Search results for: classifiers ensemble.

Authors: Zhaoan Wang

Abstract:

Research in predictive maintenance modeling has improved in the recent years to predict failures and needed maintenance with high accuracy, saving cost and improving manufacturing efficiency. However, classic prediction models provide little valuable insight towards the most important features contributing to the failure. By analyzing and quantifying feature importance in predictive maintenance models, cost saving can be optimized based on business goals. First, multiple classifiers are evaluated with cross-validation to predict the multi-class of failures. Second, predictive performance with features provided by different feature selection algorithms are further analyzed. Third, features selected by different algorithms are ranked and combined based on their predictive power. Finally, linear explainer SHAP (SHapley Additive exPlanations) is applied to interpret classifier behavior and provide further insight towards the specific roles of features in both local predictions and global model behavior. The results of the experiments suggest that certain features play dominant roles in predictive models while others have significantly less impact on the overall performance. Moreover, for multi-class prediction of machine failures, the most important features vary with type of machine failures. The results may lead to improved productivity and cost saving by prioritizing sensor deployment, data collection, and data processing of more important features over less importance features.

Keywords: Automated supply chain, intelligent manufacturing, predictive maintenance machine learning, feature engineering, model interpretation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1934

Authors: Abobaker Mohammed Qasem Farhan, Shangming Yang, Mohammed Al-Nehari

Abstract:

Pneumonia is a kind of lung disease that creates congestion in the chest. Such pneumonic conditions lead to loss of life due to the severity of high congestion. Pneumonic lung disease is caused by viral pneumonia, bacterial pneumonia, or COVID-19 induced pneumonia. The early prediction and classification of such lung diseases help reduce the mortality rate. We propose the automatic Computer-Aided Diagnosis (CAD) system in this paper using the deep learning approach. The proposed CAD system takes input from raw computerized tomography (CT) scans of the patient's chest and automatically predicts disease classification. We designed the Hybrid Deep Learning Algorithm (HDLA) to improve accuracy and reduce processing requirements. The raw CT scans are pre-processed first to enhance their quality for further analysis. We then applied a hybrid model that consists of automatic feature extraction and classification. We propose the robust 2D Convolutional Neural Network (CNN) model to extract the automatic features from the pre-processed CT image. This CNN model assures feature learning with extremely effective 1D feature extraction for each input CT image. The outcome of the 2D CNN model is then normalized using the Min-Max technique. The second step of the proposed hybrid model is related to training and classification using different classifiers. The simulation outcomes using the publicly available dataset prove the robustness and efficiency of the proposed model compared to state-of-art algorithms.

Keywords: CT scans, COVID-19, deep learning, image processing, pneumonia, lung disease.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 478

Authors: Rohit Kumar Shrivastava, Amalia Thomas, Nathalie Vriend, Stefan Luding

Abstract:

A mechanical wave or vibration propagating through granular media exhibits a specific signature in time. A coherent pulse or wavefront arrives first with multiply scattered waves (coda) arriving later. The coherent pulse is micro-structure independent i.e. it depends only on the bulk properties of the disordered granular sample, the sound wave velocity of the granular sample and hence bulk and shear moduli. The coherent wavefront attenuates (decreases in amplitude) and broadens with distance from its source. The pulse attenuation and broadening effects are affected by disorder (polydispersity; contrast in size of the granules) and have often been attributed to dispersion and scattering. To study the effect of disorder and initial amplitude (non-linearity) of the pulse imparted to the system on the coherent wavefront, numerical simulations have been carried out on one-dimensional sets of particles (granular chains). The interaction force between the particles is given by a Hertzian contact model. The sizes of particles have been selected randomly from a Gaussian distribution, where the standard deviation of this distribution is the relevant parameter that quantifies the effect of disorder on the coherent wavefront. Since, the coherent wavefront is system configuration independent, ensemble averaging has been used for improving the signal quality of the coherent pulse and removing the multiply scattered waves. The results concerning the width of the coherent wavefront have been formulated in terms of scaling laws. An experimental set-up of photoelastic particles constituting a granular chain is proposed to validate the numerical results.

Keywords: Discrete elements, Hertzian Contact, polydispersity, weakly nonlinear, wave propagation.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 876

Authors: Hashim Isam Jameel Al-Safi, P. Ranjan Sarukkalige

Abstract:

Climate change is likely to impact the Australian continent by changing the trends of rainfall, increasing temperature, and affecting the accessibility of water quantity and quality. This study investigates the possible impacts of future climate change on the hydrological system of the Harvey River catchment in Western Australia by using the conceptual modelling approach (HBV mode). Daily observations of rainfall and temperature and the long-term monthly mean potential evapotranspiration, from six weather stations, were available for the period (1961-2015). The observed streamflow data at Clifton Park gauging station for 33 years (1983-2015) in line with the observed climate variables were used to run, calibrate and validate the HBV-model prior to the simulation process. The calibrated model was then forced with the downscaled future climate signals from a multi-model ensemble of fifteen GCMs of the CMIP3 model under three emission scenarios (A2, A1B and B1) to simulate the future runoff at the catchment outlet. Two periods were selected to represent the future climate conditions including the mid (2046-2065) and late (2080-2099) of the 21^st century. A control run, with the reference climate period (1981-2000), was used to represent the current climate status. The modelling outcomes show an evident reduction in the mean annual streamflow during the mid of this century particularly for the A1B scenario relative to the control run. Toward the end of the century, all scenarios show a relatively high reduction trends in the mean annual streamflow, especially the A1B scenario, compared to the control run. The decline in the mean annual streamflow ranged between 4-15% during the mid of the current century and 9-42% by the end of the century.

