Search results for: virtual Machine

Authors: Edward Guillén, Jhordany Rodriguez, Rafael Páez

Abstract:

Anomaly detection techniques have been focused on two main components: data extraction and selection and the second one is the analysis performed over the obtained data. The goal of this paper is to analyze the influence that each of these components has over the system performance by evaluating detection over network scenarios with different setups. The independent variables are as follows: the number of system inputs, the way the inputs are codified and the complexity of the analysis techniques. For the analysis, some approaches of artificial neural networks are implemented with different number of layers. The obtained results show the influence that each of these variables has in the system performance.

Keywords: network intrusion detection, machine learning, artificial neural network, anomaly detection module

Procedia PDF Downloads 343

Authors: Tomasz Robert Kuczerski

Abstract:

The high complexity of the algorithm of the autonomous tandem detonator system creates an optimization problem due to the parallel operation of several machine states of the system. Many years of experience and classic analyses have led to a partially optimized model. Limitations on the energy resources of this class of autonomous systems make it necessary to search for more effective methods of optimisation. The use of the Quantum Approximate Optimization Algorithm (QAOA) in these studies shows the most promising results. With the help of multiple evaluations of several qubit quantum circuits, proper results of variable parameter optimization were obtained. In addition, it was observed that the increase in the number of assessments does not result in further efficient growth due to the increasing complexity of optimising variables. The tests confirmed the effectiveness of the QAOA optimization method.

Keywords: algorithm analysis, autonomous system, quantum optimization, tandem detonator

Procedia PDF Downloads 92

Authors: Florin Leon

Abstract:

This paper describes a simulator of traffic scenarios tailored to facilitate autonomous driving model training for urban environments. With the rising prominence of self-driving vehicles, the need for diverse datasets is very important. The proposed simulator provides a flexible framework that allows the generation of custom scenarios needed for the validation and enhancement of trajectory prediction algorithms. Its controlled yet dynamic environment addresses the challenges associated with real-world data acquisition and ensures adaptability to diverse driving scenarios. By providing an adaptable solution for scenario creation and algorithm testing, this tool proves to be a valuable resource for advancing autonomous driving technology that aims to ensure safe and efficient self-driving vehicles.

Keywords: autonomous driving, car simulator, machine learning, model training, urban simulation environment

Procedia PDF Downloads 62

Authors: Lazlo Fauth, Andreas Ligocki

Abstract:

For cost-effective use of harvesters, expensive repairs and unplanned downtimes must be reduced as far as possible. The predictive detection of failing systems and the calculation of intelligent service intervals, necessary to avoid these factors, require in-depth knowledge of the machines' behavior. Such know-how needs permanent monitoring of the machine state from different technical perspectives. In this paper, three approaches will be presented as they are currently pursued in the publicly funded project PreForst at Ostfalia University of Applied Sciences. These include the intelligent linking of workshop and service data, sensors on the harvester, and a special online hydraulic oil condition monitoring system. Furthermore the paper shows potentials as well as challenges for the use of these data in the conception of a predictive maintenance system.

Keywords: predictive maintenance, condition monitoring, forest harvesting, forest engineering, oil data, hydraulic data

Procedia PDF Downloads 145

Authors: Raja V., Chokkalingam B.

Abstract:

For rough cutting, in many industries and educational institutions using carbon steels or HSS single point cutting tools in center lathe machine. In power hacksaw blades, only the cutter teeth region used to parting off the given material. The portions other than the teeth can be used as a single point cutting tool for rough turning and facing on soft materials. The hardness and Tensile strength of this used Power hacksaw blade is almost same as conventional cutting tools. In this paper, the effect of power hacksaw blades over conventional tool has been compared. Thickness of the blade (1.6 mm) is very small compared to its length and width. Hence, a special tool holding device is designed to hold the tool.

