Search results for: gradient training method

Authors: Fahad Alahmadi, Norhan Abd Rahman, Mohammad Abdulrazzak, Zulikifli Yusop

Abstract:

Runoff prediction of ungauged catchment is still a challenging task especially in arid regions with a unique land cover such as volcanic basalt rocks where geological weathering and fractures are highly significant. In this study, Bathan catchment in Madinah western Saudi Arabia was selected for analysis. The aim of this paper is to evaluate different rainfall loss methods; soil conservation Services curve number (SCS-CN), green-ampt and initial-constant rate. Different direct runoff methods were evaluated: soil conservation services dimensionless unit hydrograph (SCS-UH), Snyder unit hydrograph and Clark unit hydrograph. The study showed the superiority of SCS-CN loss method and Clark unit hydrograph method for ungauged catchment where there is no observed runoff data.

Keywords: urban runoff modelling, arid regions, ungauged catchments, volcanic rocks, Madinah, Saudi Arabia

Procedia PDF Downloads 387

Authors: Alija Uzunovic, Sasa Pilipovic, Aida Sapcanin, Zahida Ademovic, Berina Pilipović

Abstract:

Capsaicin is a naturally occurring alkaloid extracted from capsicum fruit extracts of different of Capsicum species. It has been employed topically to treat many diseases such as rheumatoid arthritis, osteoarthritis, cancer pain and nerve pain in diabetes. The high degree of pre-systemic metabolism of intragastrical capsaicin and the short half-life of capsaicin by intravenous administration made topical application of capsaicin advantageous. In this study, we have evaluated differences in the dissolution characteristics of capsaicin patch 11 mg (purchased from market) at different dissolution rotation speed. The proposed patch area is 308 cm2 (22 cm x 14 cm; it contains 36 µg of capsaicin per square centimeter of adhesive). USP Apparatus 5 (Paddle Over Disc) is used for transdermal patch testing. The dissolution study was conducted using USP apparatus 5 (n=6), ERWEKA DT800 dissolution tester (paddle-type) with addition of a disc. The fabricated patch of 308 cm2 is to be cut into 9 cm2 was placed against a disc (delivery side up) retained with the stainless-steel screen and exposed to 500 mL of phosphate buffer solution pH 7.4. All dissolution studies were carried out at 32 ± 0.5 °C and different rotation speed (50± 5; 100± 5 and 150± 5 rpm). 5 ml aliquots of samples were withdrawn at various time intervals (1, 4, 8 and 12 hours) and replaced with 5 ml of dissolution medium. Withdrawn were appropriately diluted and analyzed by reversed-phase liquid chromatography (RP-LC). A Reversed Phase Liquid Chromatography (RP-LC) method has been developed, optimized and validated for the separation and quantitation of capsaicin in a transdermal patch. The method uses a ProntoSIL 120-3-C18 AQ 125 x 4,0 mm (3 μm) column maintained at 600C. The mobile phase consisted of acetonitrile: water (50:50 v/v), the flow rate of 0.9 mL/min, the injection volume 10 μL and the detection wavelength 222 nm. The used RP-LC method is simple, sensitive and accurate and can be applied for fast (total chromatographic run time was 4.0 minutes) and simultaneous analysis of capsaicin and dihydrocapsaicin in a transdermal patch. According to the results obtained in this study, we can conclude that the relative difference of dissolution rate of capsaicin after 12 hours was elevated by increase of dissolution rotation speed (100 rpm vs 50 rpm: 84.9± 11.3% and 150 rpm vs 100 rpm: 39.8± 8.3%). Although several apparatus and procedures (USP apparatus 5, 6, 7 and a paddle over extraction cell method) have been used to study in vitro release characteristics of transdermal patches, USP Apparatus 5 (Paddle Over Disc) could be considered as a discriminatory test. would be able to point out the differences in the dissolution rate of capsaicin at different rotation speed.

Keywords: capsaicin, in vitro, patch, RP-LC, transdermal

Procedia PDF Downloads 212

Authors: Sheela Tiwari, R. Naresh, R. Jha

Abstract:

The paper presents an investigation into the effect of neural network predictive control of UPFC on the transient stability performance of a multi-machine power system. The proposed controller consists of a neural network model of the test system. This model is used to predict the future control inputs using the damped Gauss-Newton method which employs ‘backtracking’ as the line search method for step selection. The benchmark 2 area, 4 machine system that mimics the behavior of large power systems is taken as the test system for the study and is subjected to three phase short circuit faults at different locations over a wide range of operating conditions. The simulation results clearly establish the robustness of the proposed controller to the fault location, an increase in the critical clearing time for the circuit breakers and an improved damping of the power oscillations as compared to the conventional PI controller.

