Search results for: cable splicing machine

Authors: Amarnath M., Ramachandra C. G., H. Chelladurai, P..Sateesh Kumar, K. Santhosh Kumar

Abstract:

Journal bearings are extensively used sliding contact machine elements to support radial/axial loaded rotors used in various applications viz. automobile crankshaft, turbine propeller shaft, rope conveyer, heavy duty electric motors. The primary reasons for the failures of these bearings include unstable lubricant film, oil degradation, misalignment, etc. This paper describes the results of experimental investigations carried out to detect surface fatigue wear developed on load bearing the contact surfaces of journal bearing. The test bearing was subjected to fatigue load cycles over a period of 600 hours. The vibration signals were acquired from the journal bearing at regular intervals of 100 hrs. These signals were post-processed by using the vibration analysis technique to obtain diagnostic information of wear propagated in the journal-bearing system.

Keywords: fatigue, journal bearing, sound signals, vibration signals, wear

Procedia PDF Downloads 81

Authors: Glaoui Hachemi

Abstract:

In this paper, a scheme based on multi-input multi output Fuzzy Sliding Mode control (MIMO-FSMC) for linear speed regulation of winding system is proposed. Once the uncoupled model of the winding system was obtained, a smooth control function with a threshold was selected to indicate how far away the case was from the sliding surface. nevertheless, this control function depends closely on the higher bound of the uncertainties, which generates overlap. So, this size has to be chosen with broad care to obtain high performances. Usually, the upper bound of uncertainties is difficult to know before motor operation, so, a Fuzzy Sliding Mode controller is investigated to resolve this problem, a simple Fuzzy inference mechanism is used to decrease the chattering phenomenon by simple adjustments. A simulation study is achieved and that the indicate fuzzy sliding mode controllers have great potential for use as an alternative to the conventional sliding mode control.

Keywords: Winding system, induction machine, Mechanical tension, Proportional-integral (PI), sliding mode control, Fuzzy logic

Procedia PDF Downloads 96

Authors: B. G. Patel, A. K. Desai, S. G. Shah

Abstract:

The main of objective of this paper is to develop new measurement technique without touching the object. DIC is advance measurement technique use to measure displacement of particle with very high accuracy. This powerful innovative technique which is used to correlate two image segments to determine the similarity between them. For this study, nine geometrically similar beam specimens of different sizes with (steel fibers and glass fibers) and without fibers were tested under three-point bending in a closed loop servo-controlled machine with crack mouth opening displacement control with a rate of opening of 0.0005 mm/sec. Digital images were captured before loading (unreformed state) and at different instances of loading and were analyzed using correlation techniques to compute the surface displacements, crack opening and sliding displacements, load-point displacement, crack length and crack tip location. It was seen that the CMOD and vertical load-point displacement computed using DIC analysis matches well with those measured experimentally.

Keywords: Digital Image Correlation, fibres, self compacting concrete, size effect

Procedia PDF Downloads 389

Authors: Boughedaoui Rachid, Belaidi Idir, Ouali Mohamed

Abstract:

The formatting by removal of material remains an indispensable means for obtaining different forms of pieces. The objective of this work is to study the influence of parameters of the vibratory regime of the system PTM 'Piece-Tool-Machine, in the case of the machining of the thin pieces on the surface finish. As a first step, an analytical study of essential dynamic models 2D slice will be presented. The stability lobes will be thus obtained. In a second step, a characterization of PTM system will be realized. This system will be instrumented with accelerometric sensors but also a laser vibrometer so as to have the information closer to the cutting area. Dynamometers three components will be used for the analysis of cutting forces. Surface states will be measured and the condition of the cutting edge will be visualized thanks to a binocular microscope coupled to a data acquisition system. This information will allow quantifying the influence of chatter on the dimensional quality of the parts. From lobes stabilities previously determined experimental validation allow for the development a method for detecting of the phenomenon of chatter and so an approach will be proposed.

