Search results for: residual LSTM

Authors: Danilo López, Nelson Vera, Luis Pedraza

Abstract:

This paper analyzes fundamental ideas and concepts related to neural networks, which provide the reader a theoretical explanation of Long Short-Term Memory (LSTM) networks operation classified as Deep Learning Systems, and to explicitly present the mathematical development of Backward Pass equations of the LSTM network model. This mathematical modeling associated with software development will provide the necessary tools to develop an intelligent system capable of predicting the behavior of licensed users in wireless cognitive radio networks.

Keywords: neural networks, multilayer perceptron, long short-term memory, recurrent neuronal network, mathematical analysis

Procedia PDF Downloads 383

Authors: Doan Van Chinh, Bui Thi Kien Trinh

Abstract:

This paper outlines some statistical properties of residual sea level (RSL) at six representative tidal stations located along the coast of Vietnam. It was found that the positive RSL varied on average between 9.82 and 19.96cm and the negative RSL varied on average between -16.62 and -9.02cm. The maximum positive RSL varied on average between 102.8 and 265.5cm with the maximum negative RSL varied on average between -250.4 and -66.4cm. It is seen that the biggest positive RSL ere appeared in the summer months and the biggest negative RSL ere appeared in the winter months. The cumulative frequency of RSL less than 50 cm occurred between 95 and 99% of the times while the frequency of RSL higher than 100 cm accounted for between 0.01 and 0.2%. It also was found that the cumulative frequency of duration of RSL less than 24 hours occurred between 90 and 99% while the frequency of duration longer than 72 hours was in the order of 0.1 and 1%.

Keywords: coast of Vietnam, residual sea level, residual water, surge, cumulative frequency

Procedia PDF Downloads 254

Authors: Venkat S. Somayajula

Abstract:

Ground surface temperature history prediction model plays a vital role in determining standards for international nuclear waste management. International standards for borehole based nuclear waste disposal require paleoclimate cycle predictions on scale of a million forward years for the place of waste disposal. This research focuses on developing a paleoclimate cycle prediction model using Bayesian long-short term memory (LSTM) neural architecture operated on accumulated borehole temperature history data. Bayesian models have been previously used for paleoclimate cycle prediction based on Monte-Carlo weight method, but due to limitations pertaining model coupling with certain other prediction networks, Bayesian models in past couldn’t accommodate prediction cycle’s over 1000 years. LSTM has provided frontier to couple developed models with other prediction networks with ease. Paleoclimate cycle developed using this process will be trained on existing borehole data and then will be coupled to surface temperature history prediction networks which give endpoints for backpropagation of LSTM network and optimize the cycle of prediction for larger prediction time scales. Trained LSTM will be tested on past data for validation and then propagated for forward prediction of temperatures at borehole locations. This research will be beneficial for study pertaining to nuclear waste management, anthropological cycle predictions and geophysical features

Keywords: Bayesian long-short term memory neural network, borehole temperature, ground surface temperature history, paleoclimate cycle

Procedia PDF Downloads 101

Authors: Xiang-Yao Zheng, Jen-Cheng Wang, Joe-Air Jiang

Abstract:

Improving the line ampacity while using existing power grids is an important issue that electricity dispatchers are now facing. Using the information provided by the dynamic thermal rating (DTR) of transmission lines, an overhead power grid can operate safely. However, dispatchers usually lack real-time DTR information. Thus, this study proposes a long-short-term memory (LSTM)-based method, which is one of the neural network models. The LSTM-based method predicts the DTR of lines using the weather data provided by Central Weather Bureau (CWB) of Taiwan. The possible thermal bottlenecks at different locations along the line and the margin of line ampacity can be real-time determined by the proposed LSTM-based prediction method. A case study that targets the 345 kV power grid of TaiPower in Taiwan is utilized to examine the performance of the proposed method. The simulation results show that the proposed method is useful to provide the information for the smart grid application in the future.

Keywords: electricity dispatch, line ampacity prediction, dynamic thermal rating, long-short-term memory neural network, smart grid

Procedia PDF Downloads 257

Authors: Vikrant Gupta, Amrit Goswami

Abstract:

