Search results for: agile mixture model

Authors: Shrabani Ghosh

Abstract:

Automatic offensive language detection in social media has become a stirring task in today's NLP. Manual Offensive language detection is tedious and laborious work where automatic methods based on machine learning are only alternatives. Previous works have done sentiment analysis over social media in different ways such as supervised, semi-supervised, and unsupervised manner. Domain adaptation in a semi-supervised way has also been explored in NLP, where the source domain and the target domain are different. In domain adaptation, the source domain usually has a large amount of labeled data, while only a limited amount of labeled data is available in the target domain. Pretrained transformers like BERT, RoBERTa models are fine-tuned to perform text classification in an unsupervised manner to perform further pre-train masked language modeling (MLM) tasks. In previous work, hate speech detection has been explored in Gab.ai, which is a free speech platform described as a platform of extremist in varying degrees in online social media. In domain adaptation process, Twitter data is used as the source domain, and Gab data is used as the target domain. The performance of domain adaptation also depends on the cross-domain similarity. Different distance measure methods such as L2 distance, cosine distance, Maximum Mean Discrepancy (MMD), Fisher Linear Discriminant (FLD), and CORAL have been used to estimate domain similarity. Certainly, in-domain distances are small, and between-domain distances are expected to be large. The previous work finding shows that pretrain masked language model (MLM) fine-tuned with a mixture of posts of source and target domain gives higher accuracy. However, in-domain performance of the hate classifier on Twitter data accuracy is 71.78%, and out-of-domain performance of the hate classifier on Gab data goes down to 56.53%. Recently self-supervised learning got a lot of attention as it is more applicable when labeled data are scarce. Few works have already been explored to apply self-supervised learning on NLP tasks such as sentiment classification. Self-supervised language representation model ALBERTA focuses on modeling inter-sentence coherence and helps downstream tasks with multi-sentence inputs. Self-supervised attention learning approach shows better performance as it exploits extracted context word in the training process. In this work, a self-supervised attention mechanism has been proposed to detect hate speech on Gab.ai. This framework initially classifies the Gab dataset in an attention-based self-supervised manner. On the next step, a semi-supervised classifier trained on the combination of labeled data from the first step and unlabeled data. The performance of the proposed framework will be compared with the results described earlier and also with optimized outcomes obtained from different optimization techniques.

Keywords: attention learning, language model, offensive language detection, self-supervised learning

Procedia PDF Downloads 107

Authors: Nicolas Plamondon, Stuart Atkinson, Shuzi Zhou

Abstract:

The insurance industry is usually not the first topic one has in mind when thinking about applications of data science. However, the use of data science in the finance and insurance industry is growing quickly for several reasons, including an abundance of reliable customer data, ferocious competition requiring more accurate pricing, etc. Among the top use cases of data science, we find pricing optimization, customer segmentation, customer risk assessment, fraud detection, marketing, and triage analytics. The objective of this paper is to present an application of the generalized additive model (GAM) on a commercial automobile insurance product: an individually rated commercial automobile. These are vehicles used for commercial purposes, but for which there is not enough volume to apply pricing to several vehicles at the same time. The GAM model was selected as an improvement over GLM for its ease of use and its wide range of applications. The model was trained using the largest split of the data to determine model parameters. The remaining part of the data was used as testing data to verify the quality of the modeling activity. We used the Gini coefficient to evaluate the performance of the model. For long-term monitoring, commonly used metrics such as RMSE and MAE will be used. Another topic of interest in the insurance industry is to process of producing the model. We will discuss at a high level the interactions between the different teams with an insurance company that needs to work together to produce a model and then monitor the performance of the model over time. Moreover, we will discuss the regulations in place in the insurance industry. Finally, we will discuss the maintenance of the model and the fact that new data does not come constantly and that some metrics can take a long time to become meaningful.

Keywords: insurance, data science, modeling, monitoring, regulation, processes

Procedia PDF Downloads 76

Authors: T. Kobayashi, K. Shiba, T. Kaburagi, Y. Kurihara

Abstract:

For the elderly living alone, falls can be a serious problem encountered in daily life. Some elderly people are unable to stand up without the assistance of a caregiver. They may become unconscious after a fall, which can lead to serious aftereffects such as hypothermia, dehydration, and sometimes even death. We treat the subject as an inverted pendulum and model its angle from the equilibrium position and its angular velocity. As the model is non-linear, we implement the filtering method with a particle filter which can estimate true states of the non-linear model. In order to evaluate the accuracy of the particle filter estimation results, we calculate the root mean square error (RMSE) between the estimated angle/angular velocity and the true values generated by the simulation. The experimental results give the highest accuracy RMSE of 0.0141 rad and 0.1311 rad/s for the angle and angular velocity, respectively.