Keywords: chatter, dynamic, milling, lobe stability

Procedia PDF Downloads 357

Authors: Daniel Benson, Yundan Gong, Hannah Kirk

Abstract:

Advancement in national language processing techniques has led to increased data processing speeds, and reduced the need for cumbersome, manual data processing that is often required when processing data from multilateral organizations for specific purposes. As such, using named entity recognition (NER) modeling and the Organisation of Economically Developed Countries (OECD) Creditor Reporting System database, we present the first geotagged dataset of OECD donor Official Development Assistance (ODA) projects on a global, subnational basis. Our resulting data contains 52,086 ODA projects geocoded to subnational locations across 115 countries, worth a combined $87.9bn. This represents the first global, OECD donor ODA project database with geocoded projects. We use this new data to revisit old questions of how ‘well’ donors allocate ODA to the developing world. This understanding is imperative for policymakers seeking to improve ODA effectiveness.

Keywords: international aid, geocoding, subnational data, natural language processing, machine learning

Procedia PDF Downloads 78

Authors: Kevin Smith

Abstract:

Publicly traded company in the US are legally obligated to host earnings calls that discuss their most recent financial reports. During these calls, investors are able to ask these companies questions about these financial reports and on the future direction of the company. This paper examines whether common institutional owners use these calls as a way to indirectly signal to companies in their portfolio to not take actions that could hurt the common owner's interests. This paper uses transcripts taken from the earnings calls of the six largest health insurance companies in the US from 2014 to 2019. This data is analyzed using text analysis and sentiment analysis to look for patterns in the statements made by common owners. The analysis found that common owners where more likely to recommend against direct price competition and instead redirect the insurance companies towards more passive actions, like investing in new technologies. This result indicates a mechanism that common owners use to reduce competition in the health insurance market.

Keywords: common ownership, text analysis, sentiment analysis, machine learning

Procedia PDF Downloads 74

Authors: Sungyani Tripathy, Atul Desai

Abstract:

Turbo machinery Frame Foundation is very important for power generation, gas, steam, hydro, geothermal and nuclear power plants. The Turbo machinery Foundation system was simulated in SAP: 2000 software and dynamic response of foundation was analysed. In this paper, the detailed study of turbo machinery foundation with different running speed has considered. The different revolution per minute considered in this study is 4000 rpm, 6000 rpm, 8000 rpm, 1000 rpm and 12000 rpm. The above analysis has been carried out considering Winkler spring soil model, solid finite element modelling and dynamic analysis of Turbo machinery foundations. The comparison of frequency and time periods at various mode shapes are addressed in this study. Current work investigates the effect of damping on the response spectra curve at the foundation top deck, considering the dynamic machine load. It has been found that turbo generator foundation with haunches remains more elastic during seismic action for different running speeds.

Keywords: turbo machinery, SAP: 2000, response spectra, running speeds

Procedia PDF Downloads 255

Authors: Aliasghar Baziar, Hassan Masoumi, Alireza Ale Saadi

Abstract:

The control of the response of electric motors plays a significant role in the damping of transient responses. In this regard, this paper presents a static VAR compensator (SVC) based on a fuzzy logic which is applied to an industrial power network consisting of three phase synchronous, asynchronous and DC motor loads. The speed and acceleration variations of a specific machine are the inputs of the proposed fuzzy logic controller (FLC). In order to verify the effectiveness and proficiency of the proposed Fuzzy Logic based SVC (FLSVC), several non-linear time-domain digital simulation tests are performed. The proposed fuzzy model can properly control the response of electric motors. The results show that the FLSVC is successful to improve the voltage profile significantly over a wide range of operating conditions and disturbances thus improving the overall dynamic performance of the network.