Keywords: hardness, high speed steels, power hacksaw blade, tensile strength

Procedia PDF Downloads 457

Authors: Antti Nurminen, Avleen Malhi

Abstract:

Significant progress in intelligent systems in combination with exceedingly wide application domains having machine learning as the core technology are usually opaque, non-intuitive, and commonly complex for human users. We use innovative IoT technology which monitors and analyzes moisture, humidity, luminosity and temperature levels to assist end users for optimization of environmental conditions for their houseplants. For plant health monitoring, we construct a system yielding the Normalized Difference Vegetation Index (NDVI), supported by visual validation by users. We run the system for a selected plant, basil, in varying environmental conditions to cater for typical home conditions, and bootstrap our AI with the acquired data. For end users, we implement a web based user interface which provides both instructions and explanations.

Keywords: explainable artificial intelligence, intelligent agent, IoT, NDVI

Procedia PDF Downloads 163

Authors: M. Peña Aguilar Juan, Ríos Hernández Ezequiel, R. Valencia Luis

Abstract:

For certification, certain goods of public interest, such as medicines and food, it is required the preparation and delivery of a dossier. For its elaboration, legal and administrative knowledge must be taken, as well as organization of the documents of the process, and an order that allows the file verification. Therefore, a virtual platform was developed to support the process of management and elaboration of the dossier, providing accessibility to the information and interfaces that allow the user to know the status of projects. The development of dossier system on the cloud allows the inclusion of the technical requirements for the software management, including the validation and the manufacturing in the field industry. The platform guides and facilitates the dossier elaboration (report, file or history), considering Mexican legislation and regulations, it also has auxiliary tools for its management. This technological alternative provides organization support for documents and accessibility to the information required to specify the successful development of a dossier. The platform divides into the following modules: System control, catalog, dossier and enterprise management. The modules are designed per the structure required in a dossier in those areas. However, the structure allows for flexibility, as its goal is to become a tool that facilitates and does not obstruct processes. The architecture and development of the software allows flexibility for future work expansion to other fields, this would imply feeding the system with new regulations.

Keywords: electronic dossier, cloud management software, pharmaceutical industry, sanitary registration

Procedia PDF Downloads 297

Authors: Liang Zhou, Qinke Sun

Abstract:

Taking the Yangtze River Economic Belt as an example, using long-term nighttime lighting data from DMSP/OLS from 1992 to 2012, support vector machine classification (SVM) was used to quantitatively extract urban built-up areas of economic belts, and spatial analysis of expansion intensity index, standard deviation ellipse, etc. was introduced. The model conducts detailed and in-depth discussions on the strength, direction, and type of the expansion of the middle and lower reaches of the economic belt and the key node cities. The results show that: (1) From 1992 to 2012, the built-up areas of the major cities in the Yangtze River Valley showed a rapid expansion trend. The built-up area expanded by 60,392 km², and the average annual expansion rate was 31%, that is, from 9615 km² in 1992 to 70007 km² in 2012. The spatial gradient analysis of the watershed shows that the expansion of urban built-up areas in the middle and lower reaches of the river basin takes Shanghai as the leading force, and the 'bottom-up' model shows an expanding pattern of 'upstream-downstream-middle-range' declines. The average annual rate of expansion is 36% and 35%, respectively. 17% of which the midstream expansion rate is about 50% of the upstream and downstream. (2) The analysis of expansion intensity shows that the urban expansion intensity in the Yangtze River Basin has generally shown an upward trend, the downstream region has continued to rise, and the upper and middle reaches have experienced different amplitude fluctuations. To further analyze the strength of urban expansion at key nodes, Chengdu, Chongqing, and Wuhan in the upper and middle reaches maintain a high degree of consistency with the intensity of regional expansion. Node cities with Shanghai as the core downstream continue to maintain a high level of expansion. (3) The standard deviation ellipse analysis shows that the overall center of gravity of the Yangtze River basin city is located in Anqing City, Anhui Province, and it showed a phenomenon of reciprocating movement from 1992 to 2012. The nighttime standard deviation ellipse distribution range increased from 61.96 km² to 76.52 km². The growth of the major axis of the ellipse was significantly larger than that of the minor axis. It had obvious east-west axiality, in which the nighttime lights in the downstream area occupied in the entire luminosity scale urban system leading position.