The fixed income market forms the basis of the modern financial market. All other assets in financial markets derive their value from the bond market. Owing to its over-the-counter nature, corporate bonds have relatively less data publicly available and thus is researched upon far less compared to Equities. Bond price prediction is a complex financial time series forecasting problem and is considered very crucial in the domain of finance. The bond prices are highly volatile and full of noise which makes it very difficult for traditional statistical time-series models to capture the complexity in series patterns which leads to inefficient forecasts. To overcome the inefficiencies of statistical models, various machine learning techniques were initially used in the literature for more accurate forecasting of time-series. However, simple machine learning methods such as linear regression, support vectors, random forests fail to provide efficient results when tested on highly complex sequences such as stock prices and bond prices. hence to capture these intricate sequence patterns, various deep learning-based methodologies have been discussed in the literature. In this study, a recurrent neural network-based deep learning model using long short term networks for prediction of corporate bond prices has been discussed. Long Short Term networks (LSTM) have been widely used in the literature for various sequence learning tasks in various domains such as machine translation, speech recognition, etc. In recent years, various studies have discussed the effectiveness of LSTMs in forecasting complex time-series sequences and have shown promising results when compared to other methodologies. LSTMs are a special kind of recurrent neural networks which are capable of learning long term dependencies due to its memory function which traditional neural networks fail to capture. In this study, a simple LSTM, Stacked LSTM and a Masked LSTM based model has been discussed with respect to varying input sequences (three days, seven days and 14 days). In order to facilitate faster learning and to gradually decompose the complexity of bond price sequence, an Empirical Mode Decomposition (EMD) has been used, which has resulted in accuracy improvement of the standalone LSTM model. With a variety of Technical Indicators and EMD decomposed time series, Masked LSTM outperformed the other two counterparts in terms of prediction accuracy. To benchmark the proposed model, the results have been compared with traditional time series models (ARIMA), shallow neural networks and above discussed three different LSTM models. In summary, our results show that the use of LSTM models provide more accurate results and should be explored more within the asset management industry.

Keywords: bond prices, long short-term memory, time series forecasting, empirical mode decomposition

Procedia PDF Downloads 105

Authors: Sharifullah Ahmed

Abstract:

Residual settlement of cohesionless or non-plastic soil of different strength underlying highway embankment overlain by stronger soil layer highway embankment is studied. A parametric study is carried out for different height of embankment and for different ESAL factor. The sum of elastic settlements of cohesionless subsoil due to axle induced stress and due to self-weight of pavement layers is termed as the residual settlement. The values of residual settlement (Sr) for different heights of road embankment (He) are obtained and presented as design charts for different SPT Value (N60) and ESAL factor. For rigid pavement and flexible pavement in approach to bridge or culvert, the tolerable residual settlement is 0.100m. This limit is taken as 0.200m for flexible pavement in general sections of highway without approach to bridge or culvert. A simplified guideline is developed for design of highway embankment underlain by very loose to loose cohesionless subsoil overlain by a stronger soil layer for limiting value of the residual settlement. In the current research study range of ESAL factor is 1-10 and range of SPT value (N60) is 1-10. That is found that, ground improvement is not required if the overlying stronger layer is minimum 1.5m and 4.0m for general road section of flexible pavement except bridge or culvert approach and for rigid pavement or flexible pavement in bridge or culvert approach. Tables and charts are included in the prepared guideline to obtain minimum allowable height of highway embankment to limit the residual settlement with in mentioned tolerable limit. Allowable values of the embankment height (He) are obtained corresponding to tolerable or limiting level of the residual settlement of loose subsoil for different SPT value, thickness of stronger layer (d) and ESAL factor. The developed guideline is may be issued to be used in assessment of the necessity of ground improvement in case of cohesionless subsoil underlying highway embankment overlain by stronger subsoil layer for limiting residual settlement. The ground improvement is only to be required if the residual settlement of subsoil is more than tolerable limit.

Keywords: axle pressure, equivalent single axle load, ground improvement, highway embankment, tolerable residual settlement

Procedia PDF Downloads 85

Authors: Kwangin Kim, Taewon Yoo, Haksoo Han

Abstract:

Polyimide is a superior polymer in the electronics industry, and we conducted a study to synthesize poly(amide-imide) at low temperatures. Poly(amide-imide) was synthesized at low-temperature curing to offer a thermal stable membrane with low residual stress and good processability. As a result, the low crack polymer with good processability could be used to various applications such as semiconductors, integrated circuits, coating materials, membranes, and display. The synthesis of poly(amide-imide) at low temperatures was confirmed by Fourier transform infrared spectroscopy (FT-IR). Thermal stabilities of the polymer was confirmed by thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC).

Keywords: poly(amide-imide), residual stress, thermal stability

Procedia PDF Downloads 396

Authors: Akinbo Razak Yinka, Adesanya Kehinde Kazeem, Oladokun Oluwagbenga Peter

Abstract:

Time series data are sequences of observations collected over a period of time. Such data can be used to predict health outcomes, such as disease progression, mortality, hospitalization, etc. The Statistical approach is based on mathematical models that capture the patterns and trends of the data, such as autocorrelation, seasonality, and noise, while Neural methods are based on artificial neural networks, which are computational models that mimic the structure and function of biological neurons. This paper compared both parametric and non-parametric time series models of patients treated for malaria in Maternal and Child Health Centres in Lagos State, Nigeria. The forecast methods considered linear regression, Integrated Moving Average, ARIMA and SARIMA Modeling for the parametric approach, while Multilayer Perceptron (MLP) and Long Short-Term Memory (LSTM) Network were used for the non-parametric model. The performance of each method is evaluated using the Mean Absolute Error (MAE), R-squared (R2) and Root Mean Square Error (RMSE) as criteria to determine the accuracy of each model. The study revealed that the best performance in terms of error was found in MLP, followed by the LSTM and ARIMA models. In addition, the Bootstrap Aggregating technique was used to make robust forecasts when there are uncertainties in the data.

