Search results for: perceptual linear prediction (PLP’s)

Authors: Babak Bahri, Fatemeh Yassaee Meybodi, Changiz Eslahchi

Abstract:

In the pursuit of effective cancer therapies, the exploration of combinatorial drug regimens is crucial to leverage synergistic interactions between drugs, thereby improving treatment efficacy and overcoming drug resistance. However, identifying synergistic drug pairs poses challenges due to the vast combinatorial space and limitations of experimental approaches. This study introduces ClusterSyn, a machine learning (ML)-powered framework for classifying anti-cancer drug synergy scores. ClusterSyn employs a two-step approach involving drug clustering and synergy score prediction using a fully connected deep neural network. For each cell line in the training dataset, a drug graph is constructed, with nodes representing drugs and edge weights denoting synergy scores between drug pairs. Drugs are clustered using the Markov clustering (MCL) algorithm, and vectors representing the similarity of drug pairs to each cluster are input into the deep neural network for synergy score prediction (synergy or antagonism). Clustering results demonstrate effective grouping of drugs based on synergy scores, aligning similar synergy profiles. Subsequently, neural network predictions and synergy scores of the two drugs on others within their clusters are used to predict the synergy score of the considered drug pair. This approach facilitates comparative analysis with clustering and regression-based methods, revealing the superior performance of ClusterSyn over state-of-the-art methods like DeepSynergy and DeepDDS on diverse datasets such as Oniel and Almanac. The results highlight the remarkable potential of ClusterSyn as a versatile tool for predicting anti-cancer drug synergy scores.

Keywords: drug synergy, clustering, prediction, machine learning., deep learning

Procedia PDF Downloads 79

Authors: Deepika Christopher, Garima Anand

Abstract:

To determine the best model for churn prediction in the telecom industry, this paper compares 11 machine learning algorithms, namely Logistic Regression, Support Vector Machine, Random Forest, Decision Tree, XGBoost, LightGBM, Cat Boost, AdaBoost, Extra Trees, Deep Neural Network, and Hybrid Model (MLPClassifier). It also aims to pinpoint the top three factors that lead to customer churn and conducts customer segmentation to identify vulnerable groups. According to the data, the Logistic Regression model performs the best, with an F1 score of 0.6215, 81.76% accuracy, 68.95% precision, and 56.57% recall. The top three attributes that cause churn are found to be tenure, Internet Service Fiber optic, and Internet Service DSL; conversely, the top three models in this article that perform the best are Logistic Regression, Deep Neural Network, and AdaBoost. The K means algorithm is applied to establish and analyze four different customer clusters. This study has effectively identified customers that are at risk of churn and may be utilized to develop and execute strategies that lower customer attrition.

Keywords: attrition, retention, predictive modeling, customer segmentation, telecommunications

Procedia PDF Downloads 57

Authors: Housein Deli, Loui Al-Shrouf, Hammoud Al Joumaa, Mohieddine Jelali

Abstract:

When forming metallic materials, fluctuations in material properties, process conditions, and wear lead to deviations in the component geometry. Several hundred features sometimes need to be measured, especially in the case of functional and safety-relevant components. These can only be measured offline due to the large number of features and the accuracy requirements. The risk of producing components outside the tolerances is minimized but not eliminated by the statistical evaluation of process capability and control measurements. The inspection intervals are based on the acceptable risk and are at the expense of productivity but remain reactive and, in some cases, considerably delayed. Due to the considerable progress made in the field of condition monitoring and measurement technology, permanently installed sensor systems in combination with machine learning and artificial intelligence, in particular, offer the potential to independently derive forecasts for component geometry and thus eliminate the risk of defective products - actively and preventively. The reliability of forecasts depends on the quality, completeness, and timeliness of the data. Measuring all geometric characteristics is neither sensible nor technically possible. This paper, therefore, uses the example of car seat rail production to discuss the necessary first step of feature selection and reduction by correlation analysis, as otherwise, it would not be possible to forecast components in real-time and inline. Four different car seat rails with an average of 130 features were selected and measured using a coordinate measuring machine (CMM). The run of such measuring programs alone takes up to 20 minutes. In practice, this results in the risk of faulty production of at least 2000 components that have to be sorted or scrapped if the measurement results are negative. Over a period of 2 months, all measurement data (> 200 measurements/ variant) was collected and evaluated using correlation analysis. As part of this study, the number of characteristics to be measured for all 6 car seat rail variants was reduced by over 80%. Specifically, direct correlations for almost 100 characteristics were proven for an average of 125 characteristics for 4 different products. A further 10 features correlate via indirect relationships so that the number of features required for a prediction could be reduced to less than 20. A correlation factor >0.8 was assumed for all correlations.