Keywords: urban space, support vector machine, spatial characteristics, night lights, Yangtze River Economic Belt

Procedia PDF Downloads 114

Authors: Han-Woong Ahn, In-Gun Kim, Hyun-Seok Hong, Dong-Woo Kang, Ju Lee

Abstract:

It is important to accurately identify machine parameters for direct vector control. To obtain the parameter values, traditional methods can be used such as no-load and rotor locked tests. However, there are many differences between values obtained from the traditional tests and actual values. In addition, there are drawbacks that additional equipment and cost are required for the experiment. Therefore, it is hard to temporary operation to estimate induction motor parameters. Therefore, this paper deals with the estimation algorithm of induction motor parameters without a motor operation and the measurement from additional equipment such as sensors and dynamometer. The validity and usefulness of the estimation algorithm considering inverter nonlinearity is verified by comparing the conventional method with the proposed method.

Keywords: induction motor, parameter, off-line estimation, inverter nonlinearity

Procedia PDF Downloads 530

Authors: Ramanjaneyulu Mogili, G.V.R.K. Acharyulu

Abstract:

Indian organized retail sector is growing at a faster pace and gaining popularity. Indian Brand Equity Foundation (IBEF) reveals that the current market size of Indian retail industry is about US$ 520 billion with for growth rate 14 to 15 percent annually by 2018 the Indian retail sector is likely to grow at a CAGR of 13% to reach a size of US$ 950 billion. Developments in Information Technology have enabled online Retail sector that empowers customers to order products, conduct transactions without the need to interact physically with the retailers. In recent years, the online shopping industry has gained popularity to the point where certain categories of customers would consider buying electronic products online rather than visiting the stores. Conventionally the physical location of a store is seen as a source of competitive advantage. Online Retailing service sites provide virtual shopping space to the customers. Online Retail services are gaining momentum in India, with internet penetration improving in the country and smartphones becoming affordable along with changing lifestyles and preferences of customers. Although online shoppers prefer the convenience and choice available in online shopping, certain issues raised due to the occurrence of service failure. The proposed study attempts to measure the service recovery and failure process of electronic goods in Indian retail channels.

Keywords: service recovery, customer satisfaction, e-shopping, service failure

Procedia PDF Downloads 228

Authors: Nikhil Byanna, Abdolghani Ebrahimi, Diego Klabjan

Abstract:

How does one objectively measure the performance of an individual offensive lineman in the NFL? The existing literature proposes various measures that rely on subjective assessments of game film, but has yet to develop an objective methodology to evaluate performance. Using a variety of statistics related to an offensive lineman’s performance, we develop a framework to objectively analyze the overall performance of an individual offensive lineman and determine specific linemen who are overvalued or undervalued relative to their salary. We identify eight players across the 2013-2014 and 2014-2015 NFL seasons that are considered to be overvalued or undervalued and corroborate the results with existing metrics that are based on subjective evaluation. To the best of our knowledge, the techniques set forth in this work have not been utilized in previous works to evaluate the performance of NFL players at any position, including offensive linemen.

Keywords: offensive lineman, player performance, NFL, machine learning

Procedia PDF Downloads 144

Authors: Akash Amalan, Rui Wang, Yanqi Qiao, Emmanouil Panaousis, Kaitai Liang

Abstract:

Federated learning is an emerging concept in the domain of distributed machine learning. This concept has enabled General Adversarial Networks (GANs) to benefit from the rich distributed training data while preserving privacy. However, in a non-IID setting, current federated GAN architectures are unstable, struggling to learn the distinct features, and vulnerable to mode collapse. In this paper, we propose an architecture MULTI-FLGAN to solve the problem of low-quality images, mode collapse, and instability for non-IID datasets. Our results show that MULTI-FLGAN is four times as stable and performant (i.e., high inception score) on average over 20 clients compared to baseline FLGAN.