Keywords: ARIMA, bootstrap aggregation, MLP, LSTM, SARIMA, time-series analysis

Procedia PDF Downloads 34

Authors: Danladi Ali, Onah Festus Iloabuchi

Abstract:

In this paper, GSM signal strength was measured in order to detect the type of the signal fading phenomenon using one-dimensional multilevel wavelet residual method and neural network clustering to determine the average GSM signal strength received in the study area. The wavelet residual method predicted that the GSM signal experienced slow fading and attenuated with MSE of 3.875dB. The neural network clustering revealed that mostly -75dB, -85dB and -95dB were received. This means that the signal strength received in the study is a weak signal.

Keywords: one-dimensional multilevel wavelets, path loss, GSM signal strength, propagation, urban environment

Procedia PDF Downloads 314

Authors: Ephrem Getahun, Shengwen Qi, Songfeng Guo, Yu Zou, Melesse Alemayehu

Abstract:

Residual shear characteristics of slide soil profile sections (SSPS) were examined using ring shear tests to know the relative residual shear behaviors among the sections of slide soil. The multistage-multiphase shearing techniques were employed to perform the experiment for each soil specimen continuously towards large displacements. The grain distribution analysis of SSPS samples was characterized by coarsening upward from bottom slip to the top sections; however, the slip surface was considered as a sheared zone that endorses their low shear resistance for failure. There is an average range of 1-2.5 mm axial displacement on each stage of loadings and phases of shearing that depicts the significant effect of dilation and compression of soil specimen. The middle section has the largest consolidation percentage (10-29%), and vertical displacement compared to other sections and showed high shear strengthening behavior having maximum shear stress of 189kPa at 240kPa loading compared to basal and top sections. It is found that the middle section of SSPS has relatively high shear resistance behavior for large displacement shearing. The residual shear assessment indicates that there is a significant influence of large displacement and rate on the friction coefficient behaviors; it resulted in shear weakening effect to attain their residual condition.

Keywords: comparison, displacements, residual shear stress, shear behavior, slide soils

Procedia PDF Downloads 121

Authors: Yang Liu, Mathias Agmell, Aylin Ahadi, Jan-Eric Stahl, Jinming Zhou

Abstract:

Tool wear and tool geometry have significant effects on the residual stresses in the component produced by high-speed machining. In this paper, Coupled Eulerian and Lagrangian (CEL) model is adopted to investigate the residual stress in high-speed machining of Inconel718 with a CBN170 cutting tool. The result shows that the mesh with the smallest size of 5 um yields cutting forces and chip morphology in close agreement with the experimental data. The analysis of thermal loading and mechanical loading are performed to study the effect of segmented chip morphology on the machined surface topography and residual stress distribution. The effects of cutting edge radius and flank wear on residual stresses formation and distribution on the workpiece were also investigated. It is found that the temperature within 100um depth of the machined surface increases drastically due to the more friction heat generation with the contact area of tool and workpiece increasing when a larger edge radius and flank wear are used. With the depth further increasing, the temperature drops rapidly for all cases due to the low conductivity of Inconel718. Consequently, higher and deeper tensile residual stress is generated on the superficial. Furthermore, an increased depth of plastic deformation and compressive residual stress is noticed in the subsurface, which is attributed to the reduction of the yield strength under the thermal effect. Besides, the ploughing effect produced by a larger tool edge radius contributes more than flank wear. The magnitude variation of the compressive residual stress caused by various edge radius and flank wear have a totally opposite trend, which depends on the magnitude of the ploughing and friction pressure acting on the machined surface.

Keywords: Coupled Eulerian Lagrangian, segmented chip, residual stress, tool wear, edge radius, Inconel718

Procedia PDF Downloads 116

Authors: A. T. Kuda, J. J. Dayya, A. Jimoh

Abstract:

This paper investigates the fault detection and Isolation technique of measurement data sets from a three tank system using analytical model-based temporal redundancy which is based on residual generation using parity equations/space approach. It further briefly outlines other approaches of model-based residual generation. The basic idea of parity space residual generation in temporal redundancy is dynamic relationship between sensor outputs and actuator inputs (input-output model). These residuals where then used to detect whether or not the system is faulty and indicate the location of the fault when it is faulty. The method obtains good results by detecting and isolating faults from the considered data sets measurements generated from the system.

