of re-entry conditions is the specific total enthalpy. This is the whole energy content of the fluid, and it represents how severe could be the conditions around a spacecraft re-entering from a space mission or, in our case, inside a hypersonic wind tunnel. It is possible to reach very high values of enthalpy (up to 45 MJ/kg) that, together with the large allowable size of the models, represent huge possibilities for making on-ground experiments regarding the atmospheric re-entry field. The maximum nozzle exit section diameter is 1950 mm, where values of Mach number very much higher than 1 can be reached. The specific total enthalpy is evaluated by means of a number of measurements, each of them concurring with its value and its uncertainty. The scope of the present paper is the evaluation of the sensibility of the uncertainty of the specific total enthalpy versus all the parameters and measurements involved. The sensors that, if improved, could give the highest advantages have so been individuated. Several simulations in Python with the METAS library and by means of Monte Carlo simulations are presented together with the obtained results and discussions about them.

Keywords: hypersonic, uncertainty, enthalpy, simulations

Procedia PDF Downloads 98

Authors: G. St-Pierre-Lemieux, E. Askari Mahvelati, D. Groleau, P. Proulx

Abstract:

Mathematical modeling has become an important tool for the study of foam behavior. Computational Fluid Dynamic (CFD) can be used to investigate the behavior of foam around foam breakers to better understand the mechanisms leading to the ‘destruction’ of foam. The focus of this investigation was the simple cone foam breaker, whose performance has been identified in numerous studies. While the optimal pumping angle is known from the literature, the contribution of pressure drop, shearing, and centrifugal forces to the foam syneresis are subject to speculation. This work provides a screening of those factors against changes in the cone angle and foam rheology. The CFD simulation was made with the open source OpenFOAM toolkits on a full three-dimensional model discretized using hexahedral cells. The geometry was generated using a python script then meshed with blockMesh. The OpenFOAM Volume Of Fluid (VOF) method was used (interFOAM) to obtain a detailed description of the interfacial forces, and the model k-omega SST was used to calculate the turbulence fields. The cone configuration allows the use of a rotating wall boundary condition. In each case, a pair of immiscible fluids, foam/air or water/air was used. The foam was modeled as a shear thinning (Herschel-Buckley) fluid. The results were compared to our measurements and to results found in the literature, first by computing the pumping rate of the cone, and second by the liquid break-up at the exit of the cone. A 3D printed version of the cones submerged in foam (shaving cream or soap solution) and water, at speeds varying between 400 RPM and 1500 RPM, was also used to validate the modeling results by calculating the torque exerted on the shaft. While most of the literature is focusing on cone behavior using Newtonian fluids, this works explore its behavior in shear thinning fluid which better reflects foam apparent rheology. Those simulations bring new light on the cone behavior within the foam and allow the computation of shearing, pressure, and velocity of the fluid, enabling to better evaluate the efficiency of the cones as foam breakers. This study contributes to clarify the mechanisms behind foam breaker performances, at least in part, using modern CFD techniques.

Keywords: bioreactor, CFD, foam breaker, foam mitigation, OpenFOAM

Procedia PDF Downloads 206

Authors: Monica Gonzalez Machorro

Abstract:

Dementia is hard to diagnose because of the lack of early physical symptoms. Early dementia recognition is key to improving the living condition of patients. Speech technology is considered a valuable biomarker for this challenge. Recent works have utilized conventional acoustic features and machine learning methods to detect dementia in speech. BERT-like classifiers have reported the most promising performance. One constraint, nonetheless, is that these studies are either based on human transcripts or on transcripts produced by automatic speech recognition (ASR) systems. This research contribution is to explore a method that does not require transcriptions to detect early Alzheimer’s disease (AD) and mild cognitive impairment (MCI). This is achieved by fine-tuning a pre-trained ASR model for the downstream early AD and MCI tasks. To do so, a subset of the thoroughly studied Pitt Corpus is customized. The subset is balanced for class, age, and gender. Data processing also involves cropping the samples into 10-second segments. For comparison purposes, a baseline model is defined by training and testing a Random Forest with 20 extracted acoustic features using the librosa library implemented in Python. These are: zero-crossing rate, MFCCs, spectral bandwidth, spectral centroid, root mean square, and short-time Fourier transform. The baseline model achieved a 58% accuracy. To fine-tune HuBERT as a classifier, an average pooling strategy is employed to merge the 3D representations from audio into 2D representations, and a linear layer is added. The pre-trained model used is ‘hubert-large-ls960-ft’. Empirically, the number of epochs selected is 5, and the batch size defined is 1. Experiments show that our proposed method reaches a 69% balanced accuracy. This suggests that the linguistic and speech information encoded in the self-supervised ASR-based model is able to learn acoustic cues of AD and MCI.