Keywords: fall, microwave Doppler sensor, non-linear dynamics model, particle filter

Procedia PDF Downloads 217

Authors: K. Paradkar, S. N. Mudliar, A. Sharma, A. B. Pandit, R. A. Pandey

Abstract:

The study explores the hybrid combination of Hydrodynamic Cavitation (HC) and Subcritical Wet Air Oxidation-based pretreatment of complex industrial effluent to enhance the biodegradability selectively (without major COD destruction) to facilitate subsequent enhanced downstream processing via anaerobic or aerobic biological treatment. Advanced oxidation based techniques can be less efficient as standalone options and a hybrid approach by combining Hydrodynamic Cavitation (HC), and Wet Air Oxidation (WAO) can lead to a synergistic effect since both the options are based on common free radical mechanism. The HC can be used for initial turbulence and generation of hotspots which can begin the free radical attack and this agitating mixture then can be subjected to less intense WAO since initial heat (to raise the activation energy) can be taken care by HC alone. Lab-scale venturi-based hydrodynamic cavitation and wet air oxidation reactor with biomethanated distillery wastewater (BMDWW) as a model effluent was examined for establishing the proof-of-concept. The results indicated that for a desirable biodegradability index (BOD: COD - BI) enhancement (up to 0.4), the Cavitation (standalone) pretreatment condition was: 5 bar and 88 min reaction time with a COD reduction of 36 % and BI enhancement of up to 0.27 (initial BI - 0.17). The optimum WAO condition (standalone) was: 150oC, 6 bar and 30 minutes with 31% COD reduction and 0.33 BI. The hybrid pretreatment (combined Cavitation + WAO) worked out to be 23.18 min HC (at 5 bar) followed by 30 min WAO at 150oC, 6 bar, at which around 50% COD was retained yielding a BI of 0.55. FTIR & NMR analysis of pretreated effluent indicated dissociation and/or reorientation of complex organic compounds in untreated effluent to simpler organic compounds post-pretreatment.

Keywords: hybrid, hydrodynamic cavitation, wet air oxidation, biodegradability index

Procedia PDF Downloads 618

Authors: Pompa Marcello, Mauro Capocelli, Vincenzo Piemonte

Abstract:

This work focuses on a mathematical model able to describe the gastro-intestinal physiology and providing a rational tool for the design of an artificial gastro-intestinal system. This latter is mainly devoted to analyse the absorption and bioavailability of drugs and nutrients through in vitro tests in order to overcome (or, at least, to partially replace) in vivo trials. The provided model realizes a conjunction ring (with extended prediction capability) between in vivo tests and mechanical-laboratory models emulating the human body. On this basis, no empirical equations controlling the gastric emptying are implemented in this model as frequent in the cited literature and all the sub-unit and the related system of equations are physiologically based. More in detail, the model structure consists of six compartments (stomach, duodenum, jejunum, ileum, colon and blood) interconnected through pipes and valves. Paracetamol, Ketoprofen, Irbesartan and Ketoconazole are considered and analysed in this work as reference drugs. The mathematical model has been validated against in vivo literature data. Results obtained show a very good model reliability and highlight the possibility to realize tailored simulations for different couples patient-drug, including food adsorption dynamics.

Keywords: gastro-intestinal model, drugs bioavailability, paracetamol, ketoprofen

Procedia PDF Downloads 170

Authors: Yilei Song, Linlin Tian, Ning Zhao

Abstract:

Turbulence energetics and structures in the wake of large-scale wind turbines under the stably-stratified atmospheric boundary layer (SABL) can be complicated due to the presence of low-level jets (LLJs), a region of higher wind speeds than the geostrophic wind speed. With a modified one-k-equation, eddy viscosity model specified for atmospheric flows as the sub-grid scale (SGS) model, a realistic atmospheric state of the stable ABL is well reproduced by large-eddy simulation (LES) techniques. Corresponding to the precursor stably stratification, the detailed wake properties of a standard 5-MW wind turbine represented as an actuator line model are provided. An engineering model is proposed for wake prediction based on the simulation statistics and gets validated. Results confirm that the proposed wake model can provide good predictions for wind turbines under the SABL.