Keywords: fuzzy logic controller, VAR compensator, single cage asynchronous motor, DC motor

Procedia PDF Downloads 628

Authors: Donghee Noh, Seonhyeong Kim, Junhwan Choi, Heegon Kim, Sooho Jung, Keunho Park

Abstract:

In this paper, we describe an image post-processing method to enhance the performance of the parcel segmentation method using deep learning-based aerial images conducted in previous studies. The study results were evaluated using a confusion matrix, IoU, Precision, Recall, and F1-Score. In the case of the confusion matrix, it was observed that the false positive value, which is the result of misclassification, was greatly reduced as a result of image post-processing. The average IoU was 0.9688 in the image post-processing, which is higher than the deep learning result of 0.8362, and the F1-Score was also 0.9822 in the image post-processing, which was higher than the deep learning result of 0.8850. As a result of the experiment, it was found that the proposed technique positively complements the deep learning results in segmenting the parcel of interest.

Keywords: aerial image, image process, machine vision, open field smart farm, segmentation

Procedia PDF Downloads 81

Authors: Arezou Javadi

Abstract:

The use of algorithmic models dependent on software calculations and simulation of new government management assays with the help of specialized software had increased the productivity and efficiency of the government management system recently. This has caused the management approach to change from the old bitch & fix model, which has low efficiency and less usefulness, to the capable management model with higher efficiency called the partnership with resident model. By using Govpilot TM software, the relationship between people in a system and the government was examined. The method of two tailed interaction was the outsourcing of a goal in a system, which is formed in the order of goals, qualified executive people, optimal executive model, and finally, summarizing additional activities at the different statistical levels. The results showed that the participation of people in a financial implementation system with a statistical potential of P≥5% caused a significant increase in investment and initial capital in the government system with maximum implement project in a smart government.

Keywords: machine learning, financial income, statistical potential, govpilot

Procedia PDF Downloads 88

Authors: Arezou Javadi

Abstract:

The use of algorithmic models dependent on software calculations and simulation of new government management assays with the help of specialized software had increased the productivity and efficiency of the government management system recently. This has caused the management approach to change from the old bitch & fix model, which has low efficiency and less usefulness, to the capable management model with higher efficiency called the partnership with resident model. By using Govpilot TM software, the relationship between people in a system and the government was examined. The method of two tailed interaction was the outsourcing of a goal in a system, which is formed in the order of goals, qualified executive people, optimal executive model, and finally, summarizing additional activities at the different statistical levels. The results showed that the participation of people in a financial implementation system with a statistical potential of P≥5% caused a significant increase in investment and initial capital in the government system with maximum implement project in a smart government.

Keywords: machine learning, financial income, statistical potential, govpilot

Procedia PDF Downloads 70

Authors: Changqiao Wu, Guoqing Ding, Xin Chen

Abstract:

In this paper, ways of modeling dynamic measurement systems are discussed. Specially, for linear system with single-input single-output, it could be modeled with shallow neural network. Then, gradient based optimization algorithms are used for searching the proper coefficients. Besides, method with normal equation and second order gradient descent are proposed to accelerate the modeling process, and ways of better gradient estimation are discussed. It shows that the mathematical essence of the learning objective is maximum likelihood with noises under Gaussian distribution. For conventional gradient descent, the mini-batch learning and gradient with momentum contribute to faster convergence and enhance model ability. Lastly, experimental results proved the effectiveness of second order gradient descent algorithm, and indicated that optimization with normal equation was the most suitable for linear dynamic models.

Keywords: dynamic system modeling, neural network, normal equation, second order gradient descent

Procedia PDF Downloads 127

Authors: Ali Ulvi Uzer

Abstract:

Data reduction is an important topic in the field of pattern recognition applications. The basic concept is the reduction of multitudinous amounts of data down to the meaningful parts. The Principal Component Analysis (PCA) method is frequently used for data reduction. The Support Vector Machine (SVM) method is a discriminative classifier formally defined by a separating hyperplane. In other words, given labeled training data, the algorithm outputs an optimal hyperplane which categorizes new examples. This study offers a hybrid approach that uses the PCA for data reduction and Support Vector Machines (SVM) for classification. In order to detect the accuracy of the suggested system, two boreholes taken from the soil sample was used. The classification accuracies for this dataset were obtained through using ten-fold cross-validation method. As the results suggest, this system, which is performed through size reduction, is a feasible system for faster recognition of dataset so our study result appears to be very promising.

