Search results for: Gaussian kernel

Authors: Jennifer Cuellar, Angie L. Parada, Kevin N. S. Chacon, Sol M. Mejia

Abstract:

The purification of bio-ethanol through distillation methods is an unresolved issue at the biofuel industry because of the ethanol-water azeotrope formation, which increases the steps of the purification process and subsequently increases the production costs. Therefore, understanding the mixture nature at the molecular level could provide new insights for improving the current methods and/or designing new and more efficient purification methods. For that reason, the present study focuses on the evaluation and analysis of (ethanol)₉-water heterodecamers, as the systems with the minimum molecular proportion that represents the azeotropic concentration (96 %m/m in ethanol). The computational modelling was carried out with B3LYP-D3/6-311++G(d,p) in Gaussian 09. Initial explorations of the potential energy surface were done through two methods: annealing simulated runs and molecular dynamics trajectories besides intuitive structures obtained from smaller (ethanol)n-water heteroclusters, n = 7, 8 and 9. The energetic order of the seven stable heterodecamers determines the most stable heterodecamer (Hdec-1) as a structure forming a bicyclic geometry with the O-H---O hydrogen bonds (HBs) where the water is a double proton donor molecule. Hdec-1 combines 1 water molecule and the same quantity of every ethanol conformer; this is, 3 trans, 3 gauche 1 and 3 gauche 2; its abundance is 89%, its decamerization energy is -80.4 kcal/mol, i.e. 13 kcal/mol most stable than the less stable heterodecamer. Besides, a way to understand why methanol does not form an azeotropic mixture with water, analogous systems ((ethanol)10, (methanol)10, and (methanol)9-water)) were optimized. Topologic analysis of the electron density reveals that Hec-1 forms 33 weak interactions in total: 11 O-H---O, 8 C-H---O, 2 C-H---C hydrogen bonds and 12 H---H interactions. The strength and abundance of the most unconventional interactions (H---H, C-H---O and C-H---O) seem to explain the preference of the ethanol for forming heteroclusters instead of clusters. Besides, O-H---O HBs present a significant covalent character according to topologic parameters as the Laplacian of electron density and the relationship between potential and kinetic energy densities evaluated at the bond critical points; obtaining negatives values and values between 1 and 2, for those two topological parameters, respectively.

Keywords: ADMP, DFT, ethanol-water azeotrope, Grimme dispersion correction, simulated annealing, weak interactions

Procedia PDF Downloads 85

Authors: Asif Arshid, Ying Huang, Denver Tolliver

Abstract:

Finite Element (FE) method coupled with ever-increasing computational powers has substantially advanced the reliability of deterministic three dimensional structural analyses of a structure with uniform material properties. However, railways trackbed is made up of diverse group of materials including steel, wood, rock and soil, while each material has its own varying levels of heterogeneity and imperfections. It is observed that the application of probabilistic methods for trackbed structural analysis while incorporating the material and geometric variabilities is deeply underworked. The authors developed and validated a 3-dimensional FE based numerical trackbed model and in this study, they investigated the influence of variability in Young modulus and thicknesses of granular layers (Ballast and Subgrade) on the reliability index (-index) of the subgrade layer. The influence of these factors is accounted for by changing their Coefficients of Variance (COV) while keeping their means constant. These variations are formulated using Gaussian Normal distribution. Two failure mechanisms in subgrade namely Progressive Shear Failure and Excessive Plastic Deformation are examined. Preliminary results of risk-based probabilistic analysis for Progressive Shear Failure revealed that the variations in Ballast depth are the most influential factor for vertical stress at the top of subgrade surface. Whereas, in case of Excessive Plastic Deformations in subgrade layer, the variations in its own depth and Young modulus proved to be most important while ballast properties remained almost indifferent. For both these failure moods, it is also observed that the reliability index for subgrade failure increases with the increase in COV of ballast depth and subgrade Young modulus. The findings of this work is of particular significance in studying the combined effect of construction imperfections and variations in ground conditions on the structural performance of railroad trackbed and evaluating the associated risk involved. In addition, it also provides an additional tool to supplement the deterministic analysis procedures and decision making for railroad maintenance.

Keywords: finite element analysis, numerical modeling, probabilistic methods, risk and reliability analysis, subgrade

Procedia PDF Downloads 102

Authors: Ayodeji Oluleye

Abstract:

This study used Total Ozone Mapping Spectrometer (TOMS) Aerosol Index (AI) and reanalysis dataset of thirty years (1983-2012) to investigate the influence of the atmospheric circulation on dust transport during the Harmattan period over WestAfrica using TOMS data. The Harmattan dust mobilization and atmospheric circulation pattern were evaluated using a kernel density estimate which shows the areas where most points are concentrated between the variables. The evolution of the Inter-Tropical Discontinuity (ITD), Sea surface Temperature (SST) over the Gulf of Guinea, and the North Atlantic Oscillation (NAO) index during the Harmattan period (November-March) was also analyzed and graphs of the average ITD positions, SST and the NAO were observed on daily basis. The Pearson moment correlation analysis was also employed to assess the effect of atmospheric circulation on Harmattan dust transport. The results show that the departure (increased) of TOMS AI values from the long-term mean (1.64) occurred from around 21st of December, which signifies the rich dust days during winter period. Strong TOMS AI signal were observed from January to March with the maximum occurring in the latter months (February and March). The inter-annual variability of TOMSAI revealed that the rich dust years were found between 1984-1985, 1987-1988, 1997-1998, 1999-2000, and 2002-2004. Significantly, poor dust year was found between 2005 and 2006 in all the periods. The study has found strong north-easterly (NE) trade winds were over most of the Sahelianregion of West Africa during the winter months with the maximum wind speed reaching 8.61m/s inJanuary.The strength of NE winds determines the extent of dust transport to the coast of Gulf of Guinea during winter. This study has confirmed that the presence of the Harmattan is strongly dependent on theSST over Atlantic Ocean and ITD position. The locus of the average SST and ITD positions over West Africa could be described by polynomial functions. The study concludes that the evolution of near surface wind field at 925 hpa, and the variations of SST and ITD positions are the major large scale atmospheric circulation systems driving the emission, distribution, and transport of Harmattan dust aerosols over West Africa. However, the influence of NAO was shown to have fewer significance effects on the Harmattan dust transport over the region.

Keywords: atmospheric circulation, dust aerosols, Harmattan, West Africa

Procedia PDF Downloads 289

Authors: Thomas Bryan, Veton Kepuska, Ivica Kostanic

Abstract:

Speaker recognition is performed in high Additive White Gaussian Noise (AWGN) environments using principals of Computational Auditory Scene Analysis (CASA). CASA methods often classify sounds from images in the time-frequency (T-F) plane using spectrograms or cochleargrams as the image. In this paper atomic decomposition implemented by matching pursuit performs a transform from time series speech signals to the T-F plane. The atomic decomposition creates a sparsely populated T-F vector in “weight space” where each populated T-F position contains an amplitude weight. The weight space vector along with the atomic dictionary represents a denoised, compressed version of the original signal. The arraignment or of the atomic indices in the T-F vector are used for classification. Unsupervised feature learning implemented by a sparse autoencoder learns a single dictionary of basis features from a collection of envelope samples from all speakers. The approach is demonstrated using pairs of speakers from the TIMIT data set. Pairs of speakers are selected randomly from a single district. Each speak has 10 sentences. Two are used for training and 8 for testing. Atomic index probabilities are created for each training sentence and also for each test sentence. Classification is performed by finding the lowest Euclidean distance between then probabilities from the training sentences and the test sentences. Training is done at a 30dB Signal-to-Noise Ratio (SNR). Testing is performed at SNR’s of 0 dB, 5 dB, 10 dB and 30dB. The algorithm has a baseline classification accuracy of ~93% averaged over 10 pairs of speakers from the TIMIT data set. The baseline accuracy is attributable to short sequences of training and test data as well as the overall simplicity of the classification algorithm. The accuracy is not affected by AWGN and produces ~93% accuracy at 0dB SNR.