Keywords: fault detection, fault isolation, disturbing influences, system failure, parity equation/relation, structured parity equations

Procedia PDF Downloads 270

Authors: P. Zamani, S. Mohajerzadeh, R. Masoudinejad, K. Farhangdoost

Abstract:

The riveting process is one of the important ways to keep fastening the lap joints in aircraft structures. Failure of aircraft lap joints directly depends on the stress field in the joint. An important application of riveting process is in the construction of aircraft fuselage structures. In this paper, a 3D finite element method is carried out in order to optimize residual stress field in a riveted lap joint and also to estimate its fatigue life. In continue, a number of experiments are designed and analyzed using design of experiments (DOE). Then, Taguchi method is used to select an optimized case between different levels of each factor. Besides that, the factor which affects the most on residual stress field is investigated. Such optimized case provides the maximum residual stress field. Fatigue life of the optimized joint is estimated by Paris-Erdogan law. Stress intensity factors (SIFs) are calculated using both finite element analysis and experimental formula. In addition, the effect of residual stress field, geometry, and secondary bending are considered in SIF calculation. A good agreement is found between results of such methods. Comparison between optimized fatigue life and fatigue life of other joints has shown an improvement in the joint’s life.

Keywords: fatigue life, residual stress, riveting process, stress intensity factor, Taguchi method

Procedia PDF Downloads 414

Authors: Zuhier Altawallbeh

Abstract:

This paper investigates the approximate analytical solutions of general form of Volterra integro-differential equations system by using the residual power series method (for short RPSM). The proposed method produces the solutions in terms of convergent series requires no linearization or small perturbation and reproduces the exact solution when the solution is polynomial. Some examples are given to demonstrate the simplicity and efficiency of the proposed method. Comparisons with the Laplace decomposition algorithm verify that the new method is very effective and convenient for solving system of pantograph equations.

Keywords: integro-differential equation, pantograph equations, system of initial value problems, residual power series method

Procedia PDF Downloads 396

Authors: Debneil Saha Roy

Abstract:

Weather forecasting is a complex topic and is well suited for analysis by deep learning approaches. With the wide availability of weather observation data nowadays, these approaches can be utilized to identify immediate comparisons between historical weather forecasts and current observations. This work explores the application of deep learning techniques to weather forecasting in order to accurately predict the weather over a given forecast hori­zon. Three deep neural networks are used in this study, namely, Multi-Layer Perceptron (MLP), Long Short Tunn Memory Network (LSTM) and a combination of Convolutional Neural Network (CNN) and LSTM. The predictive performance of these models is compared using two evaluation metrics. The results show that forecasting accuracy increases with an increase in the complexity of deep neural networks.

Keywords: convolutional neural network, deep learning, long short term memory, multi-layer perceptron

Procedia PDF Downloads 137

Authors: M. Kiran Reddy, K. Sreenivasa Rao

Abstract:

The conventional Hidden Markov Model (HMM)-based speech synthesis system (HTS) uses only a pulse excitation model, which significantly differs from natural excitation signal. Hence, buzziness can be perceived in the speech generated using HTS. This paper proposes an efficient excitation modeling method that can significantly reduce the buzziness, and improve the quality of HMM-based speech synthesis. The proposed approach models the pitch-synchronous residual frames extracted from the residual excitation signal. Each pitch synchronous residual frame is parameterized using 30 wavelet coefficients. These 30 wavelet coefficients are found to accurately capture the perceptually important information present in the residual waveform. In synthesis phase, the residual frames are reconstructed from the generated wavelet coefficients and are pitch-synchronously overlap-added to generate the excitation signal. The proposed excitation modeling method is integrated into HMM-based speech synthesis system. Evaluation results indicate that the speech synthesized by the proposed excitation model is significantly better than the speech generated using state-of-the-art excitation modeling methods.

Keywords: excitation modeling, hidden Markov models, pitch-synchronous frames, speech synthesis, wavelet coefficients

Procedia PDF Downloads 218

Authors: Vishwanath Pethri Kamath, Jayantha Gowda Sarapanahalli, Vishal Mishra, Siddhesh Balwant Bandgar

Abstract:

Recognition of emotional information is essential in any form of communication. Growing HCI (Human-Computer Interaction) in recent times indicates the importance of understanding of emotions expressed and becomes crucial for improving the system or the interaction itself. In this research work, textual data for emotion recognition is used. The text being the least expressive amongst the multimodal resources poses various challenges such as contextual information and also sequential nature of the language construction. In this research work, the proposal is made for a neural architecture to resolve not less than 8 emotions from textual data sources derived from multiple datasets using google pre-trained word2vec word embeddings and a Multi-head attention-based bidirectional LSTM model with a one-vs-all Multi-Level Classification. The emotions targeted in this research are Anger, Disgust, Fear, Guilt, Joy, Sadness, Shame, and Surprise. Textual data from multiple datasets were used for this research work such as ISEAR, Go Emotions, Affect datasets for creating the emotions’ dataset. Data samples overlap or conflicts were considered with careful preprocessing. Our results show a significant improvement with the modeling architecture and as good as 10 points improvement in recognizing some emotions.