Keywords: automatic speech recognition, early Alzheimer’s recognition, mild cognitive impairment, speech impairment

Procedia PDF Downloads 127

Authors: Zafer Yüce, Paşa Yayla, Alev Taşkın

Abstract:

There are heaps of analytical methods to estimate the crack growth life of a component. Soft computing methods have an increasing trend in predicting fatigue life. Their ability to build complex relationships and capability to handle huge amounts of data are motivating researchers and industry professionals to employ them for challenging problems. This study focuses on soft computing methods, especially random forest regressors and artificial neural networks with hyperparameter optimization algorithms such as grid search and random grid search, to estimate the crack growth life of an aircraft wing splice joint under variable amplitude loading. TensorFlow and Scikit-learn libraries of Python are used to build the machine learning models for this study. The material considered in this work is 7050-T7451 aluminum, which is commonly preferred as a structural element in the aerospace industry, and regarding the crack type; corner crack is used. A finite element model is built for the joint to calculate fastener loads and stresses on the structure. Since finite element model results are validated with analytical calculations, findings of the finite element model are fed to AFGROW software to calculate analytical crack growth lives. Based on Fighter Aircraft Loading Standard for Fatigue (FALSTAFF), 90 unique fatigue loading spectra are developed for various load levels, and then, these spectrums are utilized as inputs to the artificial neural network and random forest regression models for predicting crack growth life. Finally, the crack growth life predictions of the machine learning models are compared with analytical calculations. According to the findings, a good correlation is observed between analytical and predicted crack growth lives.

Keywords: aircraft, fatigue, joint, life, optimization, prediction.

Procedia PDF Downloads 178

Authors: Evan Lowhorn, Rocio Alba-Flores

Abstract:

The purpose of this work is to develop a method for classifying hand signals and using the output in a drone control algorithm. To achieve this, methods based on Convolutional Neural Networks (CNN) were applied. CNN's are a subset of deep learning, which allows grid-like inputs to be processed and passed through a neural network to be trained for classification. This type of neural network allows for classification via imaging, which is less intrusive than previous methods using biosensors, such as EMG sensors. Classification CNN's operate purely from the pixel values in an image; therefore they can be used without additional exteroceptive sensors. A development bench was constructed using a desktop computer connected to a high-definition webcam mounted on a scissor arm. This allowed the camera to be pointed downwards at the desk to provide a constant solid background for the dataset and a clear detection area for the user. A MATLAB script was created to automate dataset image capture at the development bench and save the images to the desktop. This allowed the user to create their own dataset of 12,000 images within three hours. These images were evenly distributed among seven classes. The defined classes include forward, backward, left, right, idle, and land. The drone has a popular flip function which was also included as an additional class. To simplify control, the corresponding hand signals chosen were the numerical hand signs for one through five for movements, a fist for land, and the universal “ok” sign for the flip command. Transfer learning with PyTorch (Python) was performed using a pre-trained 18-layer residual learning network (ResNet-18) to retrain the network for custom classification. An algorithm was created to interpret the classification and send encoded messages to a Ryze Tello drone over its 2.4 GHz Wi-Fi connection. The drone’s movements were performed in half-meter distance increments at a constant speed. When combined with the drone control algorithm, the classification performed as desired with negligible latency when compared to the delay in the drone’s movement commands.