Keywords: federated learning, generative adversarial network, inference attack, non-IID data distribution

Procedia PDF Downloads 158

Authors: K. E. Ch. Vidyasagar, M. Moghavvemi, T. S. S. T. Prabhat

Abstract:

Epilepsy is a global problem, and with seizures eluding even the smartest of diagnoses a requirement for automatic detection of the same using electroencephalogram (EEG) would have a huge impact in diagnosis of the disorder. Among a multitude of methods for automatic epilepsy detection, one should find the best method out, based on accuracy, for classification. This paper reasons out, and rationalizes, the best methods for classification. Accuracy is based on the classifier, and thus this paper discusses classifiers like quadratic discriminant analysis (QDA), classification and regression tree (CART), support vector machine (SVM), naive Bayes classifier (NBC), linear discriminant analysis (LDA), K-nearest neighbor (KNN) and artificial neural networks (ANN). Results show that ANN is the most accurate of all the above stated classifiers with 97.7% accuracy, 97.25% specificity and 98.28% sensitivity in its merit. This is followed closely by SVM with 1% variation in result. These results would certainly help researchers choose the best classifier for detection of epilepsy.

Keywords: classification, seizure, KNN, SVM, LDA, ANN, epilepsy

Procedia PDF Downloads 520

Authors: Leen Sulaimani, Maryam Hafiz, Naba Ali, Roba Alsharif

Abstract:

One of Amazon’s most crucial online systems is artificial intelligence. Amazon would not have a worldwide successful online store, an easy and secure way of payment, and other services if it weren’t for artificial intelligence and machine learning. Amazon uses AI to expand its operations and enhance them by upgrading the website daily; having a strong base of artificial intelligence in a worldwide successful business can improve marketing, decision-making, feedback, and more qualities. Aiming to have a rational AI system in one’s business should be the start of any process; that is why Amazon is fortunate that they keep taking care of the base of their business by using modern artificial intelligence, making sure that it is stable, reaching their organizational goals, and will continue to thrive more each and every day. Artificial intelligence is used daily in our current world and is still being amplified more each day to reach consumer satisfaction and company short and long-term goals.

Keywords: artificial intelligence, Amazon, business, customer, decision making

Procedia PDF Downloads 110

Authors: Titanilla Komenda, Viktorio Malisa

Abstract:

In this paper, the influence of diversity-related factors on the design of collaborative scenarios is analysed. Based on the evaluation, a framework for simulating human-robot-collaboration is presented that considers both human factors as well as the overall system performance. The implementation of the model is shown on a real-life scenario from industry and validated in terms of traceability, safety and physical limitations. By comparing scenarios that consider diversity with those only meeting system performance, an overall understanding of individually adapted human-robot-collaborative workspaces is reached. A diversity-related guideline for human-robot-collaborations provides a summary of the research and aids in optimizing future applications. Finally, limitations and future amendments of the model are discussed.

Keywords: diversity, human-machine system, human-robot collaboration, simulation

Procedia PDF Downloads 304

Authors: Hebberly Ahatlan

Abstract:

The process of digitizing energy systems involves transforming traditional energy infrastructure into interconnected, data-driven systems that enhance efficiency, sustainability, and responsiveness. As smart grids become increasingly integral to the efficient distribution and management of electricity from both fossil and renewable energy sources, the energy industry faces strategic challenges associated with digitalization and interoperability — particularly in the context of modern energy business models, such as virtual power plants (VPPs). The critical challenge in modern smart grids is to seamlessly integrate diverse technologies and systems, including virtualization, grid computing and service-oriented architecture (SOA), across the entire energy ecosystem. Achieving this requires addressing issues like semantic interoperability, IT/OT convergence, and digital asset scalability, all while ensuring security and risk management. This paper proposes a four-layer digitalization framework to tackle these challenges, encompassing persistent data protection, trusted key management, secure messaging, and authentication of IoT resources. Data assets generated through this framework enable AI systems to derive insights for improving smart grid operations, security, and revenue generation. Furthermore, this paper also proposes a Trusted Energy Interoperability Alliance as a universal guiding standard in the development of this digitalization framework to support more dynamic and interoperable energy markets.