Keywords: identification, neural networks, predictive control, transient stability, UPFC

Procedia PDF Downloads 361

Authors: Swarali Hingse, Shraddha Digole, Uday Annapure

Abstract:

The present study investigates the biotransformation of ferulic acid to vanillin by Pycnoporus cinnabarinus and its optimization using one-factor-at-a-time method as well as statistical approach. Effect of various physicochemical parameters and medium components was studied using one-factor-at-a-time method. Screening of the significant factors was carried out using L25 Taguchi orthogonal array and then these selected significant factors were further optimized using response surface methodology (RSM). Significant media components obtained using Taguchi L25 orthogonal array were glucose, KH2PO4 and yeast extract. Further, a Box Behnken design was used to investigate the interactive effects of the three most significant media components. The final medium obtained after optimization using RSM containing glucose (34.89 g/L), diammonium tartrate (1 g/L), yeast extract (1.47 g/L), MgSO4•7H2O (0.5 g/L), KH2PO4 (0.15 g/L), and CaCl2•2H2O (20 mg/L) resulted in amplification of vanillin production from 30.88 mg/L to 187.63 mg/L.

Keywords: ferulic acid, pycnoporus cinnabarinus, response surface methodology, vanillin

Procedia PDF Downloads 367

Authors: Tasuku Hongo, Hiroya Mamori, Naoya Fukushima, Makoto Yamamoto

Abstract:

A tsunami induced by an earthquake gives a severe disaster in coastal area. As well known, the huge earthquake in Japan 2011 induced a huge tsunami and the tsunami caused serious damage in the Tohoku and Kanto area. Although breakwaters were constructed in the coast to suppress the tsunami, these were collapsed, and it resulted in severe disasters. In order to decrease the tsunami disaster, we propose the submerged breakwaters and investigate its effect on the tsunami behavior by means of numerical simulations. In order to reproduce tsunami and capture its interface, we employed a moving particle method which is one of the Lagragian methods. Different from ordinary breakwaters, the present breakwater is located in the under-sea. An effective installation condition is investigated by the parametric study. The results show that the submerged breakwater can decrease the wave force by the tsunami. Moreover, the combination of two submerged breakwaters can reduce the tsunami safely and effectively. Therefore, the present results give the effective condition of the installation of the under-sea breakwaters and its mechanism.

Keywords: coastal area, tsunami force reduction, MPS method, submerged breakwater

Procedia PDF Downloads 152

Authors: Archana Thangavelu, Han-Ik Park, Young-Chul Park, Joon-Hong Park

Abstract:

In this numerical study, we have proposed a method for evaluation of backward extrusion process of pressure vessel made up of steel. Demand for lighter and stiffer products have been increasing in the last years especially in automobile engineering. Through detailed finite element analysis, effective stress, strain and velocity profile have been obtained with optimal range. The process design of a forward and backward extrusion axe-symmetric part has been studied. Forging is mainly carried out because forged products are highly reliable and possess superior mechanical properties when compared to normal products. Performing computational simulations of 3D hot forging with various dimensions of billet and optimization of weight is carried out using Taguchi Orthogonal Array (OA) Optimization technique. The technique used in this study can be used for newly developed materials to investigate its forgeability for much complicated shapes in closed hot die forging process.

Keywords: backward extrusion, hot forging, optimization, finite element analysis, Taguchi method

Procedia PDF Downloads 296

Authors: Kassoul Ilyes

Abstract:

The field of error correcting codes has been revolutionized by the introduction of iteratively decoded codes. Among these, LDPC codes are now a preferred solution thanks to their remarkable performance and low complexity. The binary version of LDPC codes showed even better performance, although it’s decoding introduced greater complexity. This thesis studies the performance of binary LDPC codes using simplified weighted decisions. Information is transported between a transmitter and a receiver by digital transmission systems, either by propagating over a radio channel or also by using a transmission medium such as the transmission line. The purpose of the transmission system is then to carry the information from the transmitter to the receiver as reliably as possible. These codes have not generated enough interest within the coding theory community. This forgetfulness will last until the introduction of Turbo-codes and the iterative principle. Then it was proposed to adopt Pearl's Belief Propagation (BP) algorithm for decoding these codes. Subsequently, Luby introduced irregular LDPC codes characterized by a parity check matrix. And finally, we study simplifications on binary LDPC codes. Thus, we propose a method to make the exact calculation of the APP simpler. This method leads to simplifying the implementation of the system.