Keywords: long-term SHM, condition monitoring, machine learning, correlation analysis, component prediction, wear prediction, regressions analysis

Procedia PDF Downloads 48

Authors: Xia Chen, Yulan Wang

Abstract:

purpose: To compare the accuracy of Haigis, SRK/T, T2, Holladay 1, Hoffer Q, Barrett Universal II, Emmetropia Verifying Optical (EVO) and Kane for intraocular lens power calculation in patients with axial length (AL) ≥ 28 mm. Methods: In this retrospective single-center study, 50 eyes of 41 patients with AL ≥ 28 mm that underwent uneventful cataract surgery were enrolled. The actual postoperative refractive results were compared to the predicted refraction calculated with different formulas (Haigis, SRK/T, T2, Holladay 1, Hoffer Q, Barrett Universal II, EVO and Kane). The mean absolute prediction errors (MAE) 1 month postoperatively were compared. Results: The MAE of different formulas were as follows: Haigis (0.509), SRK/T (0.705), T2 (0.999), Holladay 1 (0.714), Hoffer Q (0.583), Barrett Universal II (0.552), EVO (0.463) and Kane (0.441). No significant difference was found among the different formulas (P = .122). The Kane and EVO formulas achieved the lowest level of mean prediction error (PE) and median absolute error (MedAE) (p < 0.05). Conclusion: The Kane and EVO formulas had a better success rate than others in predicting IOL power in high myopic eyes with AL longer than 28 mm in this study.

Keywords: cataract, power calculation formulas, intraocular lens, long axial length

Procedia PDF Downloads 83

Authors: P. Byrnes, F. A. DiazDelaO

Abstract:

The Support Vector Machine (SVM) has become widely recognised as one of the leading algorithms in machine learning for both regression and binary classification. It expresses predictions in terms of a linear combination of kernel functions, referred to as support vectors. Despite its popularity amongst practitioners, SVM has some limitations, with the most significant being the generation of point prediction as opposed to predictive distributions. Stemming from this issue, a probabilistic model namely, Probabilistic Classification Vector Machines (PCVM), has been proposed which respects the original functional form of SVM whilst also providing a predictive distribution. As physical system designs become more complex, an increasing number of classification tasks involving industrial applications consist of more than two classes. Consequently, this research proposes a framework which allows for the extension of PCVM to a multi class setting. Additionally, the original PCVM framework relies on the use of type II maximum likelihood to provide estimates for both the kernel hyperparameters and model evidence. In a high dimensional multi class setting, however, this approach has been shown to be ineffective due to bad scaling as the number of classes increases. Accordingly, we propose the application of Markov Chain Monte Carlo (MCMC) based methods to provide a posterior distribution over both parameters and hyperparameters. The proposed framework will be validated against current multi class classifiers through synthetic and real life implementations.

Keywords: probabilistic classification vector machines, multi class classification, MCMC, support vector machines

Procedia PDF Downloads 221

Authors: Adel Mohsen

Abstract:

A direct method based on the good Broyden's updates for evaluating the inverse of a nonsingular square matrix of full rank and solving related system of linear algebraic equations is studied. For a matrix A of order n whose LU-decomposition is A = LU, the multiplication count is O (n3). This includes the evaluation of the LU-decompositions of the inverse, the lower triangular decomposition of A as well as a “reduced matrix inverse”. If an explicit value of the inverse is not needed the order reduces to O (n3/2) to compute to compute inv(U) and the reduced inverse. For a symmetric matrix only O (n3/3) operations are required to compute inv(L) and the reduced inverse. An example is presented to demonstrate the capability of using the reduced matrix inverse in treating ill-conditioned systems. Besides the simplicity of Broyden's update, the method provides a mean to exploit the possible sparsity in the matrix and to derive a suitable preconditioner.