Keywords: large-eddy simulation, stably-stratified atmospheric boundary layer, wake model, wind turbine wake

Procedia PDF Downloads 174

Authors: Adrian Rettig, Silvan Schneider, Reto Tamburini, Mirko Kleingries, Ulf Christian Muller

Abstract:

Industrial adsorption processes are, mainly due to si-multaneous heat and mass transfer, characterized by a high level of complexity. The conception of such processes often does not take place systematically; instead scale-up/down respectively number-up/down methods based on existing systems are used. This paper shows how Modelica® can be used to develop a transient model enabling a more systematic design of such ad- and desorption components and processes. The core of this model is a lumped-element submodel of a single adsorbent grain, where the thermodynamic equilibria and the kinetics of the ad- and desorption processes are implemented and solved on the basis of mass-, momentum and energy balances. For validation of this submodel, a fixed bed adsorber, whose characteristics are described in detail in the literature, was modeled and simulated. The simulation results are in good agreement with the experimental results from the literature. Therefore, the model development will be continued, and the extended model will be applied to further adsorber types like rotor adsorbers and moving bed adsorbers.

Keywords: adsorption, desorption, linear driving force, dynamic model, Modelica®, integral equation approach

Procedia PDF Downloads 371

Authors: Chen-Yin Kuo, Yung-Hsin Lee

Abstract:

Previous studies applying residual income valuation (RIV) model generally use panel data and single-equation model to forecast stock prices. Unlike these, this paper uses Taiwan longitudinal data to estimate multi-equation time-series models such as Vector Autoregressive (VAR), Vector Error Correction Model (VECM), and conduct out-of-sample forecasting. Further, this work assesses their forecasting performance by two instruments. In favor of extant research, the major finding shows that VECM outperforms other three models in forecasting for three stock sectors over entire horizons. It implies that an error correction term containing long-run information contributes to improve forecasting accuracy. Moreover, the pattern of composite shows that at longer horizon, VECM produces the greater reduction in errors, and performs substantially better than VAR.

Keywords: residual income valuation model, vector error correction model, out of sample forecasting, forecasting accuracy

Procedia PDF Downloads 316

Authors: M. J. Kang, Y. N. Jo, H. H. Yoo

Abstract:

Pendulum tests were used to identify a stretch reflex and diagnose spasticity. Some researches tried to make a mathematical model to simulate the motions. Thighs are subject to compressive forces due to gravity during a pendulum test. Therefore, it affects knee trajectories. However, the most studies on the pendulum tests did not consider that conditions. We used Kelvin-Voight model as compression model of skin and muscles. In this study, we investigated viscoelastic behaviors of skin and muscles using gelatin blocks from experiments of the vibration of the compliantly supported beam. Then we calculated a dynamic stiffness and loss factors from the experiment and estimated a damping coefficient of the model. We also did pendulum tests of human lower limbs to validate the stiffness and damping coefficient of a skin model. To simulate the pendulum motion, we derive equations of motion. We used stretch reflex activation model to estimate muscle forces induced by the stretch reflex. To validate the results, we compared the activation with electromyography signals during experiments. The compression behavior of skin and muscles in this study can be applied to analyze sitting posture as wee as developing surgical techniques.

Keywords: Kelvin-Voight model, pendulum test, skin and muscles under compression, stretch reflex

Procedia PDF Downloads 447

Authors: Aymen Rhouma, Khaled Hcheichi, Sami Hafsi

Abstract:

The article presents an application of Fractional Model Predictive Control (FMPC) to a fractional order thermal system using Controlled Auto Regressive Integrated Moving Average (CARIMA) model obtained by discretization of a continuous fractional differential equation. Moreover, the output deviation approach is exploited to design the K -step ahead output predictor, and the corresponding control law is obtained by solving a quadratic cost function. Experiment results onto a thermal system are presented to emphasize the performances and the effectiveness of the proposed predictive controller.

Keywords: fractional model predictive control, fractional order systems, thermal system, predictive control

Procedia PDF Downloads 412

Authors: Belyayev P. E., Makeyeva I. R., Mastyuk D. A., Pigasov E. E.