Keywords: classification, computer vision, convolutional neural networks, drone control

Procedia PDF Downloads 212

Authors: Tesfaye Mengistu

Abstract:

Power cables play a crucial role in the transmission and distribution of electrical energy. As the electricity generation, transmission, distribution, and storage systems become smarter, there is a growing emphasis on incorporating intelligent approaches to ensure the reliability of power cables. Various types of electrical cables are employed for transmitting and distributing electrical energy, with cross-linked polyethylene (XLPE) cables being widely utilized due to their exceptional electrical and mechanical properties. However, insulation defects can occur in XLPE cables due to subpar manufacturing techniques during production and cable joint installation. To address this issue, experts have proposed different methods for monitoring XLPE cables. Some suggest the use of interdigital capacitive (IDC) technology for online monitoring, while others propose employing continuous wave (CW) terahertz (THz) imaging systems to detect internal defects in XLPE plates used for power cable insulation. In this study, we have developed models that employ a custom dataset collected locally to classify the physical safety status of individual power cables. Our models aim to replace physical inspections with computer vision and image processing techniques to classify defective power cables from non-defective ones. The implementation of our project utilized the Python programming language along with the TensorFlow package and a convolutional neural network (CNN). The CNN-based algorithm was specifically chosen for power cable defect classification. The results of our project demonstrate the effectiveness of CNNs in accurately classifying power cable defects. We recommend the utilization of similar or additional datasets to further enhance and refine our models. Additionally, we believe that our models could be used to develop methodologies for detecting power cable defects from live video feeds. We firmly believe that our work makes a significant contribution to the field of power cable inspection and maintenance. Our models offer a more efficient and cost-effective approach to detecting power cable defects, thereby improving the reliability and safety of power grids.

Keywords: artificial intelligence, computer vision, defect detection, convolutional neural net

Procedia PDF Downloads 114

Authors: Wisam H. Muragaa, Kamaruzzaman Seman, Mohd Fadzli Marhusin

Abstract:

Software Defined Networking (SDN) is a new norm of networks. It is designed to facilitate the way of managing, measuring, debugging and controlling the network dynamically, and to make it suitable for the modern applications. Generally, measurement methods can be divided into two categories: Active and passive methods. Active measurement method is employed to inject test packets into the network in order to monitor their behaviour (ping tool as an example). Meanwhile the passive measurement method is used to monitor the traffic for the purpose of deriving measurement values. The measurement methods, both active and passive, are useful for the collection of traffic statistics, and monitoring of the network traffic. Although there has been a work focusing on measuring traffic statistics in SDN environment, it was only meant for measuring packets and bytes rates for non-web traffic. In this study, a feasible method will be designed to measure the number of packets and bytes in a certain time, and facilitate obtaining statistics for both web traffic and non-web traffic. Web traffic refers to HTTP requests that use application layer; while non-web traffic refers to ICMP and TCP requests. Thus, this work is going to be more comprehensive than previous works. With a developed module on POX OpenFlow controller, information will be collected from each active flow in the OpenFlow switch, and presented on Command Line Interface (CLI) and wireshark interface. Obviously, statistics that will be displayed on CLI and on wireshark interfaces include type of protocol, number of bytes and number of packets, among others. Besides, this module will show the number of flows added to the switch whenever traffic is generated from and to hosts in the same statistics list. In order to carry out this work effectively, our Python module will send a statistics request message to the switch requesting its current ports and flows statistics in every five seconds; while the switch will reply with the required information in a message called statistics reply message. Thus, POX controller will be notified and updated with any changes could happen in the entire network in a very short time. Therefore, our aim of this study is to prepare a list for the important statistics elements that are collected from the whole network, to be used for any further researches; particularly, those that are dealing with the detection of the network attacks that cause a sudden rise in the number of packets and bytes like Distributed Denial of Service (DDoS).