Keywords: Broyden's updates, matrix inverse, inverse factorization, solution of linear algebraic equations, ill-conditioned matrices, preconditioning

Procedia PDF Downloads 479

Authors: Grienggrai Rajchakit

Abstract:

This paper is concerned with the exponential stability of switched linear systems with interval time-varying delays. The time delay is any continuous function belonging to a given interval, in which the lower bound of delay is not restricted to zero. By constructing a suitable augmented Lyapunov-Krasovskii functional combined with Leibniz-Newton's formula, a switching rule for the exponential stability of switched linear systems with interval time-varying delays and new delay-dependent sufficient conditions for the exponential stability of the systems are first established in terms of LMIs. Finally, some examples are exploited to illustrate the effectiveness of the proposed schemes.

Keywords: exponential stability, hybrid systems, time-varying delays, lyapunov-krasovskii functional, leibniz-newton's formula

Procedia PDF Downloads 544

Authors: Raziyeh Shamsi

Abstract:

In this paper, we present a multiple objective linear programming (MOLP) problem in full fuzzy case and find Data Envelopment Analysis(DEA) targets. In the presented model, we are seeking the least inputs and the most outputs in the production possibility set (PPS) with the variable return to scale (VRS) assumption, so that the efficiency projection is obtained for all decision making units (DMUs). Then, we provide an algorithm for finding DEA targets interactively in the full fuzzy case, which solves the full fuzzy problem without defuzzification. Owing to the use of interactive methods, the targets obtained by our algorithm are more applicable, more realistic, and they are according to the wish of the decision maker. Finally, an application of the algorithm in 21 educational institutions is provided.

Keywords: DEA, MOLP, full fuzzy, target

Procedia PDF Downloads 302

Authors: Y. Khulief, S. Bashmal, S. Said, D. Al-Otaibi, K. Mansour

Abstract:

The prediction of flow rates at which the vibration-induced instability takes place in tubular heat exchangers due to cross-flow is of major importance to the performance and service life of such equipment. In this paper, the semi-analytical model for square tube arrays was extended and utilized to study the triangular tube patterns. A laboratory test rig with instrumented test section is used to measure the fluidelastic coefficients to be used for tuning the mathematical model. The test section can be made of any bundle pattern. In this study, two test sections were constructed for both the normal triangular and the rotated triangular tube arrays. The developed scheme is utilized in predicting the onset of flow-induced instability in the two triangular tube arrays. The results are compared to those obtained for two other bundle configurations. The results of the four different tube patterns are viewed in the light of TEMA predictions. The comparison demonstrated that TEMA guidelines are more conservative in all configurations considered

Keywords: fluid-structure interaction, cross-flow, heat exchangers,

Procedia PDF Downloads 277

Authors: Sabyasachi Biswas, Indrajit Ghosh

Abstract:

The random and heterogeneous behavior of vehicles in India puts up a greater challenge for researchers. Stop-and-go modeling at signalized intersections under heterogeneous traffic conditions has remained one of the most sought-after fields. Vehicles are often caught up in the dilemma zone and are unable to take quick decisions whether to stop or cross the intersection. This hampers the traffic movement and may lead to accidents. The purpose of this work is to develop a stop and go prediction model that depicts the drivers’ decision during the yellow time at signalised intersections. To accomplish this, certain traffic parameters were taken into account to develop surrogate model. This research investigated the Stop and Go behavior of the drivers by collecting data from 4-signalized intersections located in two major Indian cities. Model was developed to predict the drivers’ decision making during the yellow phase of the traffic signal. The parameters used for modeling included distance to stop line, time to stop line, speed, and length of the vehicle. A Kriging base surrogate model has been developed to investigate the drivers’ decision-making behavior in amber phase. It is observed that the proposed approach yields a highly accurate result (97.4 percent) by Gaussian function. It was observed that the accuracy for the crossing probability was 95.45, 90.9 and 86.36.11 percent respectively as predicted by the Kriging models with Gaussian, Exponential and Linear functions.