Keywords: LDPC, parity check matrix, 5G, BER, SNR

Procedia PDF Downloads 141

Authors: Zongjie Wang, Zhizhong Guo

Abstract:

While the regular optimal power flow model focuses on a single time scan, the optimization of power systems is typically intended for a time duration with respect to a desired objective function. In this paper, a temporal optimal power flow model for a time period is proposed. To reduce the computation burden needed for calculating temporal optimal power flow, a characteristic optimal power flow model is proposed, which employs different characteristic load patterns to represent the objective function and security constraints. A numerical method based on the interior point method is also proposed for solving the characteristic optimal power flow model. Both the temporal optimal power flow model and characteristic optimal power flow model can improve the systems’ desired objective function for the entire time period. Numerical studies are conducted on the IEEE 14 and 118-bus test systems to demonstrate the effectiveness of the proposed characteristic optimal power flow model.

Keywords: optimal power flow, time period, security, economy

Procedia PDF Downloads 433

Authors: Liu Ying-jie, Lu Wen-bo, Peng Cheng-jian

Abstract:

With the development of the automotive technology and electric vehicle, the contribution of the wind noise on the interior noise becomes the main source of noise. The main transfer path which the exterior excitation is transmitted through is the greenhouse panels and side windows. Simulating the wind noise transmitted into the vehicle accurately in the early development stage can be very challenging. The basic methodologies of this study were based on the Lighthill analogy; the exterior flow field around a passenger car was computed using unsteady Computational Fluid Dynamics (CFD) firstly and then a Finite Element Method (FEM) was used to compute the interior acoustic response. The major findings of this study include: 1) The Sound Pressure Level (SPL) response at driver’s ear locations is mainly induced by the turbulence pressure fluctuation; 2) Peaks were found over the full frequency range. It is found that the methodology used in this study could predict the interior wind noise induced by the exterior aerodynamic excitation in industry.

Keywords: wind noise, computational fluid dynamics, finite element method, passenger car

Procedia PDF Downloads 153

Authors: Van Diem Thi, Mourad Khelifa, Mohammed El Ganaoui, Yann Rogaume

Abstract:

An efficient numerical method has been developed to incorporate the effects of heat transfer in timber panels on partition walls exposed to real building fires. The procedure has been added to the software package Abaqus/Standard as a user-defined subroutine (UMATHT) and has been verified using both time-and spatially dependent heat fluxes in two- and three-dimensional problems. The aim is to contribute to the development of simulation tools needed to assist structural engineers and fire testing laboratories in technical assessment exercises. The presented method can also be used under the developmental stages of building components to optimize performance in real fire conditions. The accuracy of the used thermal properties and the finite element models was validated by comparing the predicted results with three different available fire tests in literature. It was found that the model calibrated to results from standard fire conditions provided reasonable predictions of temperatures within assemblies exposed to real building fire.

Keywords: Timber panels, heat transfer, thermal properties, standard fire tests

Procedia PDF Downloads 323

Authors: Eric Yi-Hsiu Huang, Meng-Chun Kao, Wen-Chuan Kuo

Abstract:

For every laparoscopic surgery, a safe pneumoperitoneumcreation (gaining access to the peritoneal cavity) is the first and essential step. However, closed pneumoperitoneum is usually obtained by blind insertion of a Veress needle into the peritoneal cavity, which may carry potential risks suchas bowel and vascular injury.Until now, there remains no definite measure to visually confirm the position of the needle tip inside the peritoneal cavity. Therefore, this study established an image-guided Veress needle method by combining a fiber probe with optical coherence tomography (OCT). An algorithm was also proposed for determining the exact location of the needle tip through the acquisition of OCT images. Our method not only generates a series of “live” two-dimensional (2D) images during the needle puncture toward the peritoneal cavity but also can eliminate operator variation in image judgment, thus improving peritoneal access safety. This study was approved by the Ethics Committee of Taipei Veterans General Hospital (Taipei VGH IACUC 2020-144). A total of 2400 in vivo OCT images, independent of each other, were acquired from experiments of forty peritoneal punctures on two piglets. Characteristic OCT image patterns could be observed during the puncturing process. The ROC curve demonstrates the discrimination capability of these quantitative image features of the classifier, showing the accuracy of the classifier for determining the inside vs. outside of the peritoneal was 98% (AUC=0.98). In summary, the present study demonstrates the ability of the combination of our proposed automatic identification method and OCT imaging for automatically and objectively identifying the location of the needle tip. OCT images translate the blind closed technique of peritoneal access into a visualized procedure, thus improving peritoneal access safety.

Keywords: pneumoperitoneum, optical coherence tomography, automatic identification, veress needle

Procedia PDF Downloads 114

Authors: Fei Gao, Rodolfo C. Raga Jr.