Keywords: digitalization, IT/OT convergence, semantic interoperability, VPP, energy blockchain

Procedia PDF Downloads 183

Authors: Pauli P.Y. Lai

Abstract:

Students with Special Education Needs (SEN) face many challenges in learning. Various challenges include difficulty in handwriting, slow understanding and assimilation, difficulty in paying attention during class, and lack of communication skills. To engage students with Special Education Needs in class with general students, Blackboard Collaborate is used as a teaching and learning tool to deliver a lecture with interactive activities. Blackboard Collaborate provides a good platform to create and enhance active, collaborative and interactive learning experience whereby the SEN students can easily interact with their general peers and the instructor by using the features of drawing on a virtual whiteboard, file sharing, classroom chatter, breakout room, hand-raising feature, polling, etc. By integrating a blended learning approach with Blackboard Collaborate, the students with Special Education Needs could engage in interactive activities with ease in class. Our research aims at exploring and discovering the use of Blackboard Collaborate for inclusive education based on a qualitative design with in-depth interviews. Being served in a general education environment, three university students with different kinds of learning disabilities have participated in our study. All participants agreed that functions provided by Blackboard Collaborate have enhanced their learning experiences and helped them learn better. Their academic performances also showed that SEN students could perform well with the help of technology. This research studies different aspects of using Blackboard Collaborate to create an inclusive learning environment for SEN students.

Keywords: blackboard collaborate, enhanced learning experience, inclusive education, special education needs

Procedia PDF Downloads 134

Authors: J. S. Rudas, J. M. Gutiérrez Cabeza, A. Corz Rodríguez, L. M. Gómez, A. O. Toro

Abstract:

The prediction of the wear rate of rubbing pairs has attracted the interest of many researchers for years. It has been recently proposed that the sliding wear rate can be inferred from the calculation of the energy rate dissipated by the tribological pair. In this paper some of the dissipative mechanisms present in a pin-on-disc configuration are discussed and both analytical and numerical calculations are carried out. Three dissipative mechanisms were studied: First, the energy release due to temperature gradients within the solid; second, the heat flow from the solid to the environment, and third, the energy loss due to abrasive damage of the surface. The Finite Element Method was used to calculate the dynamics of heat transfer within the solid, with the aid of commercial software. Validation the FEM model was assisted by virtual and laboratory experimentation using different operating points (sliding velocity and geometry contact). The materials for the experiments were Ti6Al4V alloy and Tungsten Carbide (WC-Co). The results showed that the sliding wear rate has a linear relationship with the energy dissipation flow. It was also found that energy loss due to micro-cutting is relevant for the system. This mechanism changes if the sliding velocity and pin geometry are modified though the degradation coefficient continues to present a linear behavior. We found that the less relevant dissipation mechanism for all the cases studied is the energy release by temperature gradients in the solid.

Keywords: degradation, dissipative mechanism, dry sliding, entropy, friction, wear

Procedia PDF Downloads 502

Authors: Vinay Patil, Swapnil Kekade, Ashish Supare, Vinayak Pawar, Shital Jadhav, Rajkumar Singh

Abstract:

In recent days, bainitic steel is used in automobile and non-automobile sectors due to its high strength. Bainitic steel is difficult to machine because of its high hardness, hence in this paper machinability of bainitic steel is studied by using Taguchi design of experiments (DOE) approach. Convectional turning experiments were done by using L16 orthogonal array for three input parameters viz. cutting speed, depth of cut and feed. The Taguchi method is applied to study the performance characteristics of machining parameters with surface roughness (Ra), cutting force and tool wear rate. By using Taguchi analysis, optimized process parameters for best surface finish and minimum cutting forces were analyzed.