Keywords: feature selection, sequential forward selection, support vector machines, soil sample

Procedia PDF Downloads 455

Authors: Qi Xiu, Hiromu Hota, Yohsuke Ishii, Takuya Oda

Abstract:

Recently, collectable manufacturing data are rapidly increasing. On the other hand, mega recall is getting serious as a social problem. Under such circumstances, there are increasing needs for preventing mega recalls by defect analysis such as root cause analysis and abnormal detection utilizing manufacturing data. However, the time to classify strings in manufacturing data by traditional method is too long to meet requirement of quick defect analysis. Therefore, we present String Length Distribution Classification method (SLDC) to correctly classify strings in a short time. This method learns character features, especially string length distribution from Product ID, Machine ID in BOM and asset list. By applying the proposal to strings in actual manufacturing data, we verified that the classification time of strings can be reduced by 80%. As a result, it can be estimated that the requirement of quick defect analysis can be fulfilled.

Keywords: string classification, data quality, feature selection, probability distribution, string length

Procedia PDF Downloads 318

Authors: Ashish Kumar, Kaptan Singh, Amit Saxena

Abstract:

Today, it is believed that crimes have the greatest impact on a person's ability to progress financially and personally. Identifying places where individuals shouldn't go is crucial for preventing crimes and is one of the key considerations. As society and technologies have advanced significantly, so have crimes and the harm they wreak. When there is a concentration of people in one place and changes happen quickly, it is even harder to prevent. Because of this, many crime prevention strategies have been embraced as a component of the development of smart cities in numerous cities. However, crimes can occur anywhere; all that is required is to identify the pattern of their occurrences, which will help to lower the crime rate. In this paper, an analysis related to crime has been done; information related to crimes is collected from all over India that can be accessed from anywhere. The purpose of this paper is to investigate the relationship between several factors and India's crime rate. The review has covered information related to every state of India and their associated regions of the period going in between 2001- 2014. However various classes of violations have a marginally unique scope over the years.

Keywords: K-nearest neighbor, random forest, decision tree, pre-processing

Procedia PDF Downloads 93

Authors: Fuad M. Alkoot

Abstract:

We aim at highlighting the benefits of classifier systems especially in supporting educational management decisions. The paper aims at using classifiers in an educational application where an outcome is predicted based on given input parameters that represent various conditions at the institute. We present a classifier system that is designed using a limited training set with data for only one semester. The achieved system is able to reach at previously known outcomes accurately. It is also tested on new input parameters representing variations of input conditions to see its prediction on the possible outcome value. Given the supervised expectation of the outcome for the new input we find the system is able to predict the correct outcome. Experiments were conducted on one semester data from two departments only, Switching and Mathematics. Future work on other departments with larger training sets and wider input variations will show additional benefits of classifier systems in supporting the management decisions at an educational institute.

Keywords: machine learning, pattern recognition, classifier design, educational management, outcome estimation

Procedia PDF Downloads 278

Authors: Akella Amarendra Babu, Rama Devi Yellasiri, Natukula Sainath

Abstract:

Where human beings can easily learn and adopt pronunciation variations, machines need training before put into use. Also humans keep minimum vocabulary and their pronunciation variations are stored in front-end of their memory for ready reference, while machines keep the entire pronunciation dictionary for ready reference. Supervised methods are used for preparation of pronunciation dictionaries which take large amounts of manual effort, cost, time and are not suitable for real time use. This paper presents an unsupervised adaptation model for building agile and dynamic pronunciation dictionaries online. These methods mimic human approach in learning the new pronunciations in real time. A new algorithm for measuring sound distances called Dynamic Phone Warping is presented and tested. Performance of the system is measured using an adaptation model and the precision metrics is found to be better than 86 percent.