Keywords: decision-making decision, dilemma zone, surrogate model, Kriging

Procedia PDF Downloads 309

Authors: M. R. Dahmardeh Ghaleno, M. Nohtani, S. Khaledi

Abstract:

Flood simulation and prediction is one of the most active research areas in surface water management. WetSpa is a distributed, continuous, and physical model with daily or hourly time step that explains precipitation, runoff, and evapotranspiration processes for both simple and complex contexts. This model uses a modified rational method for runoff calculation. In this model, runoff is routed along the flow path using Diffusion-Wave equation which depends on the slope, velocity, and flow route characteristics. Golestan Dam Basin is located in Golestan province in Iran and it is passing over coordinates 55° 16´ 50" to 56° 4´ 25" E and 37° 19´ 39" to 37° 49´ 28"N. The area of the catchment is about 224 km², and elevations in the catchment range from 414 to 2856 m at the outlet, with average slope of 29.78%. Results of the simulations show a good agreement between calculated and measured hydrographs at the outlet of the basin. Drawing upon Nash-Sutcliffe model efficiency coefficient for calibration periodic model estimated daily hydrographs and maximum flow rate with an accuracy up to 59% and 80.18%, respectively.

Keywords: watershed simulation, WetSpa, stream flow, flood prediction

Procedia PDF Downloads 244

Authors: Firas M. Tuaimah, Huda M. Abdul Abbas

Abstract:

Accurate Short Term Load Forecasting (STLF) is essential for a variety of decision making processes. However, forecasting accuracy can drop due to the presence of uncertainty in the operation of energy systems or unexpected behavior of exogenous variables. Interval Type 2 Fuzzy Logic System (IT2 FLS), with additional degrees of freedom, gives an excellent tool for handling uncertainties and it improved the prediction accuracy. The training data used in this study covers the period from January 1, 2012 to February 1, 2012 for winter season and the period from July 1, 2012 to August 1, 2012 for summer season. The actual load forecasting period starts from January 22, till 28, 2012 for winter model and from July 22 till 28, 2012 for summer model. The real data for Iraqi power system which belongs to the Ministry of Electricity.

Keywords: short term load forecasting, prediction interval, type 2 fuzzy logic systems, electric, computer systems engineering

Procedia PDF Downloads 397

Authors: Reza Zakerinejad

Abstract:

Gully erosion is a serious problem that threading the sustainability of agricultural area and rangeland and water in a large part of Iran. This type of water erosion is the main source of sedimentation in many catchment areas in the north of Iran. Since in many national assessment approaches just qualitative models were applied the aim of this study is to predict the spatial distribution of gully erosion processes by means of detail terrain analysis and GIS -based logistic regression in the loess deposition in a case study in the Golestan Province. This study the DEM with 25 meter result ion from ASTER data has been used. The Landsat ETM data have been used to mapping of land use. The TreeNet model as a stochastic modeling was applied to prediction the susceptible area for gully erosion. In this model ROC we have set 20 % of data as learning and 20 % as learning data. Therefore, applying the GIS and satellite image analysis techniques has been used to derive the input information for these stochastic models. The result of this study showed a high accurate map of potential for gully erosion.

Keywords: TreeNet model, terrain analysis, Golestan Province, Iran

Procedia PDF Downloads 535

Authors: Bakhtiar Azadbakht, Karim Adinehvand, Amin Sahebnasagh

Abstract:

The purpose of this study was to evaluate dependence of PAGAT polymer gel dosimeter 1/T2 on different electron and photon energies as well as on different mean dose rates for a standard clinically used Co-60 therapy unit and an ELECTA linear accelerator. A multi echo sequence with 32 equidistant echoes was used for the evaluation of irradiated polymer gel dosimeters. The optimal post-manufacture irradiation and post imaging times were both determined to be one day. The sensitivity of PAGAT polymer gel dosimeter with irradiation of photon and electron beams was represented by the slope of calibration curve in the linear region measured for each modality. The response of PAGAT gel with photon and electron beams is very similar in the lower dose region. The R2-dose response was linear up to 30Gy. In electron beams the R2-dose response for doses less than 3Gy is not exact, but in photon beams the R2-dose response for doses less than 2Gy is not exact. Dosimeter energy dependence was studied for electron energies of 4, 12 and 18MeV and photon energies of 1.25, 4, 6 and 18MV. Dose rate dependence was studied in 6MeV electron beam and 6MV photon beam with the use of dose rates 80, 160, 240, 320, 400, and 480cGy/min. Evaluation of dosimeters were performed on Siemens Symphony, Germany 1.5T Scanner in the head coil. In this study no trend in polymer-gel dosimeter 1/T2 dependence was found on mean dose rate and energy for electron and photon beams.