Abstract:

Previously, the phrase ”hydrogen safety” was often used in terms of NPP safety. Due to the rise of interest to “green” and, particularly, hydrogen power engineering, the problem of hydrogen safety at industrial facilities has become ever more urgent. In Russia, the industrial production of hydrogen is meant to be performed by placing a chemical engineering plant near NPP, which supplies the plant with the necessary energy. In this approach, the production of hydrogen involves a wide range of combustible gases, such as methane, carbon monoxide, and hydrogen itself. Considering probable incidents, sudden combustible gas outburst into open space with further ignition is less dangerous by itself than ignition of the combustible mixture in the presence of many pipelines, reactor vessels, and any kind of fitting frames. Even ignition of 2100 cubic meters of the hydrogen-air mixture in open space gives velocity and pressure that are much lesser than velocity and pressure in Chapman-Jouguet condition and do not exceed 80 m/s and 6 kPa accordingly. However, the space blockage, the significant change of channel diameter on the way of flame propagation, and the presence of gas suspension lead to significant deflagration acceleration and to its transition into detonation or quasi-detonation. At the same time, process parameters acquired from the experiments at specific experimental facilities are not general, and their application to different facilities can only have a conventional and qualitative character. Yet, conducting deflagration and detonation experimental investigation for each specific industrial facility project in order to determine safe infrastructure unit placement does not seem feasible due to its high cost and hazard, while the conduction of numerical experiments is significantly cheaper and safer. Hence, the development of a numerical method that allows the description of reacting flows in domains with complex geometry seems promising. The base for this method is the modification of Kuropatenko method for calculating shock waves recently developed by authors, which allows using it in Eulerian coordinates. The current work contains the results of the development process. In addition, the comparison of numerical simulation results and experimental series with flame propagation in shock tubes with orifice plates is presented.

Keywords: CFD, reacting flow, DDT, gas explosion

Procedia PDF Downloads 90

Authors: Y. S. Yusoff, G. Streftaris, H. R Waters

Abstract:

Death within 30 days is an important factor to be looked into, as there is a significant risk of deaths immediately following or soon after, Myocardial Infarction (MI) or stroke. In this paper, we will model the deaths within 30 days following a Myocardial Infarction (MI) or stroke in the UK. We will see how the probabilities of sudden deaths from MI or stroke have changed over the period 1981-2000. We will model the sudden deaths using a Generalized Linear Model (GLM), fitted using the R statistical package, under a Binomial distribution for the number of sudden deaths. We parameterize our model using the extensive and detailed data from the Framingham Heart Study, adjusted to match UK rates. The results show that there is a reduction for the sudden deaths following a MI over time but no significant improvement for sudden deaths following a stroke.

Keywords: sudden deaths, myocardial infarction, stroke, ischemic heart disease

Procedia PDF Downloads 289

Authors: C. Castro-Egler, A. Durán-Escalante, A. García-González

Abstract:

This paper presents a method to generate a finite element model of the human auditory inner ear system. The geometric model has been realized using 2D images from a virtual model of temporal bones. A point cloud has been gotten manually from those images to construct a whole mesh with hexahedral elements. The main difference with the predecessor models is the spiral shape of the cochlea with its three scales completely defined: scala tympani, scala media and scala vestibuli; which are separate by basilar membrane and Reissner membrane. To validate this model, numerical simulations have been realised with two models: an isolated inner ear and a whole model of human auditory system. Ideal conditions of displacement are applied over the oval window in the isolated Inner Ear model. The whole model is made up of the outer auditory channel, the tympani, the ossicular chain, and the inner ear. The boundary condition for the whole model is 1Pa over the auditory channel entrance. The numerical simulations by FEM have been done using a harmonic analysis with a frequency range between 100-10.000 Hz with an interval of 100Hz. The following results have been carried out: basilar membrane displacement; the scala media pressure according to the cochlea length and the transfer function of the middle ear normalized with the pressure in the tympanic membrane. The basilar membrane displacements and the pressure in the scala media make it possible to validate the response in frequency of the basilar membrane.

Keywords: finite elements method, human auditory system model, numerical analysis, 3D modelling cochlea

Procedia PDF Downloads 363

Authors: Amr Mansour Mohsen, Hesham Ahmed Hassan, Amira M. Idrees

Abstract:

Emotions classification of text documents is applied to reveal if the document expresses a determined emotion from its writer. As different supervised methods are previously used for emotion documents’ classification, in this research we present a novel model that supports the classification algorithms for more accurate results by the support of TF-IDF measure. Different experiments have been applied to reveal the applicability of the proposed model, the model succeeds in raising the accuracy percentage according to the determined metrics (precision, recall, and f-measure) based on applying the refinement of the lexicon, integration of lexicons using different perspectives, and applying the TF-IDF weighting measure over the classifying features. The proposed model has also been compared with other research to prove its competence in raising the results’ accuracy.