Keywords: time-frequency plane, atomic decomposition, envelope sampling, Gabor atoms, matching pursuit, sparse dictionary learning, sparse autoencoder

Procedia PDF Downloads 260

Authors: Ayman Mutahar Alrassas, Shaoran Ren, Renyuan Ren, Hung Vo Thanh, Mohammed Hail Hakimi, Zhenliang Guan

Abstract:

The significant effect of CO₂ on global climate and the environment has gained more concern worldwide. Enhance oil recovery (EOR) associated with sequestration of CO₂ particularly into the depleted oil reservoir is considered the viable approach under financial limitations since it improves the oil recovery from the existing oil reservoir and boosts the relation between global-scale of CO₂ capture and geological sequestration. Consequently, practical measurements are required to attain large-scale CO₂ emission reduction. This paper presents an integrated modeling workflow to construct an accurate 3D reservoir geological model to estimate the storage capacity of CO₂ under geological uncertainty in an unconventional oil reservoir of the Paleogene Shahejie Formation (Es1) in the block Nv32, Shenvsi oilfield, China. In this regard, geophysical data, including well logs of twenty-two well locations and seismic data, were combined with geological and engineering data and used to construct a 3D reservoir geological modeling. The geological modeling focused on four tight reservoir units of the Shahejie Formation (Es1-x1, Es1-x2, Es1-x3, and Es1-x4). The validated 3D reservoir models were subsequently used to calculate the theoretical CO₂ storage capacity in the block Nv32, Shenvsi oilfield. Well logs were utilized to predict petrophysical properties such as porosity and permeability, and lithofacies and indicate that the Es1 reservoir units are mainly sandstone, shale, and limestone with a proportion of 38.09%, 32.42%, and 29.49, respectively. Well log-based petrophysical results also show that the Es1 reservoir units generally exhibit 2–36% porosity, 0.017 mD to 974.8 mD permeability, and moderate to good net to gross ratios. These estimated values of porosity, permeability, lithofacies, and net to gross were up-scaled and distributed laterally using Sequential Gaussian Simulation (SGS) and Simulation Sequential Indicator (SIS) methods to generate 3D reservoir geological models. The reservoir geological models show there are lateral heterogeneities of the reservoir properties and lithofacies, and the best reservoir rocks exist in the Es1-x4, Es1-x3, and Es1-x2 units, respectively. In addition, the reservoir volumetric of the Es1 units in block Nv32 was also estimated based on the petrophysical property models and fund to be between 0.554368

Keywords: CO₂ storage capacity, 3D geological model, geological uncertainty, unconventional oil reservoir, block Nv32

Procedia PDF Downloads 145

Authors: Mohd Ayyub Khan, Nadeem Akhtar

Abstract:

Due to recent advances in wireless communication technologies and hand-held devices, there is a huge demand for video-based applications such as video surveillance, video conferencing, remote surgery, Digital Video Broadcast (DVB), IPTV, online learning courses, YouTube, WhatsApp, Instagram, Facebook, Interactive Video Games. However, the raw videos posses very high bandwidth which makes the compression a must before its transmission over the wireless channels. The High Efficiency Video Codec (HEVC) (also called H.265) is latest state-of-the-art video coding standard developed by the Joint effort of ITU-T and ISO/IEC teams. HEVC is targeted for high resolution videos such as 4K or 8K resolutions that can fulfil the recent demands for video services. The compression ratio achieved by the HEVC is twice as compared to its predecessor H.264/AVC for same quality level. The compression efficiency is generally increased by removing more correlation between the frames/pixels using complex techniques such as extensive intra and inter prediction techniques. As more correlation is removed, the chances of interdependency among coded bits increases. Thus, bit errors may have large effect on the reconstructed video. Sometimes even single bit error can lead to catastrophic failure of the reconstructed video. In this paper, we study the performance of HEVC bitstream over additive white Gaussian noise (AWGN) channel. Moreover, HEVC over Quadrature Amplitude Modulation (QAM) combined with forward error correction (FEC) schemes are also explored over the noisy channel. The video will be encoded using HEVC, and the coded bitstream is channel coded to provide some redundancies. The channel coded bitstream is then modulated using QAM and transmitted over AWGN channel. At the receiver, the symbols are demodulated and channel decoded to obtain the video bitstream. The bitstream is then used to reconstruct the video using HEVC decoder. It is observed that as the signal to noise ratio of channel is decreased the quality of the reconstructed video decreases drastically. Using proper FEC codes, the quality of the video can be restored up to certain extent. Thus, the performance analysis of HEVC presented in this paper may assist in designing the optimized code rate of FEC such that the quality of the reconstructed video is maximized over wireless channels.

Keywords: AWGN, forward error correction, HEVC, video coding, QAM

Procedia PDF Downloads 121

Authors: Temitayo Abayomi Ewemoje, Oluwamayowa Oluwafemi Oluwaniyi

Abstract:

Agriculture and food processing, industry are among the largest industrial sectors that uses large amount of energy. Thus, a larger amount of gases from their fuel combustion technologies is being released into the environment. The choice of input energy supply not only directly having affects the environment, but also poses a threat to human health. The study was therefore designed to assess each unit production processes in order to identify hotspots using life cycle assessments (LCA) approach in South-western Nigeria. Data such as machine power rating, operation duration, inputs and outputs of shea butter materials for unit processes obtained at site were used to modelled Life Cycle Impact Analysis on GaBi6 (Holistic Balancing) software. Four scenarios were drawn for the impact assessments. Material sourcing from Kaiama, Scenarios 1, 3 and Minna Scenarios 2, 4 but different heat supply sources (Liquefied Petroleum Gas ‘LPG’ Scenarios 1, 2 and 10.8 kW Diesel Heater, scenarios 3, 4). Modelling of shea butter production on GaBi6 was for 1kg functional unit of shea butter produced and the Tool for the Reduction and Assessment of Chemical and other Environmental Impacts (TRACI) midpoint assessment was tool used to was analyse the life cycle inventories of the four scenarios. Eight categories in all four Scenarios were observed out of which three impact categories; Global Warming Potential (GWP) (0.613, 0.751, 0.661, 0.799) kg CO2¬-Equiv., Acidification Potential (AP) (0.112, 0.132, 0.129, 0.149) kg H+ moles-Equiv., and Smog (0.044, 0.059, 0.049, 0.063) kg O3-Equiv., categories had the greater impacts on the environment in Scenarios 1-4 respectively. Impacts from transportation activities was also seen to contribute more to these environmental impact categories due to large volume of petrol combusted leading to releases of gases such as CO2, CH4, N2O, SO2, and NOx into the environment during the transportation of raw shea kernel purchased. The ratio of transportation distance from Minna and Kaiama to production site was approximately 3.5. Shea butter unit processes with greater impacts in all categories was the packaging, milling and with the churning processes in ascending order of magnitude was identified as hotspots that may require attention. From the 1kg shea butter functional unit, it was inferred that locating production site at the shortest travelling distance to raw material sourcing and combustion of LPG for heating would reduce all the impact categories assessed on the environment.