Abstract:

This research proposal will ascertain the major risk factors for diabetes and to design a predictive model for risk assessment. The project aims to improve diabetes early detection and management by utilizing data science techniques, which may improve patient outcomes and healthcare efficiency. The phase relation values of each attribute were used to analyze and choose the attributes that might influence the examiner's survival probability using Diabetes Health Indicators Dataset from Kaggle’s data as the research data. We compare and evaluate eight machine learning algorithms. Our investigation begins with comprehensive data preprocessing, including feature engineering and dimensionality reduction, aimed at enhancing data quality. The dataset, comprising health indicators and medical data, serves as a foundation for training and testing these algorithms. A rigorous cross-validation process is applied, and we assess their performance using five key metrics like accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (AUC-ROC). After analyzing the data characteristics, investigate their impact on the likelihood of diabetes and develop corresponding risk indicators.

Keywords: diabetes, risk factors, predictive model, risk assessment, data science techniques, early detection, data analysis, Kaggle

Procedia PDF Downloads 56

Authors: Endale Fantahun Tadesse, Sabika Khalid

Abstract:

The global call for increased male representation in early childhood education (ECE) has garnered significant attention. Emerging studies have indicated that involving men in ECE can yield positive outcomes for children's physical and psychological development. Challenging the prevailing misconception and stereotype that women dominate the ECE sector is crucial. In light of this, the present study undertakes a systematic review of nine studies on males working in ECE, revealing a dearth of male presence in the field in China as well. To address this issue, substantial structural changes must be implemented to enhance the inadequate pay and working conditions that dissuade both men and women from pursuing a sustainable career in ECE. It is recommended that school leadership raise awareness among female teachers and parents, encouraging them to support and uphold virtuous values for male teachers. Additionally, governing bodies should provide explicit guidelines during training programs to address concerns regarding potential abuse and gender biases. The findings of this review underscore the need for future studies to examine the self-identities of male teachers from various stakeholders' perspectives and explore the consequences of being in the profession through rigorous and robust methodologies that can inform policymakers.

Keywords: male teachers, Early Childhood Education (ECE), self-identity, perception of stakeholders

Procedia PDF Downloads 22

Authors: Cagri Mollamahmutoglu, Ali Mercan, Aykut Levent

Abstract:

Studies concerning the microstructures are becoming more important as the utilization of various micro-electro mechanical systems (MEMS) are increasing. Thus in recent years, thermal buckling and vibration analysis of microstructures have been subject to many investigations that are utilizing different numerical methods. In this study, thermal effects on mechanical response of a functionally graded (FG) Timoshenko micro-beam are presented in the framework of a mixed finite element formulation. Size effects are taken into consideration via modified couple stress theory. The mixed formulation is based on a function which in turn is derived via Gateaux Differential scientifically. After the resolution of all field equations of the beam, a potential operator is carefully constructed. Then this operator is used for the manufacturing of the functional. Usual procedures of finite element approximation are utilized for the derivation of the mixed finite element equations once the potential is obtained. Resulting finite element formulation allows usage of C₀ type simple linear shape functions and avoids shear-locking phenomena, which is a common shortcoming of the displacement-based formulations of moderately thick beams. The developed numerical scheme is used to obtain the effects of thermal loads on the static bending, free vibration and buckling of FG Timoshenko micro-beams for different power-law parameters, aspect ratios and boundary conditions. The versatility of the mixed formulation is presented over other numerical methods such as generalized differential quadrature method (GDQM). Another attractive property of the formulation is that it allows direct calculation of the contribution of micro effects on the overall mechanical response.

Keywords: micro-beam, functionally graded materials, thermal effect, mixed finite element method

Procedia PDF Downloads 120

Authors: Geon Bae, Cheol Hee Moon

Abstract:

Organic Light Emitting Diodes (OLEDs) are being developed rapidly as next-generation displays due to their self-luminous and flexible characteristics. OLEDs are highly susceptible to moisture and oxygen due to their structural properties. Thus, requiring a high level of encapsulation technology. Recently, encapsulation technology such as Thin Film Encapsulation (TFE) has been developed for OLED, but it is not perfect to prevent moisture permeation on the side. In this study, we propose OLED encapsulation method using Low melting Point Alloy (LMPA). The LMPA line was designed in square box shape on the outer edge of the device and was formed by screen printing method. To determine if LMPA has an effect on OLED, we fabricated solution processed OLEDs with a square-shaped LMPA line and evaluate the I-V-L characteristics of the OLEDs. Also, the resistance characteristic of the LMPA line was observed by repeatedly bending the LMPA line. It is expected that LMPA encapsulation will have a great advantage in shortening the process time and cost reduction.