Keywords: emotion detection, TF-IDF, WEKA tool, classification algorithms

Procedia PDF Downloads 484

Authors: Muhammad Shoaib, Hassan M Al-Swaidan

Abstract:

Saudi Arabia is the major date producer in the world. In order to maximize the production from date tree, pruning of the date trees is required annually. Large amount of this agriculture waste material (palm tree fronds) is available in Saudi Arabia and considered as an ideal source as a precursor for production of activated carbon (AC). The single step procedure for the preparation of micro porous activated carbon (AC) from Saudi date tree fronds using mixture of gases (N2 and CO2) is carried out at carbonization/activation temperature at 850°C and at different ramp rates of 10, 20 and 30 degree per minute. Alloy 330 horizontal reactor is used for tube furnace. Flow rate of nitrogen and carbon dioxide gases are kept at 150 ml/min and 50 ml/min respectively during the preparation. Characterization results reveal that the BET surface area, pore volume, and average pore diameter of the resulting activated carbon generally decreases with the increase in ramp rate. The activated carbon prepared at a ramp rate of 10 degrees/minute attains larger surface area and can offer higher potential to produce activated carbon of greater adsorption capacity from agriculture wastes such as date fronds. The BET surface areas of the activated carbons prepared at a ramp rate of 10, 20 and 30 degree/minute after 30 minutes activation time are 1094, 1020 and 515 m2/g, respectively. Scanning electron microscopy (SEM) for surface morphology, and FTIR for functional groups was carried out that also verified the same trend. Moreover, by increasing the ramp rate from 10 and 20 degrees/min the yield remains same, i.e. 18%, whereas at a ramp rate of 30 degrees/min the yield increases from 18 to 20%. Thus, it is feasible to produce high-quality micro porous activated carbon from date frond agro waste using N2 carbonization followed by physical activation with CO2 and N2 mixture. This micro porous activated carbon can be used as adsorbent of heavy metals from wastewater, NOx SOx emission adsorption from ambient air and electricity generation plants, purification of gases, sewage treatment and many other applications.

Keywords: activated carbon, date tree fronds, agricultural waste, applied chemistry

Procedia PDF Downloads 280

Authors: Jia-Jang Wu

Abstract:

This paper presents the scaling laws that provide the criteria of geometry and dynamic similitude between the full-size rotor-shaft system and its scale model, and can be used to predict the torsional vibration characteristics of the full-size rotor-shaft system by manipulating the corresponding data of its scale model. The scaling factors, which play fundamental roles in predicting the geometry and dynamic relationships between the full-size rotor-shaft system and its scale model, for torsional free vibration problems between scale and full-size rotor-shaft systems are firstly obtained from the equation of motion of torsional free vibration. Then, the scaling factor of external force (i.e., torque) required for the torsional forced vibration problems is determined based on the Newton’s second law. Numerical results show that the torsional free and forced vibration characteristics of a full-size rotor-shaft system can be accurately predicted from those of its scale models by using the foregoing scaling factors. For this reason, it is believed that the presented approach will be significant for investigating the relevant phenomenon in the scale model tests.

Keywords: torsional vibration, full-size model, scale model, scaling laws

Procedia PDF Downloads 397

Authors: Seyed Mohammadbagher Jafari, Elahe Mahdizadeh, Masomeh Ghahremani

Abstract:

Nowadays almost all organizations depend on information systems to run their businesses. Investment on information systems and their maintenance to keep them always in best situation to support firm business is one of the main issues for every organization. The new concept of cloud computing was developed as a technical and economic model to address this issue. In cloud computing the computing resources, including networks, applications, hardwares and services are configured as needed and are available at the moment of request. However, migration to cloud is not an easy task and there are many issues that should be taken into account. This study tries to provide a comprehensive model to assess a firm readiness to implement cloud computing. By conducting a systematic literature review, four dimensions of readiness were extracted which include technological, human, organizational and environmental dimensions. Every dimension has various criteria that have been discussed in details. This model provides a framework for cloud computing readiness assessment. Organizations that intend to migrate to cloud can use this model as a tool to assess their firm readiness before making any decision on cloud implementation.

Keywords: cloud computing, human readiness, organizational readiness, readiness assessment model

Procedia PDF Downloads 397

Authors: Shahram Payandeh

Abstract:

This paper presents an overview of a model which can be used as a part of a decision support system when fusing information from multiple sensing environment. Data fusion has been widely studied in the past few decades and numerous frameworks have been proposed to facilitate decision making process under uncertainties. Multi-sensor data fusion technology plays an increasingly significant role during people tracking and activity recognition. This paper presents an overview of a quantum model as a part of a decision-making process in the context of multi-sensor data fusion. The paper presents basic definitions and relationships associating the decision-making process and quantum model formulation in the presence of uncertainties.