Keywords: GaBi6, Life cycle assessment, shea butter production, TRACI

Procedia PDF Downloads 287

Authors: Gabriela D. Silva, Rodrigo L. O. R. Cunha, Mauricio D. Coutinho-Neto

Abstract:

In the last four decades the biological activities of selenium compounds has received great attention, particularly for hypervalent derivates from selenium (IV) used as enzyme inhibitors. The unregulated activity of cysteine proteases are related to the development of several pathologies, such as neurological disorders, cardiovascular diseases, obesity, rheumatoid arthritis, cancer and parasitic infections. These enzymes are therefore a valuable target for designing new small molecule inhibitors such as selenuranes. Even tough there has been advances in the synthesis and design of new selenuranes based inhibitors, little is known about their mechanism of action. It is a given that inhibition occurs through the reaction between the thiol group of the enzyme and the chalcogen atom. However, several open questions remain about the nature of the mechanism (associative vs. dissociative) and about the nature of the reactive species in solution under physiological conditions. In this work we performed a theoretical investigation on model systems to study the possible routes of substitution reactions. Nucleophiles may be present in biological systems, our interest is centered in the thiol groups from the cysteine proteases and the hydroxyls from the aqueous environment. We therefore expect this study to clarify the possibility of a route reaction in two stages, the first consisting of the substitution of chloro atoms by hydroxyl groups and then replacing these hydroxyl groups per thiol groups in selenuranes. The structures of selenuranes and nucleophiles were optimized using density function theory along the B3LYP functional and a 6-311+G(d) basis set. Solvent was treated using the IEFPCM method as implemented in the Gaussian 09 code. Our results indicate that hydrolysis from water react preferably with selenuranes, and then, they are replaced by the thiol group. It show the energy values of -106,0730423 kcal/mol for dople substituition by hydroxyl group and 96,63078511 kcal/mol for thiol group. The solvatation and pH reduction promotes this route, increasing the energy value for reaction with hydroxil group to -50,75637672 kcal/mol and decreasing the energy value for thiol to 7,917767189 kcal/mol. Alternative ways were analyzed for monosubstitution (considering the competition between Cl, OH and SH groups) and they suggest the same route. Similar results were obtained for aliphatic and aromatic selenuranes studied.

Keywords: chalcogenes, computational study, cysteine proteases, enzyme inhibitors

Procedia PDF Downloads 276

Authors: Inyoung Park, Youngjoo Woo, Euiseong Seo

Abstract:

With the advent of WebGL, Web apps are now able to provide high quality graphics by utilizing the underlying graphic processing units (GPUs). Despite that the Web apps are becoming common and popular, the current power management schemes, which were devised for the conventional native applications, are suboptimal for Web apps because of the additional layer, the Web browser, between OS and application. The Web browser running on a CPU issues GL commands, which are for rendering images to be displayed by the Web app currently running, to the GPU and the GPU processes them. The size and number of issued GL commands determine the processing load of the GPU. While the GPU is processing the GL commands, CPU simultaneously executes the other compute intensive threads. The actual user experience will be determined by either CPU processing or GPU processing depending on which of the two is the more demanded resource. For example, when the GPU work queue is saturated by the outstanding commands, lowering the performance level of the CPU does not affect the user experience because it is already deteriorated by the retarded execution of GPU commands. Consequently, it would be desirable to lower CPU or GPU performance level to save energy when the other resource is saturated and becomes a bottleneck in the execution flow. Based on this observation, we propose a power management scheme that is specialized for the Web app runtime environment. This approach incurs two technical challenges; identification of the bottleneck resource and determination of the appropriate performance level for unsaturated resource. The proposed power management scheme uses the CPU utilization level of the Window Manager to tell which one is the bottleneck if exists. The Window Manager draws the final screen using the processed results delivered from the GPU. Thus, the Window Manager is on the critical path that determines the quality of user experience and purely executed by the CPU. The proposed scheme uses the weighted average of the Window Manager utilization to prevent excessive sensitivity and fluctuation. We classified Web apps into three categories using the analysis results that measure frame-per-second (FPS) changes under diverse CPU/GPU clock combinations. The results showed that the capability of the CPU decides user experience when the Window Manager utilization is above 90% and consequently, the proposed scheme decreases the performance level of CPU by one step. On the contrary, when its utilization is less than 60%, the bottleneck usually lies in the GPU and it is desirable to decrease the performance of GPU. Even the processing unit that is not on critical path, excessive performance drop can occur and that may adversely affect the user experience. Therefore, our scheme lowers the frequency gradually, until it finds an appropriate level by periodically checking the CPU utilization. The proposed scheme reduced the energy consumption by 10.34% on average in comparison to the conventional Linux kernel, and it worsened their FPS by 1.07% only on average.

Keywords: interactive applications, power management, QoS, Web apps, WebGL

Procedia PDF Downloads 164

Authors: Carla Idely Palencia-Aguilar

Abstract:

Artificial wetlands play an important role at Guasca Municipality in Colombia, not only because they are used for the agroindustry, but also because more than 45 species were found, some of which are endemic and migratory birds. Remote sensing was used to determine the changes in the area occupied by water of artificial wetlands by means of Aster and Modis images for different time periods. Evapotranspiration was also determined by three methods: Surface Energy Balance System-Su (SEBS) algorithm, Surface Energy Balance- Bastiaanssen (SEBAL) algorithm, and Potential Evapotranspiration- FAO. Empirical equations were also developed to determine the relationship between Normalized Difference Vegetation Index (NDVI) versus net radiation, ambient temperature and rain with an obtained R2 of 0.83. Groundwater level fluctuations on a daily basis were studied as well. Data from a piezometer placed next to the wetland were fitted with rain changes (with two weather stations located at the proximities of the wetlands) by means of multiple regression and time series analysis, the R2 from the calculated and measured values resulted was higher than 0.98. Information from nearby weather stations provided information for ordinary kriging as well as the results for the Digital Elevation Model (DEM) developed by using PCI software. Standard models (exponential, spherical, circular, gaussian, linear) to describe spatial variation were tested. Ordinary Cokriging between height and rain variables were also tested, to determine if the accuracy of the interpolation would increase. The results showed no significant differences giving the fact that the mean result of the spherical function for the rain samples after ordinary kriging was 58.06 and a standard deviation of 18.06. The cokriging using for the variable rain, a spherical function; for height variable, the power function and for the cross variable (rain and height), the spherical function had a mean of 57.58 and a standard deviation of 18.36. Threatens of eutrophication were also studied, given the unconsciousness of neighbours and government deficiency. Water quality was determined over the years; different parameters were studied to determine the chemical characteristics of water. In addition, 600 pesticides were studied by gas and liquid chromatography. Results showed that coliforms, nitrogen, phosphorous and prochloraz were the most significant contaminants.

Keywords: DEM, evapotranspiration, geostatistics, NDVI

Procedia PDF Downloads 90

Authors: Huayuan Lu, Chunfang Yang, Ma Zhu, Baojun Qi, Yaqiong Qiao, Jiangqian Xu

Abstract:

Image geolocalization has great application prospects in fields such as autonomous driving and virtual/augmented reality. In practical application scenarios, the size of the image to be located is not fixed; it is impractical to train different networks for all possible sizes. When its size does not match the size of the input of the descriptor extraction model, existing image geolocalization methods usually directly scale or crop the image in some common ways. This will result in the loss of some information important to the geolocalization task, thus affecting the performance of the image geolocalization method. For example, excessive down-sampling can lead to blurred building contour, and inappropriate cropping can lead to the loss of key semantic elements, resulting in incorrect geolocation results. To address this problem, this paper designs a learnable image resizer and proposes an arbitrary-sized image geolocation method. (1) The designed learnable image resizer employs the self-attention mechanism to enhance the geometric features of the resized image. Firstly, it applies bilinear interpolation to the input image and its feature maps to obtain the initial resized image and the resized feature maps. Then, SKNet (selective kernel net) is used to approximate the best receptive field, thus keeping the geometric shapes as the original image. And SENet (squeeze and extraction net) is used to automatically select the feature maps with strong contour information, enhancing the geometric features. Finally, the enhanced geometric features are fused with the initial resized image, to obtain the final resized images. (2) The proposed image geolocalization method embeds the above image resizer as a fronting layer of the descriptor extraction network. It not only enables the network to be compatible with arbitrary-sized input images but also enhances the geometric features that are crucial to the image geolocalization task. Moreover, the triplet attention mechanism is added after the first convolutional layer of the backbone network to optimize the utilization of geometric elements extracted by the first convolutional layer. Finally, the local features extracted by the backbone network are aggregated to form image descriptors for image geolocalization. The proposed method was evaluated on several mainstream datasets, such as Pittsburgh30K, Tokyo24/7, and Places365. The results show that the proposed method has excellent size compatibility and compares favorably to recently mainstream geolocalization methods.