Keywords: OLED, encapsulation, LMPA, solution process

Procedia PDF Downloads 233

Authors: Pablo Andres Flores Siguenza, Noe Rodrigo Guaman Guachichullca

Abstract:

Sustainable practices have received increasing attention from academics and companies in recent decades due to, among many factors, the market advantages they generate, global commitments, and policies aimed at reducing greenhouse gas emissions, addressing resource scarcity, and rethinking waste management. The search for ways to promote sustainability leads industries to abandon classical methods and resort to the use of innovative strategies, which in turn are based on quantitative analysis methods and tools such as life cycle analysis (LCA), which is the basis for sustainable production and consumption, since it is a method that analyzes objectively, methodically, systematically, and scientifically the environmental impact caused by a process/product during its entire life cycle. The objective of this study is to develop an LCA of the antibacterial gel product throughout its entire supply chain (SC) under the methodology of ISO 14044 with the help of Gabi software and the Recipe 2016 method. The selection of the case study product was made based on its relevance in the current context of the COVID-19 pandemic and its exponential increase in production. For the development of the LCA, data from a Mexican company are used, and 3 scenarios are defined to obtain the midpoint and endpoint environmental impacts both by phases and globally. As part of the results, the most outstanding environmental impact categories are climate change, fossil fuel depletion, and terrestrial ecotoxicity, and the stage that generates the most pollution in the entire SC is the extraction of raw materials. The study serves as a basis for the development of different sustainability strategies, demonstrates the usefulness of an LCA, and agrees with different authors on the role and importance of this methodology in sustainable development.

Keywords: sustainability, sustainable development, life cycle analysis, environmental impact, antibacterial gel

Procedia PDF Downloads 34

Authors: Benoit Rougier, Alexandre Lefrancois, Herve Aubert

Abstract:

Microwave interrogation in the range 10-100 GHz is identified as an advanced technique to investigate simultaneously shock and particle velocity measurements. However, it requires the understanding of electromagnetic wave propagation in a multi-layered moving media. The existing models limit their approach to wave guides or evaluate the velocities with a fitting method, restricting therefore the domain of validity and the precision of the results. Moreover, few data of permittivity on high explosives at these frequencies under dynamic compression have been reported. In this paper, shock and particle velocities are computed concurrently for steady and unsteady shocks for various inert and reactive materials, via a propagation model based on Doppler shifts and signal amplitude. Refractive index of the material under compression is also calculated. From experimental data processing, it is demonstrated that Hugoniot curve can be evaluated. The comparison with published results proves the accuracy of the proposed method. This microwave interrogation technique seems promising for shock and detonation waves studies.

Keywords: electromagnetic propagation, experimental setup, Hugoniot measurement, shock propagation

Procedia PDF Downloads 202

Authors: Skyler Kim

Abstract:

An early diagnosis of leukemia has always been a challenge to doctors and hematologists. On a worldwide basis, it was reported that there were approximately 350,000 new cases in 2012, and diagnosing leukemia was time-consuming and inefficient because of an endemic shortage of flow cytometry equipment in current clinical practice. As the number of medical diagnosis tools increased and a large volume of high-quality data was produced, there was an urgent need for more advanced data analysis methods. One of these methods was the AI approach. This approach has become a major trend in recent years, and several research groups have been working on developing these diagnostic models. However, designing and implementing a leukemia diagnostic system in real clinical environments based on a deep learning approach with larger sets remains complex. Leukemia is a major hematological malignancy that results in mortality and morbidity throughout different ages. We decided to select acute lymphocytic leukemia to develop our diagnostic system since acute lymphocytic leukemia is the most common type of leukemia, accounting for 74% of all children diagnosed with leukemia. The results from this development work can be applied to all other types of leukemia. To develop our model, the Kaggle dataset was used, which consists of 15135 total images, 8491 of these are images of abnormal cells, and 5398 images are normal. In this paper, we design and implement a leukemia diagnostic system in a real clinical environment based on deep learning approaches with larger sets. The proposed diagnostic system has the function of detecting and classifying leukemia. Different from other AI approaches, we explore hybrid architectures to improve the current performance. First, we developed two independent convolutional neural network models: VGG19 and ResNet50. Then, using both VGG19 and ResNet50, we developed a hybrid deep learning architecture employing transfer learning techniques to extract features from each input image. In our approach, fusing the features from specific abstraction layers can be deemed as auxiliary features and lead to further improvement of the classification accuracy. In this approach, features extracted from the lower levels are combined into higher dimension feature maps to help improve the discriminative capability of intermediate features and also overcome the problem of network gradient vanishing or exploding. By comparing VGG19 and ResNet50 and the proposed hybrid model, we concluded that the hybrid model had a significant advantage in accuracy. The detailed results of each model’s performance and their pros and cons will be presented in the conference.