Keywords: mininet, OpenFlow, POX controller, SDN

Procedia PDF Downloads 237

Authors: Miguel Nino

Abstract:

When considering the challenges of a changing and demanding globalizing world, it is important to reflect on how university students will be prepared for the realities of internationalization, marketization and intercultural conversation. The present study is an interdisciplinary program designed to respond to the needs of the global community. The proposal bridges the humanities and science through three different fields: Languages, graphic design and computer science, specifically, fundamentals of programming such as python, java script and software animation. Therefore, the goal of the four year program is twofold: First, enable students for intercultural communication between English and other languages such as Spanish, Mandarin, French or German. Second, students will acquire knowledge in practical software and relevant employable skills to collaborate in assisted computer projects that most probable will require essential programing background in interpreted or compiled languages. In order to become inclusive and constructivist, the cognitive linguistics approach is suggested for the three different fields, particularly for languages that rely on the traditional method of repetition. This methodology will help students develop their creativity and encourage them to become independent problem solving individuals, as languages enhance their common ground of interaction for culture and technology. Participants in this course of study will be evaluated in their second language acquisition at the Intermediate-High level. For graphic design and computer science students will apply their creative digital skills, as well as their critical thinking skills learned from the cognitive linguistics approach, to collaborate on a group project design to find solutions for media web design problems or marketing experimentation for a company or the community. It is understood that it will be necessary to apply programming knowledge and skills to deliver the final product. In conclusion, the program equips students with linguistics knowledge and skills to be competent in intercultural communication, where English, the lingua franca, remains the medium for marketing and product delivery. In addition to their employability, students can expand their knowledge and skills in digital humanities, computational linguistics, or increase their portfolio in advertising and marketing. These students will be the global human capital for the competitive globalizing community.

Keywords: curriculum, international, languages, technology

Procedia PDF Downloads 443

Authors: Emigdio Mendoza, Patricia Fernandez, Cristian Gomez

Abstract:

Magnesium alloys have generated great interest for several industrial applications because their high specific strength and low density make them a very attractive alternative for the manufacture of various components; however, these alloys present a limitation with their hexagonal crystal structure that limits the deformation mechanisms at room temperature likewise the molding components alternatives, it is for this reason that severe plastic deformation processes have taken a huge relevance recently because these, allow high deformation rates to be applied that induce microstructural changes where the deficiency in the sliding systems is compensated with crystallographic grains reorientations or crystal twinning. The present study reports a statistical analysis of process temperature, number of passes and shear angle with respect to the shear stress in severe plastic deformation process denominated 'Equal Channel Angular Sheet Drawing (ECASD)' applied to the magnesium alloy AZ31B through Python Statsmodels libraries, additionally a Post-Hoc range test is performed using the Tukey statistical test. Statistical results show that each variable has a p-value lower than 0.05, which allows comparing the average values of shear stresses obtained, which are in the range of 7.37 MPa to 12.23 MPa, lower values in comparison to others severe plastic deformation processes reported in the literature, considering a value of 157.53 MPa as the average creep stress for AZ31B alloy. However, a higher stress level is required when the sheets are processed using a shear angle of 150°, due to a higher level of adjustment applied for the shear die of 150°. Temperature and shear passes are important variables as well, but there is no significant impact on the level of stress applied during the ECASD process. In the processing of AZ31B magnesium alloy sheets, ECASD technique is evidenced as a viable alternative in the modification of the elasto-plastic properties of this alloy, promoting the weakening of the basal texture, which means, a better response to deformation, whereby, during the manufacture of parts by drawing or stamping processes the formation of cracks on the surface can be reduced, presenting an adequate mechanical performance.