Keywords: text emotion recognition, bidirectional LSTM, multi-head attention, multi-level classification, google word2vec word embeddings

Procedia PDF Downloads 146

Authors: A. Poukens, I. Sluyts, A. Krings, J. Swartenbroekx, D. Geeroms, J. Poukens

Abstract:

Introduction and Objectives: It has been known for many years that with the administration of intravenous medication, a rather significant part of the planned to be administered infusion solution, the residual volume ( the volume that remains in the IV line and or infusion bag), does not reach the patient and is wasted. This could possibly result in under dosage and diminished therapeutic effect. Despite the important impact on the patient, the reduction of residual volume lacks attention. An optimized and clearly stated protocol concerning the reduction of residual volume in an IV line is necessary for each hospital. As described in my Master’s thesis, acquiring the degree of Master in Hospital Pharmacy, administration of intravenous medication can be optimized by reduction of the residual volume. Herewith effectiveness, user-friendliness, cost efficiency and safety were taken into account. Material and Methods: By usage of a literature study and an online questionnaire sent out to all Flemish hospitals and hospitals in the Netherlands (province Limburg), current flush methods could be mapped out. In laboratory research, possible flush methods aiming to reduce the residual volume were measured. Furthermore, a self-developed experimental method to reduce the residual volume was added to the study. The current flush methods and the self-developed experimental method were compared to each other based on cost efficiency, user-friendliness and safety. Results: There is a major difference between the Flemish and the hospitals in the Netherlands (Province Limburg) concerning the approach and method of flushing IV lines after administration of intravenous medication. The residual volumes were measured and laboratory research showed that if flushing was done minimally 1-time equivalent to the residual volume, 95 percent of glucose would be flushed through. Based on the comparison, it became clear that flushing by use of a pre-filled syringe would be the most cost-efficient, user-friendly and safest method. According to laboratory research, the self-developed experimental method is feasible and has the advantage that the remaining fraction of the medication can be administered to the patient in unchanged concentration without dilution. Furthermore, this technique can be applied regardless of the level of the residual volume. Conclusion and Recommendations: It is recommendable to revise the current infusion systems and flushing methods in most hospitals. Aside from education of the hospital staff and alignment on a uniform substantiated protocol, an optimized and clear policy on the reduction of residual volume is necessary for each hospital. It is recommended to flush all IV lines with rinsing fluid with at least the equivalent volume of the residual volume. Further laboratory and clinical research for the self-developed experimental method are needed before this method can be implemented clinically in a broader setting.

Keywords: intravenous medication, infusion therapy, IV flushing, residual volume

Procedia PDF Downloads 102

Authors: Sam Khozama, Ali M. Mayya

Abstract:

Breast cancer is one of the most common types in women. Early prediction of breast cancer helps physicians detect cancer in its early stages. Big cancer data needs a very powerful tool to analyze and extract predictions. Machine learning and deep learning are two of the most efficient tools for predicting cancer based on textual data. In this study, we developed a fusion model of two machine learning and deep learning models. To obtain the final prediction, Long-Short Term Memory (LSTM) and ensemble learning with hyper parameters optimization are used, and score-level fusion is used. Experiments are done on the Breast Cancer Surveillance Consortium (BCSC) dataset after balancing and grouping the class categories. Five different training scenarios are used, and the tests show that the designed fusion model improved the performance by 3.3% compared to the individual models.

Keywords: machine learning, deep learning, cancer prediction, breast cancer, LSTM, fusion

Procedia PDF Downloads 131

Authors: Haidy S. Ghali, Engy Serag, A. Samer Ezeldin

Abstract:

Copper is an essential raw material used in the construction industry. During the year 2021 and the first half of 2022, the global market suffered from a significant fluctuation in copper raw material prices due to the aftermath of both the COVID-19 pandemic and the Russia-Ukraine war, which exposed its consumers to an unexpected financial risk. Thereto, this paper aims to develop two ANN-LSTM price prediction models, using Python, that can forecast the average monthly copper prices traded in the London Metal Exchange; the first model is a multivariate model that forecasts the copper price of the next 1-month and the second is a univariate model that predicts the copper prices of the upcoming three months. Historical data of average monthly London Metal Exchange copper prices are collected from January 2009 till July 2022, and potential external factors are identified and employed in the multivariate model. These factors lie under three main categories: energy prices and economic indicators of the three major exporting countries of copper, depending on the data availability. Before developing the LSTM models, the collected external parameters are analyzed with respect to the copper prices using correlation and multicollinearity tests in R software; then, the parameters are further screened to select the parameters that influence the copper prices. Then, the two LSTM models are developed, and the dataset is divided into training, validation, and testing sets. The results show that the performance of the 3-Month prediction model is better than the 1-Month prediction model, but still, both models can act as predicting tools for diverse economic situations.