Keywords: pronunciation variations, dynamic programming, machine learning, natural language processing

Procedia PDF Downloads 175

Authors: Serge Provost, Yishan Zang

Abstract:

Copulas are currently utilized in finance, reliability theory, machine learning, signal processing, geodesy, hydrology and biostatistics, among several other fields of scientific investigation. It follows from Sklar's theorem that the joint distribution function of a multidimensional random vector can be expressed in terms of its associated copula and marginals. Since marginal distributions can easily be determined by making use of a variety of techniques, we address the problem of securing the distribution of the copula. This will be done by using several approaches. For example, we will obtain bivariate least-squares approximations of the empirical copulas, modify the kernel density estimation technique and propose a criterion for selecting appropriate bandwidths, differentiate linearized empirical copulas, secure Bernstein polynomial approximations of suitable degrees, and apply a corollary to Sklar's result. Illustrative examples involving actual observations will be presented. The proposed methodologies will as well be applied to a sample generated from a known copula distribution in order to validate their effectiveness.

Keywords: copulas, Bernstein polynomial approximation, least-squares polynomial approximation, kernel density estimation, density approximation

Procedia PDF Downloads 74

Authors: Khaled Ibrahim, Mathew Parackal, Damien Mather, Paul Hansen

Abstract:

The integration between marketing and technology has given brands unprecedented opportunities to reach accurate customer data and competence to change customers' behaviour. Technology has generated a transformation within brands from traditional branding to algorithmic branding. However, brands have utilised customer data in non-cognitive programmatic targeting. This algorithmic persuasion may be effective in reaching the targeted audience. But it may encounter a moral conflict simultaneously, as it might not consider our social principles. Moral branding is a critical topic; particularly, with the increasing interest in commercial settings to teaching machines human morals, e.g., autonomous vehicles and chatbots; however, it is understudied in the marketing literature. Therefore, this paper aims to investigate the recent moral branding literature. Furthermore, applying human-like mind theory as initial framing to this paper explores a more comprehensive concept involving human morals, machine behaviour, and branding.

Keywords: brand machines, conceptual framework, moral branding, moral machines

Procedia PDF Downloads 163

Authors: Usama Mohamed Ahamed

Abstract:

This research deals with the bond behavior of carbon FRP composite wraps adhered/bonded to the surface of the concrete. Four concrete mixes were designed to achieve a concrete compressive strength of 18, 22.5,25 and 30 MP after 28 days of curing. The focus of the study is on bond degradation when the hybrid structure is exposed to hot weather conditions. Specimens were exposed to 50 0C temperature duration 6 months and other specimens were sustained in laboratory temperature ( 20-24) 0C. Upon removing the specimens from their conditioning environment, tension tests were performed in the machine using a specially manufactured concrete cube holder. A lightweight mortar layer is used to protect the bonded carbon FRP layer on the concrete surface. The results show that the higher the concrete's compressive, the higher the bond strength. The high temperature decreases the bond strength between concrete and carbon fiber-reinforced polymer. The use of a protection layer is essential for concrete exposed to hot weather.

Keywords: concrete, bond, hot weather and carbon fiber, carbon fiber reinforced polymers

Procedia PDF Downloads 107

Authors: An-Jui. Li, Kelvin Lim, Chien-Kuo Chiu, Benson Hsiung

Abstract:

This paper details the utilization of artificial intelligence (AI) in the field of slope stability whereby quick and convenient solutions can be obtained using the developed tool. The AI tool used in this study is the artificial neural network (ANN), while the slope stability analysis methods are the finite element limit analysis methods. The developed tool allows for the prompt prediction of the safety factors of fill slopes and their corresponding probability of failure (depending on the degree of variation of the soil parameters), which can give the practicing engineer a reasonable basis in their decision making. In fact, the successful use of the Extreme Learning Machine (ELM) algorithm shows that slope stability analysis is no longer confined to the conventional methods of modeling, which at times may be tedious and repetitive during the preliminary design stage where the focus is more on cost saving options rather than detailed design. Therefore, similar ANN-based tools can be further developed to assist engineers in this aspect.