Keywords: acute lymphoblastic leukemia, hybrid model, leukemia diagnostic system, machine learning

Procedia PDF Downloads 176

Authors: Drew Commandeur, Ryan Brodie, Sandra Hundza, Marc Klimstra

Abstract:

The amplitude of raw electromyography (EMG) is affected by recording conditions and often requires normalization to make meaningful comparisons. Bioelectric methods normalize with an EMG signal recorded during a standardized task or from the experimental protocol itself, while biomechanical methods often involve measurements with an additional sensor such as a force transducer. Common bioelectric normalization techniques for treadmill walking include maximum voluntary isometric contraction (MVIC), dynamic EMG peak (EMGPeak) or dynamic EMG mean (EMGMean). There are several concerns with using MVICs to normalize EMG, including poor reliability and potential discomfort. A limitation of bioelectric normalization techniques is that they could result in a misrepresentation of the absolute magnitude of force generated by the muscle and impact the interpretation of EMG between functionally disparate groups. Additionally, methods that normalize to EMG recorded during the task may eliminate some real inter-individual variability due to biological variation. This study compared biomechanical and bioelectric EMG normalization techniques during treadmill walking to assess the impact of the normalization method on the functional interpretation of EMG data. For the biomechanical method, we normalized EMG to a target torque (EMGTS) and the bioelectric methods used were normalization to the mean and peak of the signal during the walking task (EMGMean and EMGPeak). The effect of normalization on muscle activation pattern, EMG amplitude, and inter-individual variability were compared between disparate cohorts of OLD (76.6 yrs N=11) and YOUNG (26.6 yrs N=11) adults. Participants walked on a treadmill at a self-selected pace while EMG was recorded from the right lower limb. EMG data from the soleus (SOL), medial gastrocnemius (MG), tibialis anterior (TA), vastus lateralis (VL), and biceps femoris (BF) were phase averaged into 16 bins (phases) representing the gait cycle with bins 1-10 associated with right stance and bins 11-16 with right swing. Pearson’s correlations showed that activation patterns across the gait cycle were similar between all methods, ranging from r =0.86 to r=1.00 with p<0.05. This indicates that each method can characterize the muscle activation pattern during walking. Repeated measures ANOVA showed a main effect for age in MG for EMGPeak but no other main effects were observed. Interactions between age*phase of EMG amplitude between YOUNG and OLD with each method resulted in different statistical interpretation between methods. EMGTS normalization characterized the fewest differences (four phases across all 5 muscles) while EMGMean (11 phases) and EMGPeak (19 phases) showed considerably more differences between cohorts. The second notable finding was that coefficient of variation, the representation of inter-individual variability, was greatest for EMGTS and lowest for EMGMean while EMGPeak was slightly higher than EMGMean for all muscles. This finding supports our expectation that EMGTS normalization would retain inter-individual variability which may be desirable, however, it also suggests that even when large differences are expected, a larger sample size may be required to observe the differences. Our findings clearly indicate that interpretation of EMG is highly dependent on the normalization method used, and it is essential to consider the strengths and limitations of each method when drawing conclusions.

Keywords: electromyography, EMG normalization, functional EMG, older adults

Procedia PDF Downloads 75

Authors: Fanqiang Kong, Chending Bian

Abstract:

Sparse unmixing is a promising approach in a semisupervised fashion by assuming that the observed pixels of a hyperspectral image can be expressed in the form of linear combination of only a few pure spectral signatures (end members) in an available spectral library. However, the sparse unmixing problem still remains a great challenge at finding the optimal subset of endmembers for the observed data from a large standard spectral library, without considering the spatial information. Under such circumstances, a sparse unmixing algorithm termed as non-local simultaneous sparse unmixing (NLSSU) is presented. In NLSSU, the non-local simultaneous sparse representation method for endmember selection of sparse unmixing, is used to finding the optimal subset of endmembers for the similar image patch set in the hyperspectral image. And then, the non-local means method, as a regularizer for abundance estimation of sparse unmixing, is used to exploit the abundance image non-local self-similarity. Experimental results on both simulated and real data demonstrate that NLSSU outperforms the other algorithms, with a better spectral unmixing accuracy.

Keywords: hyperspectral unmixing, simultaneous sparse representation, sparse regression, non-local means

Procedia PDF Downloads 233

Authors: Ebenezer O. Olaniyi, Oyebade K. Oyedotun, Khashman Adnan

Abstract:

In the food industry high quality production is required within a limited time to meet up with the demand in the society. In this research work, we have developed a model which can be used to replace the human operator due to their low output in production and slow in making decisions as a result of an individual differences in deciding the defective and healthy banana. This model can perform the vision attributes of human operators in deciding if the banana is defective or healthy for food production based. This research work is divided into two phase, the first phase is the image processing where several image processing techniques such as colour conversion, edge detection, thresholding and morphological operation were employed to extract features for training and testing the network in the second phase. These features extracted in the first phase were used in the second phase; the classification system phase where the multilayer perceptron using backpropagation neural network was employed to train the network. After the network has learned and converges, the network was tested with feedforward neural network to determine the performance of the network. From this experiment, a recognition rate of 97% was obtained and the time taken for this experiment was limited which makes the system accurate for use in the food industry.