Keywords: copper prices, prediction model, neural network, time series forecasting

Procedia PDF Downloads 77

Authors: Taposh Roy, Anna Paradowska, Ralph Abrahams, Quan Lai, Michael Law, Peter Mutton, Mehdi Soodi, Wenyi Yan

Abstract:

In this investigation, a laser cladding process with a powder feeding was used to deposit stainless steel 410L (high strength, excellent resistance to abrasion and corrosion, and great laser compatibility) onto railhead (higher strength, heat treated hypereutectoid rail grade manufactured in accordance with the requirements of European standard EN 13674 Part 1 for R400HT grade), to investigate the development and controllability of process-induced residual stress in the cladding, heat-affected zone (HAZ) and substrate and to analyse their correlation with hardness profile during two different laser cladding directions (across and along the track). Residual stresses were analysed by neutron diffraction at OPAL reactor, ANSTO. Neutron diffraction was carried out on the samples in longitudinal (parallel to the rail), transverse (perpendicular to the rail) and normal (through thickness) directions with high spatial resolution through the thickness. Due to the thick rail and thin cladding, 4 mm thick reference samples were prepared from every specimen by Electric Discharge Machining (EDM). Metallography across the laser claded sample revealed four distinct zones: The clad zone, the dilution zone, HAZ and the substrate. Compressive residual stresses were found in the clad zone and tensile residual stress in the dilution zone and HAZ. Laser cladding in longitudinally cladding induced higher tensile stress in the HAZ, whereas transversely cladding rail showed lower tensile behavior.

Keywords: laser cladding, residual stress, neutron diffraction, HAZ

Procedia PDF Downloads 238

Authors: Christoph Lohr, Hanna Wund, Peter Elsner, Kay André Weidenmann

Abstract:

Injection molding is one of the most commonly used techniques in the industrial polymer processing. In the conventional process of injection molding, the liquid polymer is injected into the cavity of the mold, where the polymer directly starts hardening at the cooled walls. To compensate the shrinkage, which is caused predominantly by the immediate cooling, holding pressure is applied. Through that whole process, residual stresses are produced by the temperature difference of the polymer melt and the injection mold and the relocation of the polymer chains, which were oriented by the high process pressures and injection speeds. These residual stresses often weaken or change the structural behavior of the parts or lead to deformation of components. One solution to reduce the residual stresses is the use of variothermal processing. Hereby the mold is heated – i.e. near/over the glass transition temperature of the polymer – the polymer is injected and before opening the mold and ejecting the part the mold is cooled. For the next cycle, the mold gets heated again and the procedure repeats. The rapid heating and cooling of the mold are realized indirectly by convection of heated and cooled liquid (here: water) which is pumped through fluid channels underneath the mold surface. In this paper, the influences of variothermal processing on the residual stresses are analyzed with samples in a larger scale (500 mm x 250 mm x 4 mm). In addition, the influence on functional elements, such as abrupt changes in wall thickness, bosses, and ribs, on the residual stress is examined. Therefore the polycarbonate samples are produced by variothermal and isothermal processing. The melt is injected into a heated mold, which has in our case a temperature varying between 70 °C and 160 °C. After the filling of the cavity, the closed mold is cooled down varying from 70 °C to 100 °C. The pressure and temperature inside the mold are monitored and evaluated with cavity sensors. The residual stresses of the produced samples are illustrated by birefringence where the effect on the refractive index on the polymer under stress is used. The colorful spectrum can be uncovered by placing the sample between a polarized light source and a second polarization filter. To show the achievement and processing effects on the reduction of residual stress the birefringence images of the isothermal and variothermal produced samples are compared and evaluated. In this comparison to the variothermal produced samples have a lower amount of maxima of each color spectrum than the isothermal produced samples, which concludes that the residual stress of the variothermal produced samples is lower.

Keywords: birefringence, injection molding, polycarbonate, residual stress, variothermal processing

Procedia PDF Downloads 255

Authors: Ahmed M. Ashteyat

Abstract:

Artificial Neural Network (ANN) models have been widely used in material modeling, inter-correlations, as well as behavior and trend predictions when the nonlinear relationship between system parameters cannot be quantified explicitly and mathematically. In this paper, ANN was used to predict the residual modulus of elasticity (RME) of self compacted concrete (SCC) damaged by heat. The ANN model was built, trained, tested and validated using a total of 112 experimental data sets, gathered from available literature. The data used in model development included temperature, relative humidity conditions, mix proportions, filler types, and fiber type. The result of ANN training, testing, and validation indicated that the RME of SCC, exposed to different temperature and relative humidity levels, could be predicted accurately with ANN techniques. The reliability between the predicated outputs and the actual experimental data was 99%. This show that ANN has strong potential as a feasible tool for predicting residual elastic modulus of SCC damaged by heat within the range of input parameter. The ANN model could be used to estimate the RME of SCC, as a rapid inexpensive substitute for the much more complicated and time consuming direct measurement of the RME of SCC.