Keywords: landslide, limit analysis, artificial neural network, soil properties

Procedia PDF Downloads 207

Authors: Ritesh Joshi, Kishan Fuse, Gopal Zinzala, Nishit Nirmal

Abstract:

Ti-6Al-4V is a titanium alloy having high strength, low weight and corrosion resistant which is a required characteristic for a material to be used in aerospace industry. Titanium, being a hard alloy is difficult to the machine via conventional methods, so it is a call to use non-conventional processes. In present work, the effects on Ti-6Al-4V by drilling a hole of Ø 6 mm using copper (99%) electrode in Electric Discharge Machining (EDM) process is analyzed. Effect of various input parameters like peak current, pulse-on time and pulse-off time on output parameters viz material removal rate (MRR) and electrode wear rate (EWR) is studied. Multi-objective optimization technique Grey relational analysis is used for process optimization. Experiments are designed using an L9 orthogonal array. ANOVA is used for finding most contributing parameter followed by confirmation tests for validating the results. Improvement of 7.45% in gray relational grade is observed.

Keywords: ANOVA, electric discharge machining, grey relational analysis, Ti-6Al-4V

Procedia PDF Downloads 363

Authors: Usha Lokala, Srinivas Nowduri, Prabhakar K. Sharma

Abstract:

Useful information has been extracted from the road accident data in United Kingdom (UK), using data analytics method, for avoiding possible accidents in rural and urban areas. This analysis make use of several methodologies such as data integration, support vector machines (SVM), correlation machines and multinomial goodness. The entire datasets have been imported from the traffic department of UK with due permission. The information extracted from these huge datasets forms a basis for several predictions, which in turn avoid unnecessary memory lapses. Since data is expected to grow continuously over a period of time, this work primarily proposes a new framework model which can be trained and adapt itself to new data and make accurate predictions. This work also throws some light on use of SVM’s methodology for text classifiers from the obtained traffic data. Finally, it emphasizes the uniqueness and adaptability of SVMs methodology appropriate for this kind of research work.

Keywords: support vector mechanism (SVM), machine learning (ML), support vector machines (SVM), department of transportation (DFT)

Procedia PDF Downloads 274

Authors: Piotr A. Kowalski, Kasper Sapala, Wiktor Warchalowski

Abstract:

PM10 is a suspended dust that primarily has a negative effect on the respiratory system. PM10 is responsible for attacks of coughing and wheezing, asthma or acute, violent bronchitis. Indirectly, PM10 also negatively affects the rest of the body, including increasing the risk of heart attack and stroke. Unfortunately, Poland is a country that cannot boast of good air quality, in particular, due to large PM concentration levels. Therefore, based on the dense network of Airly sensors, it was decided to deal with the problem of prediction of suspended particulate matter concentration. Due to the very complicated nature of this issue, the Machine Learning approach was used. For this purpose, Convolution Neural Network (CNN) neural networks have been adopted, these currently being the leading information processing methods in the field of computational intelligence. The aim of this research is to show the influence of particular CNN network parameters on the quality of the obtained forecast. The forecast itself is made on the basis of parameters measured by Airly sensors and is carried out for the subsequent day, hour after hour. The evaluation of learning process for the investigated models was mostly based upon the mean square error criterion; however, during the model validation, a number of other methods of quantitative evaluation were taken into account. The presented model of pollution prediction has been verified by way of real weather and air pollution data taken from the Airly sensor network. The dense and distributed network of Airly measurement devices enables access to current and archival data on air pollution, temperature, suspended particulate matter PM1.0, PM2.5, and PM10, CAQI levels, as well as atmospheric pressure and air humidity. In this investigation, PM2.5, and PM10, temperature and wind information, as well as external forecasts of temperature and wind for next 24h served as inputted data. Due to the specificity of the CNN type network, this data is transformed into tensors and then processed. This network consists of an input layer, an output layer, and many hidden layers. In the hidden layers, convolutional and pooling operations are performed. The output of this system is a vector containing 24 elements that contain prediction of PM10 concentration for the upcoming 24 hour period. Over 1000 models based on CNN methodology were tested during the study. During the research, several were selected out that give the best results, and then a comparison was made with the other models based on linear regression. The numerical tests carried out fully confirmed the positive properties of the presented method. These were carried out using real ‘big’ data. Models based on the CNN technique allow prediction of PM10 dust concentration with a much smaller mean square error than currently used methods based on linear regression. What's more, the use of neural networks increased Pearson's correlation coefficient (R²) by about 5 percent compared to the linear model. During the simulation, the R² coefficient was 0.92, 0.76, 0.75, 0.73, and 0.73 for 1st, 6th, 12th, 18th, and 24th hour of prediction respectively.