Keywords: Climate change impact, Harvey catchment, HBV model, hydrological modelling, GCMs, LARS-WG, Australia.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1387

Authors: L. Melzerová, T. Janda, M. Šejnoha, J. Šejnoha

Abstract:

Two finite element (FEM) models are presented in this paper to address the random nature of the response of glued timber structures made of wood segments with variable elastic moduli evaluated from 3600 indentation measurements. This total database served to create the same number of ensembles as was the number of segments in the tested beam. Statistics of these ensembles were then assigned to given segments of beams and the Latin Hypercube Sampling (LHS) method was called to perform 100 simulations resulting into the ensemble of 100 deflections subjected to statistical evaluation. Here, a detailed geometrical arrangement of individual segments in the laminated beam was considered in the construction of two-dimensional FEM model subjected to in fourpoint bending to comply with the laboratory tests. Since laboratory measurements of local elastic moduli may in general suffer from a significant experimental error, it appears advantageous to exploit the full scale measurements of timber beams, i.e. deflections, to improve their prior distributions with the help of the Bayesian statistical method. This, however, requires an efficient computational model when simulating the laboratory tests numerically. To this end, a simplified model based on Mindlin’s beam theory was established. The improved posterior distributions show that the most significant change of the Young’s modulus distribution takes place in laminae in the most strained zones, i.e. in the top and bottom layers within the beam center region. Posterior distributions of moduli of elasticity were subsequently utilized in the 2D FEM model and compared with the original simulations.

Keywords: Bayesian inference, FEM, four point bending test, laminated timber, parameter estimation, prior and posterior distribution, Young’s modulus.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2164

Authors: Dusittorn Ngamying

Abstract:

This investigation is focused on using of Mon dance in Pathum Thani Province’s tradition and has the following objectives: 1) to study the background of Mon dance in Pathum Thani Province; 2) to study Mon dance in Pathum Thani Province; and 3) to study of using Mon dance in Pathum Thani province’s tradition. This qualitative research was conducted in Pathum Thani province (in the central of Thailand). Data was collected from documentary study and field data by means of observation, interview, and group discussion. Workshops were also held with a total of 100 attendees, comprised of 20 key informants, 40 casual informants and 40 general informants. Data was validated using the triangulation technique and the findings are presented using the descriptive analysis. The results of the study show that the historical background of Mon dance in Pathum Thani Province initiated during the war evacuation from Martaban (south of Burma) to settle down in Sam Khok, Pathum Thani Province in Ayutthaya period to Rattanakosin. The study found that Mon dance typically consists of 12-13 dancing process. The melodies have 12-13 songs. Piphat Mon (Mon traditional music ensemble) is used in the performance. Performers are dressed in Mon traditional costumes. The performers are 6-12 women and depending on the employer’s demands. Length of the performance varies from the duration of music orchestration. Rituals and customs performed are paying homage to teachers before the performance. The offerings are composed of flowers, incense sticks, candles, money gifts which are well arranged on a tray with pedestal, and also liquors, tobaccos and pure water for asking propitiousness. For the use of Mon dance in Pathum Thani Province’s tradition, it is found that the dance is commonly performed in the funeral ceremonial tradition at present because the physical postures of the performance are considered graceful and exquisite. In addition, as for its value, it has long been believed since the ancient times that Mon dance was a sacred thing considered as the dignity and glorification especially for funeral ceremonies of priest or royal hierarchy classes. However, Mon dance has continued to be used in the traditions associated with Mon people activities in Pathum Thani Province for instance customary welcome for honor guest and Songkran festival.

Keywords: Mon dance, traditions, Pathum Thani Province.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1452

Authors: Maad Kamal Al-Anni

Abstract:

This study is about an algorithmic dependence of Artificial Neural Network on Multilayer Perceptron (MPL) pertaining to the classification and clustering presentations for Mobile Adhoc Network vulnerabilities. Moreover, mobile ad hoc network (MANET) is ubiquitous intelligent internetworking devices in which it has the ability to detect their environment using an autonomous system of mobile nodes that are connected via wireless links. Security affairs are the most important subject in MANET due to the easy penetrative scenarios occurred in such an auto configuration network. One of the powerful techniques used for inspecting the network packets is Intrusion Detection System (IDS); in this article, we are going to show the effectiveness of artificial neural networks used as a machine learning along with stochastic approach (information gain) to classify the malicious behaviors in simulated network with respect to different IDS techniques. The monitoring agent is responsible for detection inference engine, the audit data is collected from collecting agent by simulating the node attack and contrasted outputs with normal behaviors of the framework, whenever. In the event that there is any deviation from the ordinary behaviors then the monitoring agent is considered this event as an attack , in this article we are going to demonstrate the  signature-based IDS approach in a MANET by implementing the back propagation algorithm over ensemble-based Traffic Table (TT), thus the signature of malicious behaviors or undesirable activities are often significantly prognosticated and efficiently figured out, by increasing the parametric set-up of Back propagation algorithm during the experimental results which empirically shown its effectiveness  for the ratio of detection index up to 98.6 percentage. Consequently it is proved in empirical results in this article, the performance matrices are also being included in this article with Xgraph screen show by different through puts like Packet Delivery Ratio (PDR), Through Put(TP), and Average Delay(AD).