Keywords: banana, food processing, identification system, neural network

Procedia PDF Downloads 455

Authors: Rohith J. K., Amir Nazemi, Abbas S. Milani

Abstract:

In Paralympic sports, athletes often depend on some form of equipment to enable competitive sporting, where most of this equipment would only allow passive physiological supports and discrete physiological measurements. Active feedback physiological support and continuous detection of performance indicators, without time or space constraints, would be beneficial in more effective training and performance measures of Paralympic athletes. Moreover, occasionally the athletes suffer from fatigue and muscular stains due to improper monitoring systems. The latter challenges can be overcome by using Smart Composites technology when manufacturing, e.g., knee brace and other sports wearables utilities, where the sensors can be fused together into the fabric and an assisted system actively support the athlete. This paper shows how different sensing functionality may be created by intrinsic and extrinsic modifications onto different types of composite fabrics, depending on the level of integration and the employed functional elements. Results demonstrate that fabric sensors can be well-tailored to measure muscular strain and be used in the fabrication of a smart knee brace as a sample potential application. Materials, connectors, fabric circuits, interconnects, encapsulation and fabrication methods associated with such smart fabric technologies prove to be customizable and versatile.

Keywords: smart composites, sensors, smart fabrics, knee brace

Procedia PDF Downloads 168

Authors: Yong Bum Shin

Abstract:

This case focuses on Weighted additive difference, Conjunctive, Disjunctive, and Elimination by aspects methodologies in consumer decision-making models and the Simple additive weighting (SAW) approach in the multi-criteria decision-making (MCDM) area. Most decision-making models illustrate that the rank reversal phenomenon is unpreventable. This paper presents that rank reversal occurs in popular managerial methods such as Weighted Additive Difference (WAD), Conjunctive Method, Disjunctive Method, Elimination by Aspects (EBA) and MCDM methods as well as such as the Simple Additive Weighting (SAW) and finally Unified Commensurate Multiple (UCM) models which successfully addresses these rank reversal problems in most popular MCDM methods in decision-making area.

Keywords: multiple criteria decision making, rank inconsistency, unified commensurate multiple, analytic hierarchy process

Procedia PDF Downloads 67

Authors: Shaoying Guo, Yanyun Xu, Meng Zhang, Weiqing Huang

Abstract:

The wireless communication network is developing rapidly, thus the wireless security becomes more and more important. Specific emitter identification (SEI) is an vital part of wireless communication security as a technique to identify the unique transmitters. In this paper, a SEI method based on multiscale dispersion entropy (MDE) and refined composite multiscale dispersion entropy (RCMDE) is proposed. The algorithms of MDE and RCMDE are used to extract features for identification of five wireless devices and cross-validation support vector machine (CV-SVM) is used as the classifier. The experimental results show that the total identification accuracy is 99.3%, even at low signal-to-noise ratio(SNR) of 5dB, which proves that MDE and RCMDE can describe the communication signal series well. In addition, compared with other methods, the proposed method is effective and provides better accuracy and stability for SEI.

Keywords: cross-validation support vector machine, refined com- posite multiscale dispersion entropy, specific emitter identification, transient signal, wireless communication device

Procedia PDF Downloads 119

Authors: A. Chateauneuf, M. Mostoufi, D. Vyncke

Abstract:

Monte Carlo (MC) simulation is a technique that provides approximate solutions to a broad range of mathematical problems. A drawback of the method is its high computational cost, especially in a high-dimensional setting, such as estimating the Tail Value-at-Risk for large portfolios or pricing basket options and Asian options. For these types of problems, one can construct an upper bound in the convex order by replacing the copula by the comonotonic copula. This comonotonic upper bound can be computed very quickly, but it gives only a rough approximation. In this paper we introduce the Comonotonic Monte Carlo (CoMC) simulation, by using the comonotonic approximation as a control variate. The CoMC is of broad applicability and numerical results show a remarkable speed improvement. We illustrate the method for estimating Tail Value-at-Risk and pricing basket options and Asian options when the logreturns follow a Black-Scholes model or a variance gamma model.