Keywords: conventional turning, Taguchi method, S/N ratio, bainitic steel machining

Procedia PDF Downloads 331

Authors: John Monro, Suman Mishra

Abstract:

Choosing foods for health benefits requires information that accurately represents the relative effectiveness of foods with respect to specific health end points, or with respect to responses leading to health outcomes. At present consumers must rely on nutrient composition data, and on health claims to guide them to healthy food choices. Nutrient information may be of limited usefulness because it does not reflect the effect of food structure and food component interactions – that is, whole food effects. Health claims demand stringent criteria that exclude most foods, even though most foods have properties through which they may contribute to positive health outcomes in a diet. In this presentation, we show how the functional efficacy of foods may be expressed in the same format as nutrients, with weight units, as virtual food components that allow a nutrition information panel to show not only what a food is, but also what it does. In the presentation, two body responses linked to well-being are considered – glycaemic response and colonic bulk – in order to illustrate the concept. We show how the nutrient information on available carbohydrates and dietary fibre values obtained by food analysis methods fail to provide information of the glycaemic potency or the colonic bulking potential of foods, because of failings in the methods and approach taken to food analysis. It is concluded that a category of food values that represent the functional efficacy of foods is required to accurately guide food choices for health.

Keywords: dietary fibre, glycaemic response, food values, food effects, health

Procedia PDF Downloads 502

Authors: Retno Supriyanti, Ays Rahmadian Subhi, Yogi Ramadhani, Haris B. Widodo

Abstract:

Alzheimer is one type of disease in the elderly that may occur in the world. The severity of the Alzheimer can be measured using a scale called Clinical Dementia Rating (CDR) based on a doctor's diagnosis of the patient's condition. Currently, diagnosis of Alzheimer often uses MRI machine, to know the condition of part of the brain called Hippocampus and Ventricle. MRI image itself consists of 3 slices, namely Coronal, Sagittal and Axial. In this paper, we discussed the measurement of the area of the ventricle especially in the Coronal slice based on the severity level referring to the CDR value. We use Active Contour method to segment the ventricle’s region, therefore that ventricle’s area can be calculated automatically. The results show that this method can be used for further development in the automatic diagnosis of Alzheimer.

Keywords: Alzheimer, CDR, coronal, ventricle, active contour

Procedia PDF Downloads 266

Authors: Madhu Jain, Rakesh Kumar Meena

Abstract:

This paper investigates a finite M/G/1 fault tolerant multi-component machining system. The system incorporates the features such as standby support, threshold recovery and imperfect coverage make the study closer to real time systems. The performance prediction of M/G/1 fault tolerant system is carried out using recursive approach by treating remaining service time as a supplementary variable. The numerical results are presented to illustrate the computational tractability of analytical results by taking three different service time distributions viz. exponential, 3-stage Erlang and deterministic. Moreover, the cost function is constructed to determine the optimal choice of system descriptors to upgrading the system.

Keywords: fault tolerant, machine repair, threshold recovery policy, imperfect coverage, supplementary variable technique

Procedia PDF Downloads 292

Authors: Prajwal Prakash Vasisht, Sharath Rajamurthy, Nishanth Dara

Abstract:

In a supervised machine learning approach, metrics such as precision, accuracy, and coverage can be calculated using ground truth labels to help in model tuning, evaluation, and selection. In an unsupervised setting, however, where the data has no ground truth, there are few interpretable metrics that can guide us to do the same. Our approach creates a framework to adapt the Intersection over Union metric, referred to as Adapted IoU, usually used to evaluate supervised learning models, into the unsupervised domain, which solves the problem by factoring in subject matter expertise and intuition about the ideal output from the model. This metric essentially provides a scale that allows us to compare the performance across numerous unsupervised models or tune hyper-parameters and compare different versions of the same model.