Keywords: residual modulus of elasticity, artificial neural networks, self compacted-concrete, material modeling

Procedia PDF Downloads 504

Authors: Tahani Aljohani, Jialin Yu, Alexandra. I. Cristea

Abstract:

The more an educational system knows about a learner, the more personalised interaction it can provide, which leads to better learning. However, asking a learner directly is potentially disruptive, and often ignored by learners. Especially in the booming realm of MOOC Massive Online Learning platforms, only a very low percentage of users disclose demographic information about themselves. Thus, in this paper, we aim to predict learners’ demographic characteristics, by proposing an approach using linguistically motivated Deep Learning Architectures for Learner Profiling, particularly targeting gender prediction on a FutureLearn MOOC platform. Additionally, we tackle here the difficult problem of predicting the gender of learners based on their comments only – which are often available across MOOCs. The most common current approaches to text classification use the Long Short-Term Memory (LSTM) model, considering sentences as sequences. However, human language also has structures. In this research, rather than considering sentences as plain sequences, we hypothesise that higher semantic - and syntactic level sentence processing based on linguistics will render a richer representation. We thus evaluate, the traditional LSTM versus other bleeding edge models, which take into account syntactic structure, such as tree-structured LSTM, Stack-augmented Parser-Interpreter Neural Network (SPINN) and the Structure-Aware Tag Augmented model (SATA). Additionally, we explore using different word-level encoding functions. We have implemented these methods on Our MOOC dataset, which is the most performant one comparing with a public dataset on sentiment analysis that is further used as a cross-examining for the models' results.

Keywords: deep learning, data mining, gender predication, MOOCs

Procedia PDF Downloads 114

Authors: Xinxin Ma, Guangze Tang, Shuxin Yang, Jinguang He, Fan Zhang, Peiling Sun, Ming Liu, Minyu Sun, Liqin Wang

Abstract:

As a traditional surface hardening technique, shot peening is widely used in industry. By using shot peening, a residual compressive stress is formed in the surface which is beneficial for improving the fatigue life of metal materials. At the same time, very fine grains and high density defects are generated in the surface layer which enhances the surface hardness, either. However, most of the processes are carried out at room temperature. For high strength steel, such as M50, the thickness of the strengthen layer is limited. In order to obtain a thick strengthen surface layer, elevated temperature shot peening was carried out in this work by using Φ1mm cast ion balls with a speed of 80m/s. Considering the tempering temperature of M50 steel is about 550 oC, the processing temperature was in the range from 300 to 500 oC. The effect of processing temperature and processing time of shot peening on distribution of residual stress and surface hardness was investigated. As we known, the working temperature of M50 steel can be as high as 315 oC. Because the defects formed by shot peening are unstable when the working temperature goes higher, it is worthy to understand what happens during the shot peening process, and what happens when the strengthen samples were kept at a certain temperature. In our work, the shot peening time was selected from 2 to 10 min. And after the strengthening process, the samples were annealed at various temperatures from 200 to 500 oC up to 60 h. The results show that the maximum residual compressive stress is near 900 MPa. Compared with room temperature shot peening, the strengthening depth of 500 oC shot peening sample is about 2 times deep. The surface hardness increased with the processing temperature, and the saturation peening time decreases. After annealing, the residual compressive stress decreases, however, for 500 oC peening sample, even annealing at 500 oC for 20 h, the residual compressive stress is still over 600 MPa. However, it is clean to see from SEM that the grain size of surface layers is still very small.

Keywords: shot peening, M50 steel, residual compressive stress, elevated temperature

Procedia PDF Downloads 420

Authors: Gong Zhilin, Jing Yang, Jian Yin

Abstract:

The purpose of this paper is to develop a novel system for credit card fraud detection based on sequential modeling of data using hybrid deep learning models. The projected model encapsulates five major phases are pre-processing, imbalance-data handling, feature extraction, optimal feature selection, and fraud detection with an ensemble classifier. The collected raw data (input) is pre-processed to enhance the quality of the data through alleviation of the missing data, noisy data as well as null values. The pre-processed data are class imbalanced in nature, and therefore they are handled effectively with the K-means clustering-based SMOTE model. From the balanced class data, the most relevant features like improved Principal Component Analysis (PCA), statistical features (mean, median, standard deviation) and higher-order statistical features (skewness and kurtosis). Among the extracted features, the most optimal features are selected with the Self-improved Arithmetic Optimization Algorithm (SI-AOA). This SI-AOA model is the conceptual improvement of the standard Arithmetic Optimization Algorithm. The deep learning models like Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and optimized Quantum Deep Neural Network (QDNN). The LSTM and CNN are trained with the extracted optimal features. The outcomes from LSTM and CNN will enter as input to optimized QDNN that provides the final detection outcome. Since the QDNN is the ultimate detector, its weight function is fine-tuned with the Self-improved Arithmetic Optimization Algorithm (SI-AOA).