Keywords: image geolocalization, self-attention mechanism, image resizer, geometric feature

Procedia PDF Downloads 178

Authors: Hamza Benzerrouk, Alexander Nebylov

Abstract:

In classical GNSS/INS integration designs, the loosely coupled approach uses the GNSS derived position and the velocity as the measurements vector. This design is suboptimal from the standpoint of preventing GNSSoutliers/outages. The tightly coupled GPS/INS navigation filter mixes the GNSS pseudo range and inertial measurements and obtains the vehicle navigation state as the final navigation solution. The ultra‐tightly coupled GNSS/INS design combines the I (inphase) and Q(quadrature) accumulator outputs in the GNSS receiver signal tracking loops and the INS navigation filter function intoa single Kalman filter variant (EKF, UKF, SPKF, CKF and HCKF). As mentioned, EKF and UKF are the most used nonlinear filters in the literature and are well adapted to inertial navigation state estimation when integrated with GNSS signal outputs. In this paper, it is proposed to move a step forward with more accurate filters and modern approaches called Cubature and High Degree cubature Kalman Filtering methods, on the basis of previous results solving the state estimation based on INS/GNSS integration, Cubature Kalman Filter (CKF) and High Degree Cubature Kalman Filter with (HCKF) are the references for the recent developed generalized Cubature rule based Kalman Filter (GCKF). High degree cubature rules are the kernel of the new solution for more accurate estimation with less computational complexity compared with the Gauss-Hermite Quadrature (GHQKF). Gauss-Hermite Kalman Filter GHKF which is not selected in this work because of its limited real-time implementation in high-dimensional state-spaces. In ultra tightly or a deeply coupled GNSS/INS system is dynamics EKF is used with transition matrix factorization together with GNSS block processing which is well described in the paper and assumes available the intermediary frequency IF by using a correlator samples with a rate of 500 Hz in the presented approach. GNSS (GPS+GLONASS) measurements are assumed available and modern SPKF with Cubature Kalman Filter (CKF) are compared with new versions of CKF called high order CKF based on Spherical-radial cubature rules developed at the fifth order in this work. Estimation accuracy of the high degree CKF is supposed to be comparative to GHKF, results of state estimation are then observed and discussed for different initialization parameters. Results show more accurate navigation state estimation and more robust GNSS receiver when Ultra Tightly Coupled approach applied based on High Degree Cubature Kalman Filter.

Keywords: GNSS, INS, Kalman filtering, ultra tight integration

Procedia PDF Downloads 259

Authors: Gómez R. Marta, Martín M. Jesús María

Abstract:

The general aim of this paper identifies the areas of highest concentration of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) around the incinerator through the use of dispersion models. Atmospheric dispersion models are useful tools for estimating and prevent the impact of emissions from a particular source in air quality. These models allow considering different factors that influence in air pollution: source characteristics, the topography of the receiving environment and weather conditions to predict the pollutants concentration. The PCDD/Fs, after its emission into the atmosphere, are deposited on water or land, near or far from emission source depending on the size of the associated particles and climatology. In this way, they are transferred and mobilized through environmental compartments. The modelling of PCDD/Fs was carried out with following tools: Atmospheric Dispersion Model Software (ADMS) and Surfer. ADMS is a dispersion model Gaussian plume, used to model the impact of air quality industrial facilities. And Surfer is a program of surfaces which is used to represent the dispersion of pollutants on a map. For the modelling of emissions, ADMS software requires the following input parameters: characterization of emission sources (source type, height, diameter, the temperature of the release, flow rate, etc.) meteorological and topographical data (coordinate system), mainly. The study area was set at 5 Km around the incinerator and the first population center nearest to focus PCDD/Fs emission is about 2.5 Km, approximately. Data were collected during one year (2013) both PCDD/Fs emissions of the incinerator as meteorology in the study area. The study has been carried out during period's average that legislation establishes, that is to say, the output parameters are taking into account the current legislation. Once all data required by software ADMS, described previously, are entered, and in order to make the representation of the spatial distribution of PCDD/Fs concentration and the areas affecting them, the modelling was proceeded. In general, the dispersion plume is in the direction of the predominant winds (Southwest and Northeast). Total levels of PCDD/Fs usually found in air samples, are from <2 pg/m3 for remote rural areas, from 2-15 pg/m3 in urban areas and from 15-200 pg/m3 for areas near to important sources, as can be an incinerator. The results of dispersion maps show that maximum concentrations are the order of 10-8 ng/m3, well below the values considered for areas close to an incinerator, as in this case.

Keywords: atmospheric dispersion, dioxin, furan, incinerator

Procedia PDF Downloads 180

Authors: Magdalena Igras, Joanna Grzybowska, Mariusz Ziółko

Abstract:

The study of emotions expressed in emergency phone call is presented, covering both statistical analysis of emotions configurations and an attempt to automatically classify emotions. An emergency call is a situation usually accompanied by intense, authentic emotions. They influence (and may inhibit) the communication between caller and responder. In order to support responders in their responsible and psychically exhaustive work, we studied when and in which combinations emotions appeared in calls. A corpus of 45 hours of conversations (about 3300 calls) from emergency call center was collected. Each recording was manually tagged with labels of emotions valence (positive, negative or neutral), type (sadness, tiredness, anxiety, surprise, stress, anger, fury, calm, relief, compassion, satisfaction, amusement, joy) and arousal (weak, typical, varying, high) on the basis of perceptual judgment of two annotators. As we concluded, basic emotions tend to appear in specific configurations depending on the overall situational context and attitude of speaker. After performing statistical analysis we distinguished four main types of emotional behavior of callers: worry/helplessness (sadness, tiredness, compassion), alarm (anxiety, intense stress), mistake or neutral request for information (calm, surprise, sometimes with amusement) and pretension/insisting (anger, fury). The frequency of profiles was respectively: 51%, 21%, 18% and 8% of recordings. A model of presenting the complex emotional profiles on the two-dimensional (tension-insecurity) plane was introduced. In the stage of acoustic analysis, a set of prosodic parameters, as well as Mel-Frequency Cepstral Coefficients (MFCC) were used. Using these parameters, complex emotional states were modeled with machine learning techniques including Gaussian mixture models, decision trees and discriminant analysis. Results of classification with several methods will be presented and compared with the state of the art results obtained for classification of basic emotions. Future work will include optimization of the algorithm to perform in real time in order to track changes of emotions during a conversation.