Keywords: quantum model, sensor space, sensor network, decision support

Procedia PDF Downloads 227

Authors: K. Eryilmaz, G. Mercanoglu

Abstract:

Aim: Purification of the final product, which is the last step in the synthesis of lanthanide-labeled peptide compounds, can be accomplished by different methods. Among these methods, the two most commonly used methods are C18 solid phase extraction (SPE) and weak cation exchanger cartridge elution. SPE C18 solid phase extraction method yields high purity final product, while elution from the weak cation exchanger cartridge is pH dependent and ineffective in removing colloidal impurities. The aim of this work is to develop an additional purification method for the lanthanide-labeled peptide compound in cases where the desired radionuclidic and radiochemical purity of the final product can not be achieved because of pH problem or colloidal impurity. Material and Methods: For colloidal impurity formation, 3 mL of water for injection (WFI) was added to 30 mCi of 177LuCl3 solution and allowed to stand for 1 day. 177Lu-DOTATATE was synthesized using EZAG ML-EAZY module (10 mCi/mL). After synthesis, the final product was mixed with the colloidal impurity solution (total volume:13 mL, total activity: 40 mCi). The resulting mixture was trapped in SPE-C18 cartridge. The cartridge was washed with 10 ml saline to remove impurities to the waste vial. The product trapped in the cartridge was eluted with 2 ml of 50% ethanol and collected to the final product vial via passing through a 0.22μm filter. The final product was diluted with 10 mL of saline. Radiochemical purity before and after purification was analysed by HPLC method. (column: ACE C18-100A. 3µm. 150 x 3.0mm, mobile phase: Water-Acetonitrile-Trifluoro acetic acid (75:25:1), flow rate: 0.6 mL/min). Results: UV and radioactivity detector results in HPLC analysis showed that colloidal impurities were completely removed from the 177Lu-DOTATATE/ colloidal impurity mixture by purification method. Conclusion: The improved purification method can be used as an additional method to remove impurities that may result from the lanthanide-peptide synthesis in which the weak cation exchange purification technique is used as the last step. The purification of the final product and the GMP compliance (the final aseptic filtration and the sterile disposable system components) are two major advantages.

Keywords: lanthanide, peptide, labeling, purification, radionuclide, radiopharmaceutical, synthesis

Procedia PDF Downloads 164

Authors: Aref Ghafouri, Mohammad javad Mollakazemi, Farhad Asadi

Abstract:

In this paper, model order reduction method is used for approximation in linear and nonlinearity aspects in some experimental data. This method can be used for obtaining offline reduced model for approximation of experimental data and can produce and follow the data and order of system and also it can match to experimental data in some frequency ratios. In this study, the method is compared in different experimental data and influence of choosing of order of the model reduction for obtaining the best and sufficient matching condition for following the data is investigated in format of imaginary and reality part of the frequency response curve and finally the effect and important parameter of number of order reduction in nonlinear experimental data is explained further.

Keywords: frequency response, order of model reduction, frequency matching condition, nonlinear experimental data

Procedia PDF Downloads 404

Authors: Shan-Jen Cheng, Te-Jen Chang, Kuang-Hsiung Tan, Shou-Ling Kuo

Abstract:

Polymer Electrolyte Membrane Fuel Cell (PEMFC) is such a time-vary nonlinear dynamic system. The traditional linear modeling approach is hard to estimate structure correctly of PEMFC system. From this reason, this paper presents a nonlinear modeling of the PEMFC using Neural Network Auto-regressive model with eXogenous inputs (NNARX) approach. The multilayer perception (MLP) network is applied to evaluate the structure of the NNARX model of PEMFC. The validity and accuracy of NNARX model are tested by one step ahead relating output voltage to input current from measured experimental of PEMFC. The results show that the obtained nonlinear NNARX model can efficiently approximate the dynamic mode of the PEMFC and model output and system measured output consistently.

Keywords: PEMFC, neural network, nonlinear modeling, NNARX

Procedia PDF Downloads 382

Authors: Edlira Donefski, Lorenc Ekonomi, Tina Donefski

Abstract:

Edgeworth approximation is one of the most important statistical methods that has a considered contribution in the reduction of the sum of standard deviation of the independent variables’ coefficients in a Quantile Regression Model. This model estimates the conditional median or other quantiles. In this paper, we have applied approximating statistical methods in an economical problem. We have created and generated a quantile regression model to see how the profit gained is connected with the realized sales of the cosmetic products in a real data, taken from a local business. The Linear Regression of the generated profit and the realized sales was not free of autocorrelation and heteroscedasticity, so this is the reason that we have used this model instead of Linear Regression. Our aim is to analyze in more details the relation between the variables taken into study: the profit and the finalized sales and how to minimize the standard errors of the independent variable involved in this study, the level of realized sales. The statistical methods that we have applied in our work are Edgeworth Approximation for Independent and Identical distributed (IID) cases, Bootstrap version of the Model and the Edgeworth approximation for Bootstrap Quantile Regression Model. The graphics and the results that we have presented here identify the best approximating model of our study.