Keywords: air pollution prediction (forecasting), machine learning, regression task, convolution neural networks

Procedia PDF Downloads 149

Authors: Onan Kilicaslan, Anil Kabaklarli, Levent Subasi, Erdem Bektas, Rifat Yilmaz

Abstract:

The Aluminum-Copper casting alloys are well known for their high mechanical strength, especially when compared to more commonly used Aluminum-Silicon alloys. A201 is one of the best in terms of strength vs. weight ratio among other aluminum alloys, which makes it suitable for premium quality casting applications in aerospace and automotive industries. It is reported that A201 has low castability, but it is easy to machine. However, there is a need to specifically determine the process window for feasible machining. This research investigates the machinability of A201 alloy after T7 heat treatment in terms of chip/burr formation, surface roughness, hardness, and microstructure. The samples are cast with low-pressure sand casting method and milling experiments are performed with uncoated carbide tools using different cutting speeds and feeds. Statistical analysis is used to correlate the machining parameters to surface integrity. It is found that there is a strong dependence of the cutting conditions on machinability and a process window is determined.

Keywords: A201-T7, machinability, milling, surface integrity

Procedia PDF Downloads 196

Authors: Akeel A. Shah, Tong Zhang

Abstract:

Computational approaches to learning the properties of materials are commonplace, motivated by the need to screen or design materials for a given application, e.g., semiconductors and energy storage. Experimental approaches can be both time consuming and costly. Unfortunately, computational approaches such as ab-initio electronic structure calculations and classical or ab-initio molecular dynamics are themselves can be too slow for the rapid evaluation of materials, often involving thousands to hundreds of thousands of candidates. Machine learning assisted approaches have been developed to overcome the time limitations of purely physics-based approaches. These approaches, on the other hand, require large volumes of data for training (hundreds of thousands on many standard data sets such as QM7b). This means that they are limited by how quickly such a large data set of physics-based simulations can be established. At high fidelity, such as configuration interaction, composite methods such as G4, and coupled cluster theory, gathering such a large data set can become infeasible, which can compromise the accuracy of the predictions - many applications require high accuracy, for example band structures and energy levels in semiconductor materials and the energetics of charge transfer in energy storage materials. In order to circumvent this problem, multi-fidelity approaches can be adopted, for example the Δ-ML method, which learns a high-fidelity output from a low-fidelity result such as Hartree-Fock or density functional theory (DFT). The general strategy is to learn a map between the low and high fidelity outputs, so that the high-fidelity output is obtained a simple sum of the physics-based low-fidelity and correction, Although this requires a low-fidelity calculation, it typically requires far fewer high-fidelity results to learn the correction map, and furthermore, the low-fidelity result, such as Hartree-Fock or semi-empirical ZINDO, is typically quick to obtain, For high-fidelity outputs the result can be an order of magnitude or more in speed up. In this work, a new multi-fidelity approach is developed, based on a graph convolutional network (GCN) combined with semi-supervised learning. The GCN allows for the material or molecule to be represented as a graph, which is known to improve accuracy, for example SchNet and MEGNET. The graph incorporates information regarding the numbers of, types and properties of atoms; the types of bonds; and bond angles. They key to the accuracy in multi-fidelity methods, however, is the incorporation of low-fidelity output to learn the high-fidelity equivalent, in this case by learning their difference. Semi-supervised learning is employed to allow for different numbers of low and high-fidelity training points, by using an additional GCN-based low-fidelity map to predict high fidelity outputs. It is shown on 4 different data sets that a significant (at least one order of magnitude) increase in accuracy is obtained, using one to two orders of magnitude fewer low and high fidelity training points. One of the data sets is developed in this work, pertaining to 1000 simulations of quinone molecules (up to 24 atoms) at 5 different levels of fidelity, furnishing the energy, dipole moment and HOMO/LUMO.