Keywords: acoustic analysis, complex emotions, emotion recognition, machine learning

Procedia PDF Downloads 368

Authors: Seema Paul, Jesper Oppelstrup, Roger Thunvik, Vladimir Cvetkovic

Abstract:

Lake Victoria is the second largest fresh water body in the world, located in East Africa with a catchment area of 250,000 km², of which 68,800 km² is the actual lake surface. The hydrodynamic processes of the shallow (40–80 m deep) water system are unique due to its location at the equator, which makes Coriolis effects weak. The paper describes a St.Venant shallow water model of Lake Victoria developed in COMSOL Multiphysics software, a general purpose finite element tool for solving partial differential equations. Depth soundings taken in smaller parts of the lake were combined with recent more extensive data to resolve the discrepancies of the lake shore coordinates. The topography model must have continuous gradients, and Delaunay triangulation with Gaussian smoothing was used to produce the lake depth model. The model shows large-scale flow patterns, passive tracer concentration and water level variations in response to river and tracer inflow, rain and evaporation, and wind stress. Actual data of precipitation, evaporation, in- and outflows were applied in a fifty-year simulation model. It should be noted that the water balance is dominated by rain and evaporation and model simulations are validated by Matlab and COMSOL. The model conserves water volume, the celerity gradients are very small, and the volume flow is very slow and irrotational except at river mouths. Numerical experiments show that the single outflow can be modelled by a simple linear control law responding only to mean water level, except for a few instances. Experiments with tracer input in rivers show very slow dispersion of the tracer, a result of the slow mean velocities, in turn, caused by the near-balance of rain with evaporation. The numerical and hydrodynamical model can evaluate the effects of wind stress which is exerted by the wind on the lake surface that will impact on lake water level. Also, model can evaluate the effects of the expected climate change, as manifest in changes to rainfall over the catchment area of Lake Victoria in the future.

Keywords: bathymetry, lake flow and steady state analysis, water level validation and concentration, wind stress

Procedia PDF Downloads 191

Authors: Aprajeeta Jha, Punyadarshini P. Tripathy

Abstract:

Drying technologies for various food processing operations shares an inevitable linkage with energy, cost and environmental sustainability. Hence, solar drying of food grains has become imperative choice to combat duo challenges of meeting high energy demand for drying and to address climate change scenario. But performance and reliability of solar dryers depend hugely on sunshine period, climatic conditions, therefore, offer a limited control over drying conditions and have lower efficiencies. Solar drying technology, supported by Photovoltaic (PV) power plant and hybrid type solar air collector can potentially overpower the disadvantages of solar dryers. For development of such robust hybrid dryers; to ensure quality and shelf-life of paddy grains the optimization of process parameter becomes extremely critical. Investigation of the moisture distribution profile within the grains becomes necessary in order to avoid over drying or under drying of food grains in hybrid solar dryer. Computational simulations based on finite element modeling can serve as potential tool in providing a better insight of moisture migration during drying process. Hence, present work aims at optimizing the process parameters and to develop a 3-dimensional (3D) finite element model (FEM) for predicting moisture profile in paddy during solar drying. COMSOL Multiphysics was employed to develop a 3D finite element model for predicting moisture profile. Furthermore, optimization of process parameters (power level, air velocity and moisture content) was done using response surface methodology in design expert software. 3D finite element model (FEM) for predicting moisture migration in single kernel for every time step has been developed and validated with experimental data. The mean absolute error (MAE), mean relative error (MRE) and standard error (SE) were found to be 0.003, 0.0531 and 0.0007, respectively, indicating close agreement of model with experimental results. Furthermore, optimized process parameters for drying paddy were found to be 700 W, 2.75 m/s at 13% (wb) with optimum temperature, milling yield and drying time of 42˚C, 62%, 86 min respectively, having desirability of 0.905. Above optimized conditions can be successfully used to dry paddy in PV integrated solar dryer in order to attain maximum uniformity, quality and yield of product. PV-integrated hybrid solar dryers can be employed as potential and cutting edge drying technology alternative for sustainable energy and food security.

Keywords: finite element modeling, moisture migration, paddy grain, process optimization, PV integrated hybrid solar dryer

Procedia PDF Downloads 120

Authors: Elia Favarelli, Enrico Testi, Andrea Giorgetti

Abstract:

Autonomous structural health monitoring (SHM) of many structures and bridges became a topic of paramount importance for maintenance purposes and safety reasons. This paper proposes a set of machine learning (ML) tools to perform automatic feature selection and detection of anomalies in a bridge from vibrational data and compare different feature extraction schemes to increase the accuracy and reduce the amount of data collected. As a case study, the Z-24 bridge is considered because of the extensive database of accelerometric data in both standard and damaged conditions. The proposed framework starts from the first four fundamental frequencies extracted through operational modal analysis (OMA) and clustering, followed by density-based time-domain filtering (tracking). The fundamental frequencies extracted are then fed to a dimensionality reduction block implemented through two different approaches: feature selection (intelligent multiplexer) that tries to estimate the most reliable frequencies based on the evaluation of some statistical features (i.e., mean value, variance, kurtosis), and feature extraction (auto-associative neural network (ANN)) that combine the fundamental frequencies to extract new damage sensitive features in a low dimensional feature space. Finally, one class classifier (OCC) algorithms perform anomaly detection, trained with standard condition points, and tested with normal and anomaly ones. In particular, a new anomaly detector strategy is proposed, namely one class classifier neural network two (OCCNN2), which exploit the classification capability of standard classifiers in an anomaly detection problem, finding the standard class (the boundary of the features space in normal operating conditions) through a two-step approach: coarse and fine boundary estimation. The coarse estimation uses classics OCC techniques, while the fine estimation is performed through a feedforward neural network (NN) trained that exploits the boundaries estimated in the coarse step. The detection algorithms vare then compared with known methods based on principal component analysis (PCA), kernel principal component analysis (KPCA), and auto-associative neural network (ANN). In many cases, the proposed solution increases the performance with respect to the standard OCC algorithms in terms of F1 score and accuracy. In particular, by evaluating the correct features, the anomaly can be detected with accuracy and an F1 score greater than 96% with the proposed method.

Keywords: anomaly detection, frequencies selection, modal analysis, neural network, sensor network, structural health monitoring, vibration measurement

Procedia PDF Downloads 91

Authors: Kai Zhang, Xi Jiang

Abstract:

Effect of fuel variabilities on premixed combustion of bio-syngas mixtures is of great importance in bio-syngas utilisation. The uncertainties of concentrations of fuel constituents such as H2, CO and CH4 may lead to unpredictable combustion performances, combustion instabilities and hot spots which may deteriorate and damage the combustion hardware. Numerical modelling and simulations can assist in understanding the behaviour of bio-syngas combustion with pre-defined species concentrations, while the evaluation of variabilities of concentrations is expensive. To be more specific, questions such as ‘what is the burning velocity of bio-syngas at specific equivalence ratio?’ have been answered either experimentally or numerically, while questions such as ‘what is the likelihood of burning velocity when precise concentrations of bio-syngas compositions are unknown, but the concentration ranges are pre-described?’ have not yet been answered. Uncertainty quantification (UQ) methods can be used to tackle such questions and assess the effects of fuel compositions. An efficient probabilistic UQ method based on Polynomial Chaos Expansion (PCE) techniques is employed in this study. The method relies on representing random variables (combustion performances) with orthogonal polynomials such as Legendre or Gaussian polynomials. The constructed PCE via Galerkin Projection provides easy access to global sensitivities such as main, joint and total Sobol indices. In this study, impacts of fuel compositions on combustion (adiabatic flame temperature and laminar flame speed) of bio-syngas fuel mixtures are presented invoking this PCE technique at several equivalence ratios. High-pressure effects on bio-syngas combustion instability are obtained using detailed chemical mechanism - the San Diego Mechanism. Guidance on reducing combustion instability from upstream biomass gasification process is provided by quantifying the significant contributions of composition variations to variance of physicochemical properties of bio-syngas combustion. It was found that flame speed is very sensitive to hydrogen variability in bio-syngas, and reducing hydrogen uncertainty from upstream biomass gasification processes can greatly reduce bio-syngas combustion instability. Variation of methane concentration, although thought to be important, has limited impacts on laminar flame instabilities especially for lean combustion. Further studies on the UQ of percentage concentration of hydrogen in bio-syngas can be conducted to guide the safer use of bio-syngas.