Keywords: polymer gels, PAGAT gel, electron and photon beams, MRI

Procedia PDF Downloads 473

Authors: Rafik Djemili, Hocine Bourouba, M. C. Amara Korba

Abstract:

In this paper, we present a method in order to classify EEG signals for Brain-Computer Interfaces (BCI). EEG signals are first processed by means of spectral estimation methods to derive reliable features before classification step. Spectral estimation methods used are standard periodogram and the periodogram calculated by the Welch method; both methods are compared with Logarithm of Band Power (logBP) features. In the method proposed, we apply Linear Discriminant Analysis (LDA) followed by Support Vector Machine (SVM). Classification accuracy reached could be as high as 85%, which proves the effectiveness of classification of EEG signals based BCI using spectral methods.

Keywords: brain-computer interface, motor imagery, electroencephalogram, linear discriminant analysis, support vector machine

Procedia PDF Downloads 499

Authors: Lidija R. Jevrić, Sanja O. Podunavac-Kuzmanović, Strahinja Z. Kovačević

Abstract:

The present study deals with the selection of the most suitable quantitative structure-retention relationship (QSRR) models which should be used in prediction of the retention behavior of basic, neutral, acidic and phenolic pesticides which belong to different classes: fungicides, herbicides, metabolites, insecticides and plant growth regulators. Sum of ranking differences (SRD) approach can give a different point of view on selection of the most consistent QSRR model. SRD approach can be applied not only for ranking of the QSRR models, but also for detection of similarity or dissimilarity among them. Applying the SRD analysis, the most similar models can be found easily. In this study, selection of the best model was carried out on the basis of the reference ranking (“golden standard”) which was defined as the row average values of logarithm of retention time (logtr) defined by high performance liquid chromatography (HPLC). Also, SRD analysis based on experimental logtr values as reference ranking revealed similar grouping of the established QSRR models already obtained by hierarchical cluster analysis (HCA).

Keywords: chemometrics, chromatography, pesticides, sum of ranking differences

Procedia PDF Downloads 375

Authors: Ana Vieira, Fernando Lau, João Pedro Mortágua, Luís Cruz, Rui Santos

Abstract:

The air transport impact on environment is more than ever a limitative obstacle to the aeronautical industry continuous growth. Over the last decades, considerable effort has been carried out in order to obtain quieter aircraft solutions, whether by changing the original design or investigating more silent maneuvers. The noise propagated by rotating surfaces is one of the most important sources of annoyance, being present in most aerial vehicles. Bearing this is mind, CEIIA developed a new computational chain for noise prediction with in-house software tools to obtain solutions in relatively short time without using excessive computer resources. This work is based on the new acoustic tool, which aims to predict the rotor noise generated during steady and maneuvering flight, making use of the flexibility of the C language and the advantages of GPU programming in terms of velocity. The acoustic tool is based in the Formulation 1A of Farassat, capable of predicting two important types of noise: the loading and thickness noise. The present work describes the most important features of the acoustic tool, presenting its most relevant results and framework analyses for helicopters and UAV quadrotors.

Keywords: rotor noise, acoustic tool, GPU Programming, UAV noise

Procedia PDF Downloads 401

Authors: Emel Kızılkaya Aydogan, Mihrimah Ozmen, Yılmaz Delice

Abstract:

In recent days, there is a change and the ongoing development of the telecommunications sector in the global market. In this sector, churn analysis techniques are commonly used for analysing why some customers terminate their service subscriptions prematurely. In addition, customer churn is utmost significant in this sector since it causes to important business loss. Many companies make various researches in order to prevent losses while increasing customer loyalty. Although a large quantity of accumulated data is available in this sector, their usefulness is limited by data quality and relevance. In this paper, a cost-sensitive feature selection framework is developed aiming to obtain the feature reducts to predict customer churn. The framework is a cost based optional pre-processing stage to remove redundant features for churn management. In addition, this cost-based feature selection algorithm is applied in a telecommunication company in Turkey and the results obtained with this algorithm.