Keywords: control variate Monte Carlo, comonotonicity, option pricing, scientific computing

Procedia PDF Downloads 499

Authors: Fatma El-Zahraa El-Taher, Ayman Taha, Jane Courtney, Susan Mckeever

Abstract:

Understanding road networks plays an important role in navigation applications such as self-driving vehicles and route planning for individual journeys. Intersections of roads are essential components of road networks. Understanding the features of an intersection, from a simple T-junction to larger multi-road junctions, is critical to decisions such as crossing roads or selecting the safest routes. The identification and profiling of intersections from satellite images is a challenging task. While deep learning approaches offer the state-of-the-art in image classification and detection, the availability of training datasets is a bottleneck in this approach. In this paper, a labelled satellite image dataset for the intersection recognition problem is presented. It consists of 14,692 satellite images of Washington DC, USA. To support other users of the dataset, an automated download and labelling script is provided for dataset replication. The challenges of construction and fine-grained feature labelling of a satellite image dataset is examined, including the issue of how to address features that are spread across multiple images. Finally, the accuracy of the detection of intersections in satellite images is evaluated.

Keywords: satellite images, remote sensing images, data acquisition, autonomous vehicles

Procedia PDF Downloads 124

Authors: Jamshaid Ul Rahman, Faiza Makhdoom, Dianchen Lu

Abstract:

Activation functions play a decisive role in determining the capacity of Deep Neural Networks (DNNs) as they enable neural networks to capture inherent nonlinearities present in data fed to them. The prior research on activation functions primarily focused on the utility of monotonic or non-oscillatory functions, until Growing Cosine Unit (GCU) broke the taboo for a number of applications. In this paper, a Convolutional Neural Network (CNN) model named as ASU-CNN is proposed which utilizes recently designed activation function ASU across its layers. The effect of this non-monotonic and oscillatory function is inspected through feature map visualizations from different convolutional layers. The optimization of proposed network is offered by Adam with a fine-tuned adjustment of learning rate. The network achieved promising results on both training and testing data for the classification of CIFAR-10. The experimental results affirm the computational feasibility and efficacy of the proposed model for performing tasks related to the field of computer vision.

Keywords: amplifying sine unit, activation function, convolutional neural networks, oscillatory activation, image classification, CIFAR-10

Procedia PDF Downloads 86

Authors: Mahdi Mottahedi, Peter Zahn, Armin Lechler, Alexander Verl

Abstract:

Optimal topology of components leads to the maximum stiffness with the minimum material use. For the generation of these topologies, normally algorithms are employed, which tackle manufacturing limitations, at the cost of the optimal result. The global optimum result with penalty factor one, however, cannot be fabricated with conventional methods. In this article, an additive manufacturing method is introduced, in order to enable the production of global topology optimization results. For a benchmark, topology optimization with higher and lower penalty factors are performed. Different algorithms are employed in order to interpret the results of topology optimization with lower factors in many microstructure layers. These layers are then joined to form the final geometry. The algorithms’ benefits are then compared experimentally and numerically for the best interpretation. The findings demonstrate that by implementation of the selected algorithm, the stiffness of the components produced with this method is higher than what could have been produced by conventional techniques.

Keywords: topology optimization, additive manufacturing, 3D-printer, laminated object manufacturing

Procedia PDF Downloads 326

Authors: Suparman Suparman

Abstract:

A white noise in autoregressive (AR) model is often assumed to be normally distributed. In application, the white noise usually do not follows a normal distribution. This paper aims to estimate a parameter of AR model that has a exponential white noise. A Bayesian method is adopted. A prior distribution of the parameter of AR model is selected and then this prior distribution is combined with a likelihood function of data to get a posterior distribution. Based on this posterior distribution, a Bayesian estimator for the parameter of AR model is estimated. Because the order of AR model is considered a parameter, this Bayesian estimator cannot be explicitly calculated. To resolve this problem, a method of reversible jump Markov Chain Monte Carlo (MCMC) is adopted. A result is a estimation of the parameter AR model can be simultaneously calculated.

Keywords: autoregressive (AR) model, exponential white Noise, bayesian, reversible jump Markov Chain Monte Carlo (MCMC)

Procedia PDF Downloads 343

Authors: Sonia Perez-Gamboa, Qingquan Sun, Yan Zhang

Abstract:

Non-intrusive sensor-based human activity recognition (HAR) is utilized in a spectrum of applications, including fitness tracking devices, gaming, health care monitoring, and smartphone applications. Deep learning models such as convolutional neural networks (CNNs) and long short term memory (LSTM) recurrent neural networks (RNNs) provide a way to achieve HAR accurately and effectively. In this paper, we design a multi-layer hybrid architecture with CNN and LSTM and explore a variety of multi-layer combinations. Based on the exploration, we present a lightweight, hybrid, and multi-layer model, which can improve the recognition performance by integrating local features and scale-invariant with dependencies of activities. The experimental results demonstrate the efficacy of the proposed model, which can achieve a 94.7% activity recognition rate on a benchmark human activity dataset. This model outperforms traditional machine learning and other deep learning methods. Additionally, our implementation achieves a balance between recognition rate and training time consumption.

Keywords: deep learning, LSTM, CNN, human activity recognition, inertial sensor

Procedia PDF Downloads 131