Keywords: bio-syngas combustion, clean energy utilisation, fuel variability, PCE, targeted uncertainty reduction, uncertainty quantification

Procedia PDF Downloads 250

Authors: Natarajan Ramasamy, Gurulingamurthy Haralur, Ramesh Nivarthu, Nikhil Kumar Singha

Abstract:

Elastomers are polymeric materials with varied backbone architectures ranging from linear to dendrimeric structures and wide varieties of monomeric repeat units. These elastomers show strongly viscous and weakly elastic when it is not cross-linked. But when crosslinked, based on the extent the properties of these elastomers can range from highly flexible to highly stiff nature. Lightly cross-linked systems are well studied and reported. Understanding the nature of highly cross-linked rubber based upon chemical structure and architecture is critical for varieties of applications. One of the critical parameters is cross-link density. In the current work, we have studied the highly cross-linked state of linear, lightly branched to star-shaped branched elastomers and determined the cross-linked density by using different models. Change in hardness, shift in Tg, change in modulus and swelling behavior were measured experimentally as a function of the extent of curing. These properties were analyzed using varied models to determine cross-link density. We used hardness measurements to examine cure time. Hardness to the extent of curing relationship is determined. It is well known that micromechanical transitions like Tg and storage modulus are related to the extent of crosslinking. The Tg of the elastomer in different crosslinked state was determined by DMA, and based on plateau modulus the crosslink density is estimated by using Nielsen’s model. Usually for lightly crosslinked systems, based on equilibrium swelling ratio in solvent the cross link density is estimated by using Flory–Rhener model. When it comes to highly crosslinked system, Flory-Rhener model is not valid because of smaller chain length. So models based on the assumption of polymer as a Non-Gaussian chain like 1) Helmis–Heinrich–Straube (HHS) model, 2) Gloria M.gusler and Yoram Cohen Model, 3) Barbara D. Barr-Howell and Nikolaos A. Peppas model is used for estimating crosslink density. In this work, correction factors are determined to the existing models and based upon it structure-property relationship of highly crosslinked elastomers was studied.

Keywords: dynamic mechanical analysis, glass transition temperature, parts per hundred grams of rubber, crosslink density, number of networks per unit volume of elastomer

Procedia PDF Downloads 136

Authors: Shin Yoneda, Mikiya Kato, Kosuke Murate, Kodo Kawase

Abstract:

In recent years, there have been numerous attempts to smuggle narcotic drugs and chemicals by concealing them in international mail. Combatting this requires a non-destructive technique that can identify such illicit substances in mail. Terahertz (THz) waves can pass through a wide variety of materials, and many chemicals show specific frequency-dependent absorption, known as a spectral fingerprint, in the THz range. Therefore, it is reasonable to investigate non-destructive mail inspection techniques that use THz waves. For this reason, in this work, we tried to identify reagents under high attenuation shielding materials using injection-seeded THz-wave parametric generator (is-TPG). Our THz spectroscopic imaging system using is-TPG consisted of two non-linear crystals for emission and detection of THz waves. A micro-chip Nd:YAG laser and a continuous wave tunable external cavity diode laser were used as the pump and seed source, respectively. The pump beam and seed beam were injected to the LiNbO₃ crystal satisfying the noncollinear phase matching condition in order to generate high power THz-wave. The emitted THz wave was irradiated to the sample which was raster scanned by the x-z stage while changing the frequencies, and we obtained multispectral images. Then the transmitted THz wave was focused onto another crystal for detection and up-converted to the near infrared detection beam based on nonlinear optical parametric effects, wherein the detection beam intensity was measured using an infrared pyroelectric detector. It was difficult to identify reagents in a cardboard box because of high noise levels. In this work, we introduce improvements for noise reduction and image clarification, and the intensity of the near infrared detection beam was converted correctly to the intensity of the THz wave. A Gaussian spatial filter is also introduced for a clearer THz image. Through these improvements, we succeeded in identification of reagents hidden in a 42-mm thick cardboard box filled with several obstacles, which attenuate 56 dB at 1.3 THz, by improving analysis methods. Using this system, THz spectroscopic imaging was possible for saccharides and may also be applied to cases where illicit drugs are hidden in the box, and multiple reagents are mixed together. Moreover, THz spectroscopic imaging can be achieved through even thicker obstacles by introducing an NIR detector with higher sensitivity.

Keywords: nondestructive inspection, principal component analysis, terahertz parametric source, THz spectroscopic imaging

Procedia PDF Downloads 148

Authors: Mohd Khairul Bazli Mohd Aziz, Fadhilah Yusof, Zulkifli Yusop, Zalina Mohd Daud, Mohammad Afif Kasno

Abstract:

Recent developed rainfall network design techniques have been discussed and compared by many researchers worldwide due to the demand of acquiring higher levels of accuracy from collected data. In many studies, rain-gauge networks are designed to provide good estimation for areal rainfall and for flood modelling and prediction. In a certain study, even using lumped models for flood forecasting, a proper gauge network can significantly improve the results. Therefore existing rainfall network in Johor must be optimized and redesigned in order to meet the required level of accuracy preset by rainfall data users. The well-known geostatistics method (variance-reduction method) that is combined with simulated annealing was used as an algorithm of optimization in this study to obtain the optimal number and locations of the rain gauges. Rain gauge network structure is not only dependent on the station density; station location also plays an important role in determining whether information is acquired accurately. The existing network of 84 rain gauges in Johor is optimized and redesigned by using rainfall, humidity, solar radiation, temperature and wind speed data during monsoon season (November – February) for the period of 1975 – 2008. Three different semivariogram models which are Spherical, Gaussian and Exponential were used and their performances were also compared in this study. Cross validation technique was applied to compute the errors and the result showed that exponential model is the best semivariogram. It was found that the proposed method was satisfied by a network of 64 rain gauges with the minimum estimated variance and 20 of the existing ones were removed and relocated. An existing network may consist of redundant stations that may make little or no contribution to the network performance for providing quality data. Therefore, two different cases were considered in this study. The first case considered the removed stations that were optimally relocated into new locations to investigate their influence in the calculated estimated variance and the second case explored the possibility to relocate all 84 existing stations into new locations to determine the optimal position. The relocations of the stations in both cases have shown that the new optimal locations have managed to reduce the estimated variance and it has proven that locations played an important role in determining the optimal network.