Keywords: Mobile ad hoc network, MANET, intrusion detection system, back propagation algorithm, neural networks, traffic table, multilayer perceptron, feed-forward back-propagation, network simulator 2.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 886

Authors: Lily Ingsrisawang, Tasanee Nacharoen

Abstract:

The problems arising from unbalanced data sets generally appear in real world applications. Due to unequal class distribution, many researchers have found that the performance of existing classifiers tends to be biased towards the majority class. The k-nearest neighbors’ nonparametric discriminant analysis is a method that was proposed for classifying unbalanced classes with good performance. In this study, the methods of discriminant analysis are of interest in investigating misclassification error rates for classimbalanced data of three diabetes risk groups. The purpose of this study was to compare the classification performance between parametric discriminant analysis and nonparametric discriminant analysis in a three-class classification of class-imbalanced data of diabetes risk groups. Data from a project maintaining healthy conditions for 599 employees of a government hospital in Bangkok were obtained for the classification problem. The employees were divided into three diabetes risk groups: non-risk (90%), risk (5%), and diabetic (5%). The original data including the variables of diabetes risk group, age, gender, blood glucose, and BMI were analyzed and bootstrapped for 50 and 100 samples, 599 observations per sample, for additional estimation of the misclassification error rate. Each data set was explored for the departure of multivariate normality and the equality of covariance matrices of the three risk groups. Both the original data and the bootstrap samples showed nonnormality and unequal covariance matrices. The parametric linear discriminant function, quadratic discriminant function, and the nonparametric k-nearest neighbors’ discriminant function were performed over 50 and 100 bootstrap samples and applied to the original data. Searching the optimal classification rule, the choices of prior probabilities were set up for both equal proportions (0.33: 0.33: 0.33) and unequal proportions of (0.90:0.05:0.05), (0.80: 0.10: 0.10) and (0.70, 0.15, 0.15). The results from 50 and 100 bootstrap samples indicated that the k-nearest neighbors approach when k=3 or k=4 and the defined prior probabilities of non-risk: risk: diabetic as 0.90: 0.05:0.05 or 0.80:0.10:0.10 gave the smallest error rate of misclassification. The k-nearest neighbors approach would be suggested for classifying a three-class-imbalanced data of diabetes risk groups.

Keywords: Bootstrap, diabetes risk groups, error rate, k-nearest neighbors.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1965

Authors: Saurabh Sharma, Yong Sun, Bruce Vernham

Abstract:

Accurate prediction of NOx emission is a continuous challenge in the field of diesel engine-out emission modeling. Performing experiments for each conditions and scenario cost significant amount of money and man hours, therefore model-based development strategy has been implemented in order to solve that issue. NOx formation is highly dependent on the burn gas temperature and the O₂ concentration inside the cylinder. The current empirical models are developed by calibrating the parameters representing the engine operating conditions with respect to the measured NOx. This makes the prediction of purely empirical models limited to the region where it has been calibrated. An alternative solution to that is presented in this paper, which focus on the utilization of in-cylinder combustion parameters to form a predictive semi-empirical NOx model. The result of this work is shown by developing a fast and predictive NOx model by using the physical parameters and empirical correlation. The model is developed based on the steady state data collected at entire operating region of the engine and the predictive combustion model, which is developed in Gamma Technology (GT)-Power by using Direct Injected (DI)-Pulse combustion object. In this approach, temperature in both burned and unburnt zone is considered during the combustion period i.e. from Intake Valve Closing (IVC) to Exhaust Valve Opening (EVO). Also, the oxygen concentration consumed in burnt zone and trapped fuel mass is also considered while developing the reported model.  Several statistical methods are used to construct the model, including individual machine learning methods and ensemble machine learning methods. A detailed validation of the model on multiple diesel engines is reported in this work. Substantial numbers of cases are tested for different engine configurations over a large span of speed and load points. Different sweeps of operating conditions such as Exhaust Gas Recirculation (EGR), injection timing and Variable Valve Timing (VVT) are also considered for the validation. Model shows a very good predictability and robustness at both sea level and altitude condition with different ambient conditions. The various advantages such as high accuracy and robustness at different operating conditions, low computational time and lower number of data points requires for the calibration establishes the platform where the model-based approach can be used for the engine calibration and development process. Moreover, the focus of this work is towards establishing a framework for the future model development for other various targets such as soot, Combustion Noise Level (CNL), NO₂/NOx ratio etc.

Keywords: Diesel engine, machine learning, NOx emission, semi-empirical.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 796