Keywords: churn prediction, data mining, decision-theoretic rough set, feature selection

Procedia PDF Downloads 446

Authors: Ghazwan Al-Haji, Adeyemi Adedokun

Abstract:

Traffic safety at intersections is highly represented, given the fact that accidents occur randomly in time and space. It is necessary to judge whether the intersection is dangerous or not based on short-term observations, and not waiting for many years of assessing historical accident data. There are active and pro-active road infrastructure safety methods for assessing safety at intersections. This study aims to investigate the use of quantitative and qualitative pre-observational methods as the best practice for accident prediction, future black spot identification, and treatment. Historical accident data from STRADA (the Swedish Traffic Accident Data Acquisition) was used within Norrkoping city in Sweden. The ADT (Average Daily Traffic), capacity and speed were used to predict accident rates. Locations with the highest accident records and predicted accident counts were identified and hence audited qualitatively by using Street Audit. The results from these quantitative and qualitative methods were analyzed, validated and compared. The paper provides recommendations on the used methods as well as on how to reduce the accident occurrence at the chosen intersections.

Keywords: intersections, traffic conflict, traffic safety, street audit, accidents predictions

Procedia PDF Downloads 233

Authors: Sergey Svita, Valeriy Astapenko

Abstract:

The study is devoted to theoretical analysis of ultrashort electromagnetic pulses (USP) absorption on gold nanospheres. Dependencies of USP energy absorption on nanospheres placed in various matrix are compared. The results of calculation of absorbed energy on gold nanospheres as a function of ultrashort electromagnetic pulse carrier frequency and number of pulse cycles of carrier frequency show strong non-linear dependence of absorbed energy on number of cycles of carrier frequency, but for relatively large number of cycles on USP carrier frequency it goes to linear dependence.

Keywords: ultrashort electromagnetic pulses, absorption, nanospheres, theoretical research

Procedia PDF Downloads 256

Authors: Jin Ho Kim, Yujeong Shin, Seong-Jin Cho, Dong-Jin Kim, U-Syn Ha

Abstract:

Due to the development of the information industry and technologies, cellular phones have must not only function to communicate, but also have functions such as the Internet, e-banking, entertainment, etc. These phones are called smartphones. The performance of smartphones has improved, because of the various functions of smartphones, and the capacity of the battery has been increased gradually. Recently, linear generators have been embedded in smartphones in order to recharge the smartphone's battery. In this study, optimization is performed and an array change of permanent magnets is examined in order to increase efficiency. We propose an optimal design using design of experiments (DOE) to maximize the generated induced voltage. The thickness of the poleshoe and permanent magnet (PM), the height of the poleshoe and PM, and the thickness of the coil are determined to be design variables. We made 25 sampling points using an orthogonal array according to four design variables. We performed electromagnetic finite element analysis to predict the generated induced voltage using the commercial electromagnetic analysis software ANSYS Maxwell. Then, we made an approximate model using the Kriging algorithm, and derived optimal values of the design variables using an evolutionary algorithm. The commercial optimization software PIAnO (Process Integration, Automation, and Optimization) was used with these algorithms. The result of the optimization shows that the generated induced voltage is improved.

Keywords: smartphone, linear generator, design of experiment, approximate model, optimal design

Procedia PDF Downloads 345

Authors: Gabriele Borg, Alexei Debono, Charlie Abela

Abstract:

There on a constant rise in the availability of high volumes of data gathered from multiple sources, resulting in an abundance of unprocessed information that can be used to monitor patterns and trends in user behaviour. Similarly, year after year, Malta is also constantly experiencing ongoing population growth and an increase in mobilization demand. This research takes advantage of data which is continuously being sourced and converting it into useful information related to the traffic problem on the Maltese roads. The scope of this paper is to provide a methodology to create a custom dataset (MalTra - Malta Traffic) compiled from multiple participants from various locations across the island to identify the most common routes taken to expose the main areas of activity. This use of big data is seen being used in various technologies and is referred to as ITSs (Intelligent Transportation Systems), which has been concluded that there is significant potential in utilising such sources of data on a nationwide scale. Furthermore, a series of traffic prediction graph neural network models are conducted to compare MalTra to large-scale traffic datasets.