Keywords: geostatistics, simulated annealing, semivariogram, optimization

Procedia PDF Downloads 273

Authors: Rasha Al-Saedi, Anthony Meijer

Abstract:

The progress of new porous materials has increased rapidly over the past decade for use in applications such as catalysis, gas storage and removal of environmentally unfriendly species due to their high surface area and high thermal stability. In this work, a theoretical study of the zirconium-based metal organic framework (MOFs) were examined in order to determine their potential for gas adsorption of various guest molecules: CO2, N2, CH4 and H2. The zirconium cluster consists of an inner Zr6O4(OH)4 core in which the triangular faces of the Zr6- octahedron are alternatively capped by O and OH groups which bound to nine formate groups and three benzoate groups linkers. General formula is [Zr(μ-O)4(μ-OH)4(HCOO)9((phyO2C)3X))] where X= CH2OH, CH2NH2, CH2CONH2, n(NH2); (n = 1-3). Three types of adsorption sites on the Zr metal center have been studied, named according to capped chemical groups as the ‘−O site’; the H of (μ-OH) site removed and added to (μ-O) site, ‘–OH site’; (μ-OH) site removed, the ‘void site’ where H2O molecule removed; (μ-OH) from one site and H from other (μ-OH) site, in addition to no defect versions. A series of investigations have been performed aiming to address this important issue. First, density functional theory DFT-B3LYP method with 6-311G(d,p) basis set was employed using Gaussian 09 package in order to evaluate the gas adsorption performance of missing-linker defects in zirconium cluster. Next, study the gas adsorption behaviour on different functionalised zirconium clusters. Those functional groups as mentioned above include: amines, alcohol, amide, in comparison with non-substitution clusters. Then, dispersion-corrected density functional theory (DFT-D) calculations were performed to further understand the enhanced gas binding on zirconium clusters. Finally, study the water effect on CO2 and N2 adsorption. The small functionalized Zr clusters were found to result in good CO2 adsorption over N2, CH4, and H2 due to the quadrupole moment of CO2 while N2, CH4 and H2 weakly polar or non-polar. The adsorption efficiency was determined using the dispersion method where the adsorption binding improved as most of the interactions, for example, van der Waals interactions are missing with the conventional DFT method. The calculated gas binding strengths on the no defect site are higher than those on the −O site, −OH site and the void site, this difference is especially notable for CO2. It has been stated that the enhanced affinity of CO2 of no defect versions is most likely due to the electrostatic interactions between the negatively charged O of CO2 and the positively charged H of (μ-OH) metal site. The uptake of the gas molecule does not enhance in presence of water as the latter binds to Zr clusters more strongly than gas species which attributed to the competition on adsorption sites.

Keywords: density functional theory, gas adsorption, metal- organic frameworks, molecular simulation, porous materials, theoretical chemistry

Procedia PDF Downloads 156

Authors: Jaeyoung Lee

Abstract:

Autonomous driving systems require high reliability to provide people with a safe and comfortable driving experience. However, despite the development of a number of vehicle sensors, it is difficult to always provide high perceived performance in driving environments that vary from time to season. The image segmentation method using deep learning, which has recently evolved rapidly, provides high recognition performance in various road environments stably. However, since the system controls a vehicle in real time, a highly complex deep learning network cannot be used due to time and memory constraints. Moreover, efficient networks are optimized for GPU environments, which degrade performance in embedded processor environments equipped simple hardware accelerators. In this paper, a semantic segmentation network, matrix multiplication accelerator network (MMANet), optimized for matrix multiplication accelerator (MMA) on Texas instrument digital signal processors (TI DSP) is proposed to improve the recognition performance of autonomous driving system. The proposed method is designed to maximize the number of layers that can be performed in a limited time to provide reliable driving environment information in real time. First, the number of channels in the activation map is fixed to fit the structure of MMA. By increasing the number of parallel branches, the lack of information caused by fixing the number of channels is resolved. Second, an efficient convolution is selected depending on the size of the activation. Since MMA is a fixed, it may be more efficient for normal convolution than depthwise separable convolution depending on memory access overhead. Thus, a convolution type is decided according to output stride to increase network depth. In addition, memory access time is minimized by processing operations only in L3 cache. Lastly, reliable contexts are extracted using the extended atrous spatial pyramid pooling (ASPP). The suggested method gets stable features from an extended path by increasing the kernel size and accessing consecutive data. In addition, it consists of two ASPPs to obtain high quality contexts using the restored shape without global average pooling paths since the layer uses MMA as a simple adder. To verify the proposed method, an experiment is conducted using perfsim, a timing simulator, and the Cityscapes validation sets. The proposed network can process an image with 640 x 480 resolution for 6.67 ms, so six cameras can be used to identify the surroundings of the vehicle as 20 frame per second (FPS). In addition, it achieves 73.1% mean intersection over union (mIoU) which is the highest recognition rate among embedded networks on the Cityscapes validation set.

Keywords: edge network, embedded network, MMA, matrix multiplication accelerator, semantic segmentation network

Procedia PDF Downloads 96

Authors: Tayo P. Ogundunmade, Adedayo A. Adepoju

Abstract:

Terrorist attacks in liberal democracies bring about a few pessimistic results, for example, sabotaged public support in the governments they target, disturbing the peace of a protected environment underwritten by the state, and a limitation of individuals from adding to the advancement of the country, among others. Hence, seeking for techniques to understand the different factors involved in terrorism and how to deal with those factors in order to completely stop or reduce terrorist activities is the topmost priority of the government in every country. This research aim is to develop an efficient deep learning-based predictive model for the prediction of future terrorist activities in Nigeria, addressing low-quality prediction accuracy problems associated with the existing solution methods. The proposed predictive AI-based model as a counterterrorism tool will be useful by governments and law enforcement agencies to protect the lives of individuals in society and to improve the quality of life in general. A Heterogeneous Bayesian Neural Network (HETBNN) model was derived with Gaussian error normal distribution. Three primary transfer functions (HOTTFs), as well as two derived transfer functions (HETTFs) arising from the convolution of the HOTTFs, are namely; Symmetric Saturated Linear transfer function (SATLINS ), Hyperbolic Tangent transfer function (TANH), Hyperbolic Tangent sigmoid transfer function (TANSIG), Symmetric Saturated Linear and Hyperbolic Tangent transfer function (SATLINS-TANH) and Symmetric Saturated Linear and Hyperbolic Tangent Sigmoid transfer function (SATLINS-TANSIG). Data on the Terrorist activities in Nigeria gathered through questionnaires for the purpose of this study were used. Mean Square Error (MSE), Mean Absolute Error (MAE) and Test Error are the forecast prediction criteria. The results showed that the HETFs performed better in terms of prediction and factors associated with terrorist activities in Nigeria were determined. The proposed predictive deep learning-based model will be useful to governments and law enforcement agencies as an effective counterterrorism mechanism to understand the parameters of terrorism and to design strategies to deal with terrorism before an incident actually happens and potentially causes the loss of precious lives. The proposed predictive AI-based model will reduce the chances of terrorist activities and is particularly helpful for security agencies to predict future terrorist activities.

Keywords: activation functions, Bayesian neural network, mean square error, test error, terrorism

Procedia PDF Downloads 135

Authors: Shahram Naghizadeh Raeisi, Ali Alghooneh, Seyed Jalal Razavi Zahedkolaei

Abstract:

Herein, a shelf stable (no refrigeration required) UHT processed, aseptically packaged whey protein drink was formulated by using a new strategy in microparticulate process. Applying thermal and two-dimensional mechanical treatments simultaneously, a modified protein (MWPC-80) was produced. Then the physical, thermal and thermodynamic properties of MWPC-80 were assessed using particle size analysis, dynamic temperature sweep (DTS), and differential scanning calorimetric (DSC) tests. Finally, using MWPC-80, a new RTD beverage was formulated, and shelf stability was assessed for three months at ambient temperature (25 °C). Non-isothermal dynamic temperature sweep was performed, and the results were analyzed by a combination of classic rate equation, Arrhenius equation, and time-temperature relationship. Generally, results showed that temperature dependency of the modified sample was significantly (Pvalue<0.05) less than the control one contained WPC-80. The changes in elastic modulus of the MWPC did not show any critical point at all the processed stages, whereas, the control sample showed two critical points during heating (82.5 °C) and cooling (71.10 °C) stages. Thermal properties of samples (WPC-80 & MWPC-80) were assessed using DSC with 4 °C /min heating speed at 20-90 °C heating range. Results did not show any thermal peak in MWPC DSC curve, which suggested high thermal resistance. On the other hands, WPC-80 sample showed a significant thermal peak with thermodynamic properties of ∆G:942.52 Kj/mol ∆H:857.04 Kj/mole and ∆S:-1.22Kj/mole°K. Dynamic light scattering was performed and results showed 0.7 µm and 15 nm average particle size for MWPC-80 and WPC-80 samples, respectively. Moreover, particle size distribution of MWPC-80 and WPC-80 were Gaussian-Lutresian and normal, respectively. After verification of microparticulation process by DTS, PSD and DSC analyses, a 10% why protein beverage (10% w/w/ MWPC-80, 0.6% w/w vanilla flavoring agent, 0.1% masking flavor, 0.05% stevia natural sweetener and 0.25% citrate buffer) was formulated and UHT treatment was performed at 137 °C and 4 s. Shelf life study did not show any jellification or precipitation of MWPC-80 contained beverage during three months storage at ambient temperature, whereas, WPC-80 contained beverage showed significant precipitation and jellification after thermal processing, even at 3% w/w concentration. Consumer knowledge on nutritional advantages of whey protein increased the request for using this protein in different food systems especially RTD beverages. These results could make a huge difference in this industry.