Keywords: plastic deformation, strain, sheet drawing, magnesium

Procedia PDF Downloads 110

Authors: Habtamu Ayenew Asegie

Abstract:

Wealth, as opposed to income or consumption, implies a more stable and permanent status. Due to natural and human-made difficulties, households' economies will be diminished, and their well-being will fall into trouble. Hence, governments and humanitarian agencies offer considerable resources for poverty and malnutrition reduction efforts. One key factor in the effectiveness of such efforts is the accuracy with which low-income or poor populations can be identified. As a result, this study aims to predict a household’s wealth status using ensemble Machine learning (ML) algorithms. In this study, design science research methodology (DSRM) is employed, and four ML algorithms, Random Forest (RF), Adaptive Boosting (AdaBoost), Light Gradient Boosted Machine (LightGBM), and Extreme Gradient Boosting (XGBoost), have been used to train models. The Ethiopian Demographic and Health Survey (EDHS) dataset is accessed for this purpose from the Central Statistical Agency (CSA)'s database. Various data pre-processing techniques were employed, and the model training has been conducted using the scikit learn Python library functions. Model evaluation is executed using various metrics like Accuracy, Precision, Recall, F1-score, area under curve-the receiver operating characteristics (AUC-ROC), and subjective evaluations of domain experts. An optimal subset of hyper-parameters for the algorithms was selected through the grid search function for the best prediction. The RF model has performed better than the rest of the algorithms by achieving an accuracy of 96.06% and is better suited as a solution model for our purpose. Following RF, LightGBM, XGBoost, and AdaBoost algorithms have an accuracy of 91.53%, 88.44%, and 58.55%, respectively. The findings suggest that some of the features like ‘Age of household head’, ‘Total children ever born’ in a family, ‘Main roof material’ of their house, ‘Region’ they lived in, whether a household uses ‘Electricity’ or not, and ‘Type of toilet facility’ of a household are determinant factors to be a focal point for economic policymakers. The determinant risk factors, extracted rules, and designed artifact achieved 82.28% of the domain expert’s evaluation. Overall, the study shows ML techniques are effective in predicting the wealth status of households.

Keywords: ensemble machine learning, households wealth status, predictive model, wealth status prediction

Procedia PDF Downloads 43

Authors: Hany Elsaid Fawaz Abdallah

Abstract:

This study presents a new artificial neural network(ANN) model to predict the Nusselt Number and pressure drop for the turbulent flow in a triangular corrugated plate heat exchanger for forced air and turbulent water flow. An experimental investigation was performed to create a new dataset for the Nusselt Number and pressure drop values in the following range of dimensionless parameters: The plate corrugation angles (from 0° to 60°), the Reynolds number (from 10000 to 40000), pitch to height ratio (from 1 to 4), and Prandtl number (from 0.7 to 200). Based on the ANN performance graph, the three-layer structure with {12-8-6} hidden neurons has been chosen. The training procedure includes back-propagation with the biases and weight adjustment, the evaluation of the loss function for the training and validation dataset and feed-forward propagation of the input parameters. The linear function was used at the output layer as the activation function, while for the hidden layers, the rectified linear unit activation function was utilized. In order to accelerate the ANN training, the loss function minimization may be achieved by the adaptive moment estimation algorithm (ADAM). The ‘‘MinMax’’ normalization approach was utilized to avoid the increase in the training time due to drastic differences in the loss function gradients with respect to the values of weights. Since the test dataset is not being used for the ANN training, a cross-validation technique is applied to the ANN network using the new data. Such procedure was repeated until loss function convergence was achieved or for 4000 epochs with a batch size of 200 points. The program code was written in Python 3.0 using open-source ANN libraries such as Scikit learn, TensorFlow and Keras libraries. The mean average percent error values of 9.4% for the Nusselt number and 8.2% for pressure drop for the ANN model have been achieved. Therefore, higher accuracy compared to the generalized correlations was achieved. The performance validation of the obtained model was based on a comparison of predicted data with the experimental results yielding excellent accuracy.