Keywords: graph neural networks, traffic management, big data, mobile data patterns

Procedia PDF Downloads 131

Authors: Daniel James Pitchforth, Timothy James Rogers, Ulf Tyge Tygesen, Elizabeth Jane Cross

Abstract:

Wave loading is a primary cause of fatigue within offshore structures and its quantification presents a challenging and important subtask within the SHM framework. The accurate representation of physics in such environments is difficult, however, driving the development of data-driven techniques in recent years. Within many industrial applications, empirical laws remain the preferred method of wave loading prediction due to their low computational cost and ease of implementation. This paper aims to develop an approach that combines data-driven Gaussian process models with physical empirical solutions for wave loading, including Morison’s Equation. The aim here is to incorporate physics directly into the covariance function (kernel) of the Gaussian process, enforcing derived behaviors whilst still allowing enough flexibility to account for phenomena such as vortex shedding, which may not be represented within the empirical laws. The combined approach has a number of advantages, including improved performance over either component used independently and interpretable hyperparameters.

Keywords: offshore structures, Gaussian processes, Physics informed machine learning, Kernel design

Procedia PDF Downloads 192

Authors: Woo-Tai Jung, Sung-Yong Choi, Young-Hwan Park

Abstract:

The composite pavement system considered in this paper is composed of a functional surface layer, a fiber reinforced asphalt middle layer and a fiber reinforced lean concrete base layer. The mix design of the fiber reinforced lean concrete corresponds to the mix composition of conventional lean concrete but reinforced by fibers. The quasi-absence of research on the durability or long-term performances (fatigue, creep, etc.) of such mix design stresses the necessity to evaluate experimentally the long-term characteristics of this layer composition. This study tests the creep characteristics as one of the long-term characteristics of the fiber reinforced lean concrete layer for composite pavement using a new creep device. The test results reveal that the lean concrete mixed with fiber reinforcement and fly ash develops smaller creep than the conventional lean concrete. The results of the application of the CEB-FIP prediction equation indicate that a modified creep prediction equation should be developed to fit with the new mix design of the layer.

Keywords: creep, lean concrete, pavement, fiber reinforced concrete, base

Procedia PDF Downloads 522

Authors: Tunjo Perič, Marin Fatović

Abstract:

In this paper a new methodology for vendor selection and supply quotas determination (VSSQD) is proposed. The problem of VSSQD is solved by the model that combines revised weighting method for determining the objective function coefficients, and a multiple objective linear programming (MOLP) method based on the cooperative game theory for VSSQD. The criteria used for VSSQD are: (1) purchase costs and (2) product quality supplied by individual vendors. The proposed methodology is tested on the example of flour purchase for a bakery with two decision makers.

Keywords: cooperative game theory, multiple objective linear programming, revised weighting method, vendor selection

Procedia PDF Downloads 358

Authors: A. M. Sagir

Abstract:

The aim of this paper is to investigate the performance of the developed linear multistep block method for solving first order initial value problem of Ordinary Differential Equations (ODEs). The method calculates the numerical solution at three points simultaneously and produces three new equally spaced solution values within a block. The continuous formulations enable us to differentiate and evaluate at some selected points to obtain three discrete schemes, which were used in block form for parallel or sequential solutions of the problems. A stability analysis and efficiency of the block method are tested on ordinary differential equations involving practical applications, and the results obtained compared favorably with the exact solution. Furthermore, comparison of error analysis has been developed with the help of computer software.