Keywords: credit card, data mining, fraud detection, money transactions

Procedia PDF Downloads 102

Authors: P. D. Pedrosa, J. M. A. Rebello, M. P. Cindra Fonseca

Abstract:

Duplex stainless steels (DSS) exhibit a biphasic microstructure consisting of austenite and delta ferrite. Their high resistance to oxidation, and corrosion, even in H2S containing environments, allied to low cost when compared to conventional stainless steel, are some properties which make this material very attractive for several industrial applications. However, several of these industrial applications imposes cyclic loading to the equipments and in consequence fatigue damage needs to be a concern. A well-known way of improving the fatigue life of a component is by introducing compressive residual stress in its surface. Shot peening is an industrial working process which brings the material directly beneath component surface in a high mechanical compressive state, so inhibiting fatigue crack initiation. However, one must take into account the fact that the cyclic loading itself can reduce and even suppress these residual stresses, thus having undesirable consequences in the process of improving fatigue life by the introduction of compressive residual stresses. In the present work, shot peening was used to introduce residual stresses in several DSS samples. These were thereafter submitted to three different fatigue regimes: low, medium and high cycle fatigue. The evolution of the residual stress during loading were then examined on both surface and subsurface of the samples. It was used the DSS UNS S31803, with microstructure composed of 49% austenite and 51% ferrite. The treatment of shot peening was accomplished by the application of blasting in two Almen intensities of 0.25 and 0.39A. The residual stresses were measured by X-ray diffraction using the double exposure method and a portable equipment with CrK radiation and the (211) diffracting plane for the austenite phase and the (220) plane for the ferrite phase. It is known that residual stresses may arise when two regions of the same material experienced different degrees of plastic deformation. When these regions are separated in respect to each other on a scale that is large compared to the material's microstructure they are called macro stresses. In contrast, microstresses can largely vary over distances which are small comparable to the scale of the material's microstructure and must balance zero between the phases present. In the present work, special attention will be paid to the measurement of residual microstresses. Residual stress measurements were carried out in test pieces submitted to low, medium and high-cycle fatigue, in both longitudinal and transverse direction of the test pieces. It was found that after shot peening, the residual microstress is tensile in the austenite and compressive in the ferrite phases. It was hypothesized that the hardening behavior of the austenite after shot peening was probably due to its higher nitrogen content. Fatigue cycling can effectively change this stress state but this effect was found to be dependent of the shot peening intensity was well as the fatigue range.

Keywords: residual stresses, fatigue, duplex steel, shot peening

Procedia PDF Downloads 193

Authors: Y. Kourd, D. Lefebvre, N. Guersi

Abstract:

This paper presents a new idea for fault detection and isolation (FDI) technique which is applied to industrial system. This technique is based on Neural Networks fault-free and Faulty behaviors Models (NNFM's). NNFM's are used for residual generation, while decision tree architecture is used for residual evaluation. The decision tree is realized with data collected from the NNFM’s outputs and is used to isolate detectable faults depending on computed threshold. Each part of the tree corresponds to specific residual. With the decision tree, it becomes possible to take the appropriate decision regarding the actual process behavior by evaluating few numbers of residuals. In comparison to usual systematic evaluation of all residuals, the proposed technique requires less computational effort and can be used for on line diagnosis. An application example is presented to illustrate and confirm the effectiveness and the accuracy of the proposed approach.

Keywords: neural networks, decision trees, diagnosis, behaviors

Procedia PDF Downloads 457

Authors: Getaneh Berie Tarekegn

Abstract:

Wireless communication via unmanned aerial vehicles (UAVs) has drawn a great deal of attention due to its flexibility in establishing line-of-sight (LoS) communications. However, in complex urban and dynamic environments, the movement of UAVs can be blocked by trees and high-rise buildings that obstruct directional paths. With reconfigurable intelligent surfaces (RIS), this problem can be effectively addressed. To achieve this goal, accurate channel estimation in RIS-assisted UAV-enabled wireless communications is crucial. This paper proposes an accurate channel estimation model using long short-term memory (LSTM) for a multi-user RIS-assisted UAV-enabled wireless communication system. According to simulation results, LSTM can improve the channel estimation performance of RIS-assisted UAV-enabled wireless communication.

Keywords: channel estimation, reconfigurable intelligent surfaces, long short-term memory, unmanned aerial vehicles

Procedia PDF Downloads 19

Authors: Yuriy A. Shevchenko, Konstantin V. Nefedev

Abstract:

We consider the antiferromagnetic systems of Ising spins located at the sites of the hexagonal, triangular and pyrochlore lattices. Such systems can be diluted to a certain concentration level by randomly replacing the magnetic spins with nonmagnetic ones. Quite recently we studied density of states (DOS) was calculated by the Wang-Landau method. Based on the obtained data, we calculated the dependence of the residual entropy (entropy at a temperature tending to zero) on the dilution concentration for quite large systems (more than 2000 spins). In the current study, we obtained the same data for small systems (less than 20 spins) by a complete search of all possible magnetic configurations and compared the result with the result for large systems. The shape of the curve remains unchanged in both cases, but the specific values of the residual entropy are different because of the finite size effect.

Keywords: entropy, pyrochlore, spin ice, Wang-Landau algorithm

Procedia PDF Downloads 234