Keywords: general metric, unsupervised learning, classification, intersection over union

Procedia PDF Downloads 49

Authors: Mogale Sabone, Thabiso Ntlole

Abstract:

The pressure from globalization has made manufacturing organizations to move towards three major competitive arenas: quality, cost, and responsiveness. Quality is a universal value and has become a global issue. In order to survive and be able to provide customers with good products, manufacturing organizations’ supporting systems, tools, and structures it uses must grow or evolve. The majority of quality management concepts and strategies that are practiced recently are aimed at detecting and correcting problems which already exist and serve to limit losses. In agile manufacturing environment there is no room for defect and error so it needs a quality management which is proactively directed at problem prevention. This proactive quality management avoids losses by focusing on failure prevention, virtual elimination of the possibility of premature failure, mistake-proofing, and assuring consistently high quality in the definition and design of creation processes. To achieve this, a decision support tool for quality control and management is suggested. Current decision support tools/methods used by most manufacturing companies in Botswana for quality management and control are not integrated, for example they are not consistent since some tests results data is recorded manually only whilst others are recorded electronically. It is only a set of procedures not a tool. These procedures cannot offer interactive decision support. This point brings to light the aim of this research which is to develop a framework which will help manufacturing companies in Botswana build a decision support tool for quality control and management.

Keywords: decision support tool, manufacturing, quality control, quality management

Procedia PDF Downloads 566

Authors: Caihong Sun, Xizi Zhang

Abstract:

Rating prediction is an important problem for recommender systems. The task is to predict the rating for an item that a user would give. Most of the existing algorithms for the task ignore the effect of negative ratings rated by users on items, but the negative ratings have a significant impact on users’ purchasing decisions in practice. In this paper, we present a rating prediction algorithm based on factorization machines that consider the effect of negative ratings inspired by Loss Aversion theory. The aim of this paper is to develop a concave and a convex negative disgust function to evaluate the negative ratings respectively. Experiments are conducted on MovieLens dataset. The experimental results demonstrate the effectiveness of the proposed methods by comparing with other four the state-of-the-art approaches. The negative ratings showed much importance in the accuracy of ratings predictions.

Keywords: factorization machines, feature engineering, negative ratings, recommendation systems

Procedia PDF Downloads 242

Authors: Precious Ehiomogue, Ifechukwude Israel Ahuchaogu, Isiguzo Edwin Ahaneku

Abstract:

Vermicompost is the product of the decomposition process using various species of worms, to create a mixture of decomposing vegetable or food waste, bedding materials, and vemicast. This process is called vermicomposting, while the rearing of worms for this purpose is called vermiculture. Several works have verified the adsorption of toxic metals using vermicompost but the application is still scarce for the retention of organic compounds. This research brings to knowledge the effectiveness of earthworm waste (vermicompost) for the remediation of crude oil contaminated soils. The remediation methods adopted in this study were two soil washing methods namely, batch and column process which represent laboratory and in-situ remediation. Characterization of the vermicompost and crude oil contaminated soil were performed before and after the soil washing using Fourier transform infrared (FTIR), scanning electron microscopy (SEM), X-ray fluorescence (XRF), X-ray diffraction (XRD) and Atomic adsorption spectrometry (AAS). The optimization of washing parameters, using response surface methodology (RSM) based on Box-Behnken Design was performed on the response from the laboratory experimental results. This study also investigated the application of machine learning models [Artificial neural network (ANN), Adaptive neuro fuzzy inference system (ANFIS). ANN and ANFIS were evaluated using the coefficient of determination (R²) and mean square error (MSE)]. Removal efficiency obtained from the Box-Behnken design experiment ranged from 29% to 98.9% for batch process remediation. Optimization of the experimental factors carried out using numerical optimization techniques by applying desirability function method of the response surface methodology (RSM) produce the highest removal efficiency of 98.9% at absorbent dosage of 34.53 grams, adsorbate concentration of 69.11 (g/ml), contact time of 25.96 (min), and pH value of 7.71, respectively. Removal efficiency obtained from the multilevel general factorial design experiment ranged from 56% to 92% for column process remediation. The coefficient of determination (R²) for ANN was (0.9974) and (0.9852) for batch and column process, respectively, showing the agreement between experimental and predicted results. For batch and column precess, respectively, the coefficient of determination (R²) for RSM was (0.9712) and (0.9614), which also demonstrates agreement between experimental and projected findings. For the batch and column processes, the ANFIS coefficient of determination was (0.7115) and (0.9978), respectively. It can be concluded that machine learning models can predict the removal of crude oil from polluted soil using vermicompost. Therefore, it is recommended to use machines learning models to predict the removal of crude oil from contaminated soil using vermicompost.