Keywords: block method, first order ordinary differential equations, linear multistep, self-starting

Procedia PDF Downloads 306

Authors: Abdi Ammar, Ouazir Youcef, Laissaoui Abdelmalek

Abstract:

In this work, an approach based on transmission lines theory is presented. It is exploited for the calculation of overvoltage created by direct impacts of lightning waves on a guard cable of an overhead high-voltage line. First, we show the theoretical developments leading to the propagation equation, its discretization by finite difference time domain method (FDTD), and the resulting linear algebraic equations, followed by the calculation of the linear parameters of the line. The second step consists of solving the transmission lines system of equations by the FDTD method. This enabled us to determine the spatio-temporal evolution of the induced overvoltage.

Keywords: lightning surge, transient overvoltage, eddy current, FDTD, electromagnetic compatibility, ground wire

Procedia PDF Downloads 83

Authors: A. H. Elkholy, D. M. Almutairi

Abstract:

The successful and efficient execution of a forging process is dependent upon the correct analysis of loading and metal flow of blanks. This paper investigates the Upper Bound Technique (UBT) and its application in the analysis of open forging process when a possibility of blank bulging exists. The UBT is one of the energy rate minimization methods for the solution of metal forming process based on the upper bound theorem. In this regards, the kinematically admissible velocity field is obtained by minimizing the total forging energy rate. A computer program is developed in this research to implement the UBT. The significant advantages of this method is the speed of execution while maintaining a fairly high degree of accuracy and the wide prediction capability. The information from this analysis is useful for the design of forging processes and dies. Results for the prediction of forging loads and stresses, metal flow and surface profiles with the assured benefits in terms of press selection and blank preform design are outlined in some detail. The obtained predictions are ready for comparison with both laboratory and industrial results.

Keywords: forging, upper bound technique, metal forming, forging energy, forging die/platen

Procedia PDF Downloads 293

Authors: Sun-Young Jang, Sung-Ah Kim, Dongyoun Shin

Abstract:

Current real-estate value estimation, difficult for laymen, usually is performed by specialists. This paper presents an automated estimation process based on big data and machine-learning technology that calculates influences of building conditions on real-estate price measurement. The present study analyzed actual building sales sample data for Nonhyeon-dong, Gangnam-gu, Seoul, Korea, measuring the major influencing factors among the various building conditions. Further to that analysis, a prediction model was established and applied using RapidMiner Studio, a graphical user interface (GUI)-based tool for derivation of machine-learning prototypes. The prediction model is formulated by reference to previous examples. When new examples are applied, it analyses and predicts accordingly. The analysis process discerns the crucial factors effecting price increases by calculation of weighted values. The model was verified, and its accuracy determined, by comparing its predicted values with actual price increases.

Keywords: apartment complex, big data, life-cycle building value analysis, machine learning

Procedia PDF Downloads 374

Authors: Sergey Kopysov, Nikita Nedozhogin, Leonid Tonkov

Abstract:

The article presents a parallel iterative solver for large sparse linear systems which can be used on a heterogeneous platform. Traditionally, the problem of solving linear systems does not scale well on multi-CPU/multi-GPUs clusters. For example, most of the attempts to implement the classical conjugate gradient method were at best counted in the same amount of time as the problem was enlarged. The paper proposes the pipelined variant of the conjugate gradient method (PCG), a formulation that is potentially better suited for hybrid CPU/GPU computing since it requires only one synchronization point per one iteration instead of two for standard CG. The standard and pipelined CG methods need the vector entries generated by the current GPU and other GPUs for matrix-vector products. So the communication between GPUs becomes a major performance bottleneck on multi GPU cluster. The article presents an approach to minimize the communications between parallel parts of algorithms. Additionally, computation and communication can be overlapped to reduce the impact of data exchange. Using the pipelined version of the CG method with one synchronization point, the possibility of asynchronous calculations and communications, load balancing between the CPU and GPU for solving the large linear systems allows for scalability. The algorithm is implemented with the combined use of technologies: MPI, OpenMP, and CUDA. We show that almost optimum speed up on 8-CPU/2GPU may be reached (relatively to a one GPU execution). The parallelized solver achieves a speedup of up to 5.49 times on 16 NVIDIA Tesla GPUs, as compared to one GPU.

Keywords: conjugate gradient, GPU, parallel programming, pipelined algorithm

Procedia PDF Downloads 165