Keywords: bootstrap, edgeworth approximation, IID, quantile

Procedia PDF Downloads 159

Authors: Amir Mohammad Fathollahi Fard, Mostafa Hajiaghaei-Keshteli, Mohammad Mahdi Paydar

Abstract:

With increasing life expectancy in developed countries, the role of home care services is highlighted by both academia and industrial contributors in Home Health Care Supply Chain (HHCSC) companies. The main decisions in such supply chain systems are the location of pharmacies, the allocation of patients to these pharmacies and also the routing and scheduling decisions of nurses to visit their patients. In this study, for the first time, an integrated model is proposed to consist of all preliminary and necessary decisions in these companies, namely, location-allocation-routing model. This model is a type of NP-hard one. Therefore, an Imperialist Competitive Algorithm (ICA) is utilized to solve the model, especially in large sizes. Results confirm the efficiency of the developed model for HHCSC companies as well as the performance of employed ICA.

Keywords: home health care supply chain, location-allocation-routing problem, imperialist competitive algorithm, optimization

Procedia PDF Downloads 398

Authors: Hamdi Sana, Emna Bouazizi, Sami Faiz

Abstract:

In recent years, there is a huge increase in the use of spatio-temporal applications where data and queries are continuously moving. As a result, the need to process real-time spatio-temporal data seems clear and real-time stream data management becomes a hot topic. Sliding window model and frequent itemset mining over dynamic data are the most important problems in the context of data mining. Thus, sliding window model for frequent itemset mining is a widely used model for data stream mining due to its emphasis on recent data and its bounded memory requirement. These methods use the traditional transaction-based sliding window model where the window size is based on a fixed number of transactions. Actually, this model supposes that all transactions have a constant rate which is not suited for real-time applications. And the use of this model in such applications endangers their performance. Based on these observations, this paper relaxes the notion of window size and proposes the use of a timestamp-based sliding window model. In our proposed frequent itemset mining algorithm, support conditions are used to differentiate frequents and infrequent patterns. Thereafter, a tree is developed to incrementally maintain the essential information. We evaluate our contribution. The preliminary results are quite promising.

Keywords: real-time spatial big data, frequent itemset, transaction-based sliding window model, timestamp-based sliding window model, weighted frequent patterns, tree, stream query

Procedia PDF Downloads 165

Authors: Ling Qi, Matloob Khushi, Josiah Poon

Abstract:

This paper introduces a pre-trained Masked Candlestick Model (MCM) for trading time-series data. The pre-trained model is based on three core designs. First, we convert trading price data at each data point as a set of normalized elements and produce embeddings of each element. Second, we generate a masked sequence of such embedded elements as inputs for self-supervised learning. Third, we use the encoder mechanism from the transformer to train the inputs. The masked model learns the contextual relations among the sequence of embedded elements, which can aid downstream classification tasks. To evaluate the performance of the pre-trained model, we fine-tune MCM for three different downstream classification tasks to predict future price trends. The fine-tuned models achieved better accuracy rates for all three tasks than the baseline models. To better analyze the effectiveness of MCM, we test the same architecture for three currency pairs, namely EUR/GBP, AUD/USD, and EUR/JPY. The experimentation results demonstrate MCM’s effectiveness on all three currency pairs and indicate the MCM’s capability for signal extraction from trading data.

Keywords: masked language model, transformer, time series prediction, trading prediction, embedding, transfer learning, self-supervised learning

Procedia PDF Downloads 129

Authors: Jung–Min Yang

Abstract:

Solvability of the model matching problem for input/output switched asynchronous sequential machines is discussed in this paper. The control objective is to determine the existence condition and design algorithm for a corrective controller that can match the stable-state behavior of the closed-loop system to that of a reference model. Switching operations and correction procedures are incorporated using output feedback so that the controlled switched machine can show the desired input/output behavior. A matrix expression is presented to address reachability of switched asynchronous sequential machines with output equivalence with respect to a model. The presented reachability condition for the controller design is validated in a simple example.