Keywords: ANFIS, ANN, crude-oil, contaminated soil, remediation and vermicompost

Procedia PDF Downloads 111

Authors: Ranjith Maniyeri, Ahamed C. Saleel

Abstract:

Two-dimensional fluid flow over a stationary circular cylinder is one of the bench mark problem in the field of fluid-structure interaction in computational fluid dynamics (CFD). Motivated by this, in the present work, a two-dimensional computational model is developed using an improved version of immersed boundary method which combines the feedback forcing scheme of the virtual boundary method with Peskin’s regularized delta function approach. Lagrangian coordinates are used to represent the cylinder and Eulerian coordinates are used to describe the fluid flow. A two-dimensional Dirac delta function is used to transfer the quantities between the sold to fluid domain. Further, continuity and momentum equations governing the fluid flow are solved using fractional step based finite volume method on a staggered Cartesian grid system. The developed code is validated by comparing the values of drag coefficient obtained for different Reynolds numbers with that of other researcher’s results. Also, through numerical simulations for different Reynolds numbers flow behavior is well captured. The stability analysis of the improved version of immersed boundary method is tested for different values of feedback forcing coefficients.

Keywords: Feedback Forcing Scheme, Finite Volume Method, Immersed Boundary Method, Navier-Stokes Equations

Procedia PDF Downloads 305

Authors: Sofia Stoica

Abstract:

In this paper, we use historical data to predict the stock price of a tech company. To this end, we use a dataset consisting of the stock prices in the past five years of ten major tech companies – Adobe, Amazon, Apple, Facebook, Google, Microsoft, Netflix, Oracle, Salesforce, and Tesla. We experimented with a variety of models– a linear regressor model, K nearest Neighbors (KNN), a sequential neural network – and algorithms - Multiplicative Weight Update, and AdaBoost. We found that the sequential neural network performed the best, with a testing error of 0.18%. Interestingly, the linear model performed the second best with a testing error of 0.73%. These results show that using historical data is enough to obtain high accuracies, and a simple algorithm like linear regression has a performance similar to more sophisticated models while taking less time and resources to implement.

Keywords: finance, machine learning, opening price, stock market

Procedia PDF Downloads 190

Authors: Youssef Chahir, Messaoud Mostefai, Salah Khodja

Abstract:

This paper presents the development of a 2D visual marker, derived from a recent patented work in the field of numbering systems. The proposed fiducial uses a group of projective invariant straight-line patterns, easily detectable and remotely recognizable. Based on an efficient data coding scheme, the developed marker enables producing a large panel of unique real time identifiers with highly distinguishable patterns. The proposed marker Incorporates simultaneously decimal and binary information, making it readable by both humans and machines. This important feature opens up new opportunities for the development of efficient visual human-machine communication and monitoring protocols. Extensive experiment tests validate the robustness of the marker against acquisition and geometric distortions.

Keywords: visual markers, projective invariants, distance map, level sets

Procedia PDF Downloads 163

Authors: Robert Poenaru, Mihail Pleşa

Abstract:

In this work, a solution that addresses the protection of pre-trained neural networks is developed: Singularization. This method involves applying permutations to the weight matrices of a pre-trained model, introducing a form of structured noise that obscures the original model’s architecture. These permutations make it difficult for an attacker to reconstruct the original model, even if the permuted weights are obtained. Experimental benchmarks indicate that the application of singularization has a profound impact on model performance, often degrading it to the point where retraining from scratch becomes necessary to recover functionality, which is particularly effective for securing intellectual property in neural networks. Moreover, unlike other approaches, singularization is lightweight and computationally efficient, which makes it well suited for resource-constrained environments. Our experiments also demonstrate that this technique performs efficiently in various image classification tasks, highlighting its broad applicability and practicality in real-world scenarios.

Keywords: machine learning, ANE, CNN, security

Procedia PDF Downloads 14