Keywords: .materials screening, computational materials, machine learning, multi-fidelity, graph convolutional network, semi-supervised learning

Procedia PDF Downloads 41

Authors: Belkacem Laimouche

Abstract:

With the field of artificial intelligence (AI) experiencing exponential growth, fueled by technological advancements that pave the way for increasingly innovative and promising applications, there is an escalating need to develop rigorous methods for assessing their performance in pursuit of transparency and equity. This article proposes a metrology-inspired statistical framework for evaluating bias and explainability in AI systems. Drawing from the principles of metrology, we propose a pioneering approach, using a concrete example, to evaluate the accuracy and precision of AI models, as well as to quantify the sources of measurement uncertainty that can lead to bias in their predictions. Furthermore, we explore a statistical approach for evaluating the explainability of AI systems based on their ability to provide interpretable and transparent explanations of their predictions.

Keywords: artificial intelligence, metrology, measurement uncertainty, prediction error, bias, machine learning algorithms, probabilistic models, interlaboratory comparison, data analysis, data reliability, measurement of bias impact on predictions, improvement of model accuracy and reliability

Procedia PDF Downloads 105

Authors: Jinghui Peng, Shanyu Tang, Jia Li

Abstract:

Steganalysis seeks to detect the presence of secret data embedded in cover objects, and there is an imminent demand to detect hidden messages in streaming media. This paper shows how a steganalysis algorithm based on Fast Fourier Transform (FFT) can be used to detect the existence of secret data embedded in streaming media. The proposed algorithm uses machine parameter characteristics and a network sniffer to determine whether the Internet traffic contains streaming channels. The detected streaming data is then transferred from the time domain to the frequency domain through FFT. The distributions of power spectra in the frequency domain between original VoIP streams and stego VoIP streams are compared in turn using t-test, achieving the p-value of 7.5686E-176 which is below the threshold. The results indicate that the proposed FFT-based steganalysis algorithm is effective in detecting the secret data embedded in VoIP streaming media.

Keywords: steganalysis, security, Fast Fourier Transform, streaming media

Procedia PDF Downloads 147

Authors: Efrain Rodriguez, Sergio Pertuz, Cristhian Riano

Abstract:

Tool-path generation is an essential step in the FFF (Fused Filament Fabrication)-based Additive Manufacturing (AM) process planning. In the manufacture of a mechanical part by using additive processes, high resource consumption and prolonged production times are inherent drawbacks of these processes mainly due to non-optimized tool-path generation. In this work, we propose a heuristic-search intelligent algorithm-based approach for optimized tool-path generation for FFF-based AM. The main benefit of this approach is a significant reduction of travels without material deposition when the AM machine performs moves without any extrusion. The optimization method used reduces the number of travels without extrusion in comparison with commercial software as Slic3r or Cura Engine, which means a reduction of production time.

Keywords: additive manufacturing, tool-path optimization, fused filament fabrication, process planning

Procedia PDF Downloads 443

Authors: Tomasz Sapiński, Dorota Kamińska, Adam Pelikant, Egils Avots, Cagri Ozcinar, Gholamreza Anbarjafari

Abstract:

People express emotions through different modalities. Integration of verbal and non-verbal communication channels creates a system in which the message is easier to understand. Expanding the focus to several expression forms can facilitate research on emotion recognition as well as human-machine interaction. In this article, the authors present a Polish emotional database composed of three modalities: facial expressions, body movement and gestures, and speech. The corpora contains recordings registered in studio conditions, acted out by 16 professional actors (8 male and 8 female). The data is labeled with six basic emotions categories, according to Ekman’s emotion categories. To check the quality of performance, all recordings are evaluated by experts and volunteers. The database is available to academic community and might be useful in the study on audio-visual emotion recognition.

Keywords: body movement, emotion recognition, emotional corpus, facial expressions, gestures, multimodal database, speech

Procedia PDF Downloads 349