Keywords: artificial neural networks, corrugated channel, heat transfer enhancement, Nusselt number, pressure drop, generalized correlations

Procedia PDF Downloads 90

Authors: Elcin Timur Cakmak, Ayse Oguzlar

Abstract:

This study presents the results of the tweets sent by Twitter users on social media about the Russia-Ukraine war by bigram and trigram methods. On February 24, 2022, Russian President Vladimir Putin declared a military operation against Ukraine, and all eyes were turned to this war. Many people living in Russia and Ukraine reacted to this war and protested and also expressed their deep concern about this war as they felt the safety of their families and their futures were at stake. Most people, especially those living in Russia and Ukraine, express their views on the war in different ways. The most popular way to do this is through social media. Many people prefer to convey their feelings using Twitter, one of the most frequently used social media tools. Since the beginning of the war, it is seen that there have been thousands of tweets about the war from many countries of the world on Twitter. These tweets accumulated in data sources are extracted using various codes for analysis through Twitter API and analysed by Python programming language. The aim of the study is to find the word sequences in these tweets by the n-gram method, which is known for its widespread use in computational linguistics and natural language processing. The tweet language used in the study is English. The data set consists of the data obtained from Twitter between February 24, 2022, and April 24, 2022. The tweets obtained from Twitter using the #ukraine, #russia, #war, #putin, #zelensky hashtags together were captured as raw data, and the remaining tweets were included in the analysis stage after they were cleaned through the preprocessing stage. In the data analysis part, the sentiments are found to present what people send as a message about the war on Twitter. Regarding this, negative messages make up the majority of all the tweets as a ratio of %63,6. Furthermore, the most frequently used bigram and trigram word groups are found. Regarding the results, the most frequently used word groups are “he, is”, “I, do”, “I, am” for bigrams. Also, the most frequently used word groups are “I, do, not”, “I, am, not”, “I, can, not” for trigrams. In the machine learning phase, the accuracy of classifications is measured by Classification and Regression Trees (CART) and Naïve Bayes (NB) algorithms. The algorithms are used separately for bigrams and trigrams. We gained the highest accuracy and F-measure values by the NB algorithm and the highest precision and recall values by the CART algorithm for bigrams. On the other hand, the highest values for accuracy, precision, and F-measure values are achieved by the CART algorithm, and the highest value for the recall is gained by NB for trigrams.

Keywords: classification algorithms, machine learning, sentiment analysis, Twitter

Procedia PDF Downloads 75

Authors: Luzie Krings, Sven Liebehentze, Maximilian Gehring, Uwe Rüppel

Abstract:

The integration of renewable energy into the energy grid is essential for decarbonization, and leveraging electrified vehicles (EVs) as mobile storage units offers a pathway to address grid challenges. The decentralized nature of EVs and the intermittency of renewable energy sources, such as photovoltaic (PV) and wind power, complicate grid stability. Vehicle-to-Grid (V2G) technology presents a promising solution, enabling EVs to support grid stability through services like redispatch, congestion mitigation, and enhanced renewable energy utilization. Freight transport, contributing 38% of transport emissions, holds significant potential as its aggregated energy storage capacity can stabilize the grid and optimize renewable energy integration. This study introduces a risk-averse optimization model for marketing EV flexibilities in Germany’s energy markets, with a strong focus on improving grid stability and maximizing renewable energy potential. Using a linear optimization framework, the model incorporates technical, regulatory, and operational constraints to simulate EV fleets as scalable energy storage solutions. The integration of proprietary PV and wind energy systems is also modeled to evaluate benefits. Benchmarks compare bidirectional charging with unidirectional charging under dynamic tariffs. The methodology employs the Python-based energypilot tool to optimize participation in Day-Ahead, Intraday, and Redispatch markets, accounting for trading conditions and temporal offsets. Results demonstrate that redispatch utilization substantially supports grid stability, while bidirectional charging increased renewable energy integration by 15% and economic benefits by 20%. Longer charging cycles offered greater financial returns compared to fragmented cycles, emphasizing the potential of fleets with extended idle periods for storing renewable energy. This research highlights the critical role of EVs in stabilizing the grid and utilizing renewable energy effectively by expanding storage capacity. The optimization framework addresses key challenges in energy trading, offering a transferable methodology for broader energy storage applications. This supports the transition to a sustainable energy system by improving environmental outcomes and economic incentives.

Keywords: Electric Vehicles, Energy Grid, Energy Storages, Redispatch

Procedia PDF Downloads 11