Keywords: asynchronous sequential machines, corrective control, model matching, input/output control

Procedia PDF Downloads 344

Authors: Hycham Aboutaleb, Bruno Monsuez

Abstract:

Current systems complexity has reached a degree that requires addressing conception and design issues while taking into account all the necessary aspects. Therefore, one of the main challenges is the way complex systems are specified and designed. The exponential growing effort, cost and time investment of complex systems in modeling phase emphasize the need for a paradigm, a framework and a environment to handle the system model complexity. For that, it is necessary to understand the expectations of the human user of the model and his limits. This paper presents a generic framework for designing complex systems, highlights the requirements a system model needs to fulfill to meet human user expectations, and defines the refined functional as well as non functional requirements modeling tools needs to meet to be useful in model-based system engineering.

Keywords: system modeling, modeling language, modeling requirements, framework

Procedia PDF Downloads 532

Authors: Faramarz Khosravi, Gokhan Izbirak

Abstract:

A weighted statistical stochastic based Analytical Hierarchy Process (AHP) model for modeling the potential barriers and enablers of sustainability for measuring and assessing the sustainability level is proposed. For context-dependent potential barriers and enablers, the proposed model takes the basis of the properties of the variables describing the sustainability functions and was developed into a realistic analytical model for the sustainable behavior of an organization. This thus serves as a means for measuring the sustainability of the organization. The main focus of this paper was the application of the AHP tool in a statistically-based model for measuring sustainability. Hence a strong weighted stochastic AHP based procedure was achieved. A case study scenario of a widely reported major Canadian electric utility was adopted to demonstrate the applicability of the developed model and comparatively examined its results with those of an equal-weighted model method. Variations in the sustainability of a company, as fluctuations, were figured out during the time. In the results obtained, sustainability index for successive years changed form 73.12%, 79.02%, 74.31%, 76.65%, 80.49%, 79.81%, 79.83% to more exact values 73.32%, 77.72%, 76.76%, 79.41%, 81.93%, 79.72%, and 80,45% according to priorities of factors that have found by expert views, respectively. By obtaining relatively necessary informative measurement indicators, the model can practically and effectively evaluate the sustainability extent of any organization and also to determine fluctuations in the organization over time.

Keywords: AHP, sustainability fluctuation, environmental indicators, performance measurement

Procedia PDF Downloads 122

Authors: Nikolaos Reppas, Yilin Gui, Ben Wetenhall, Colin Davie

Abstract:

Injection of CO2 inside wellbore can induce different kind of loadings that can lead to thermal, hydraulic, and mechanical changes on the surrounding rock. A dual-porosity theoretical constitutive model will be presented for the stability analysis of the wellbore during CO2 injection. An elastoplastic damage response will be considered. A bounding yield surface will be presented considering damage effects on sandstone. The main target of the research paper is to present a theoretical constitutive model that can help industries to safely store CO2 in geological rock formations and forecast any changes on the surrounding rock of the wellbore. The fully coupled elasto-plastic damage Thermo-Hydraulic-Mechanical theoretical model will be validated from existing experimental data for sandstone after simulating some scenarios by using FEM on MATLAB software.

Keywords: carbon capture and storage, rock mechanics, THM effects on rock, constitutive model

Procedia PDF Downloads 153

Authors: Hussain Syed Asad, Richard Kwok Kit Yuen, Gongsheng Huang

Abstract:

Real-time optimization has been considered an effective approach for improving energy efficient operation of heating, ventilation, and air-conditioning (HVAC) systems. In model-based real-time optimization, model mismatches cannot be avoided. When model mismatches are significant, the performance of the real-time optimization will be impaired and hence the expected energy saving will be reduced. In this paper, the model mismatches for chiller plant on real-time optimization are considered. In the real-time optimization of the chiller plant, simplified semi-physical or grey box model of chiller is always used, which should be identified using available operation data. To overcome the model mismatches associated with the chiller model, hybrid Genetic Algorithms (HGAs) method is used for online real-time training of the chiller model. HGAs combines Genetic Algorithms (GAs) method (for global search) and traditional optimization method (i.e. faster and more efficient for local search) to avoid conventional hit and trial process of GAs. The identification of model parameters is synthesized as an optimization problem; and the objective function is the Least Square Error between the output from the model and the actual output from the chiller plant. A case study is used to illustrate the implementation of the proposed method. It has been shown that the proposed approach is able to provide reliability in decision making, enhance the robustness of the real-time optimization strategy and improve on energy performance.

Keywords: energy performance, hybrid adaptive modeling, hybrid genetic algorithms, real-time optimization, heating, ventilation, and air-conditioning

Procedia PDF Downloads 418