Keywords: high protein whey beverage, micropartiqulation, two-dimentional mechanical treatments, thermodynamic properties

Procedia PDF Downloads 40

Authors: Soyon Kim, Edward Kim

Abstract:

In order to retrieve information from the massive stream of songs in the music industry, music search by title, lyrics, artist, mood, and genre has become more important. Despite the subjectivity and controversy over the definition of music genres across different nations and cultures, automatic genre classification systems that facilitate the process of music categorization have been developed. Manual genre selection by music producers is being provided as statistical data for designing automatic genre classification systems. In this paper, an automatic music genre classification system utilizing non-negative matrix factorization (NMF) is proposed. Short-term characteristics of the music signal can be captured based on the timbre features such as mel-frequency cepstral coefficient (MFCC), decorrelated filter bank (DFB), octave-based spectral contrast (OSC), and octave band sum (OBS). Long-term time-varying characteristics of the music signal can be summarized with (1) the statistical features such as mean, variance, minimum, and maximum of the timbre features and (2) the modulation spectrum features such as spectral flatness measure, spectral crest measure, spectral peak, spectral valley, and spectral contrast of the timbre features. Not only these conventional basic long-term feature vectors, but also NMF based feature vectors are proposed to be used together for genre classification. In the training stage, NMF basis vectors were extracted for each genre class. The NMF features were calculated in the log spectral magnitude domain (NMF-LSM) as well as in the basic feature vector domain (NMF-BFV). For NMF-LSM, an entire full band spectrum was used. However, for NMF-BFV, only low band spectrum was used since high frequency modulation spectrum of the basic feature vectors did not contain important information for genre classification. In the test stage, using the set of pre-trained NMF basis vectors, the genre classification system extracted the NMF weighting values of each genre as the NMF feature vectors. A support vector machine (SVM) was used as a classifier. The GTZAN multi-genre music database was used for training and testing. It is composed of 10 genres and 100 songs for each genre. To increase the reliability of the experiments, 10-fold cross validation was used. For a given input song, an extracted NMF-LSM feature vector was composed of 10 weighting values that corresponded to the classification probabilities for 10 genres. An NMF-BFV feature vector also had a dimensionality of 10. Combined with the basic long-term features such as statistical features and modulation spectrum features, the NMF features provided the increased accuracy with a slight increase in feature dimensionality. The conventional basic features by themselves yielded 84.0% accuracy, but the basic features with NMF-LSM and NMF-BFV provided 85.1% and 84.2% accuracy, respectively. The basic features required dimensionality of 460, but NMF-LSM and NMF-BFV required dimensionalities of 10 and 10, respectively. Combining the basic features, NMF-LSM and NMF-BFV together with the SVM with a radial basis function (RBF) kernel produced the significantly higher classification accuracy of 88.3% with a feature dimensionality of 480.

Keywords: mel-frequency cepstral coefficient (MFCC), music genre classification, non-negative matrix factorization (NMF), support vector machine (SVM)

Procedia PDF Downloads 265

Authors: Man Fung Ho, Lap So, Jiaqi Zhang, Yuheng Zhao, Huiyang Lu, Tat Shing Choi, K. Y. Michael Wong

Abstract:

Nowadays, a tremendous amount of data is available in the transportation system, enabling the development of various machine learning approaches to make short-term latency predictions. A natural question is then the choice of relevant information to enable accurate predictions. Using traffic data collected from the Taiwan Freeway System, we consider the prediction of short-term latency of a freeway segment with a length of 17 km covering 5 measurement points, each collecting vehicle-by-vehicle data through the electronic toll collection system. The processed data include the past latencies of the freeway segment with different time lags, the traffic conditions of the individual segments (the accumulations, the traffic fluxes, the entrance and exit rates), the total accumulations, and the weekday latency profiles obtained by Gaussian process regression of past data. We arrive at several important conclusions about how data should be refined to obtain accurate predictions, which have implications for future system-wide latency predictions. (1) We find that the prediction of median latency is much more accurate and meaningful than the prediction of average latency, as the latter is plagued by outliers. This is verified by machine-learning prediction using XGBoost that yields a 35% improvement in the mean square error of the 5-minute averaged latencies. (2) We find that the median latency of the segment 15 minutes ago is a very good baseline for performance comparison, and we have evidence that further improvement is achieved by machine learning approaches such as XGBoost and Long Short-Term Memory (LSTM). (3) By analyzing the feature importance score in XGBoost and calculating the mutual information between the inputs and the latencies to be predicted, we identify a sequence of inputs ranked in importance. It confirms that the past latencies are most informative of the predicted latencies, followed by the total accumulation, whereas inputs such as the entrance and exit rates are uninformative. It also confirms that the inputs are much less informative of the average latencies than the median latencies. (4) For predicting the latencies of segments composed of two or three sub-segments, summing up the predicted latencies of each sub-segment is more accurate than the one-step prediction of the whole segment, especially with the latency prediction of the downstream sub-segments trained to anticipate latencies several minutes ahead. The duration of the anticipation time is an increasing function of the traveling time of the upstream segment. The above findings have important implications to predicting the full set of latencies among the various locations in the freeway system.

Keywords: data refinement, machine learning, mutual information, short-term latency prediction

Procedia PDF Downloads 144

Authors: Weikang Gong, Chunhua Li

Abstract:

Dynamics plays an essential role in function exertion of proteins. Elastic network model (ENM), a harmonic potential-based and cost-effective computational method, is a valuable and efficient tool for characterizing the intrinsic dynamical properties encoded in biomacromolecule structures and has been widely used to detect the large-amplitude collective motions of proteins. Gaussian network model (GNM) and anisotropic network model (ANM) are the two often-used ENM models. In recent years, many ENM variants have been proposed. Here, we propose a small but effective modification (denoted as modified mENM) to the multiscale ENM (mENM) where fitting weights of Kirchhoff/Hessian matrixes with the least square method (LSM) is modified since it neglects the details of pairwise interactions. Then we perform its comparisons with the original mENM, traditional ENM, and parameter-free ENM (pfENM) on reproducing dynamical properties for the six representative proteins whose molecular dynamics (MD) trajectories are available in http://mmb.pcb.ub.es/MoDEL/. In the results, for B-factor prediction, mENM achieves the best performance among the four ENM models. Additionally, it is noted that with the weights of the multiscale Kirchhoff/Hessian matrixes modified, interestingly, the modified mGNM/mANM still has a much better performance than the corresponding traditional ENM and pfENM models. As to dynamical cross-correlation map (DCCM) calculation, taking the data obtained from MD trajectories as the standard, mENM performs the worst while the results produced by the modified mENM and pfENM models are close to those from MD trajectories with the latter a little better than the former. Generally, ANMs perform better than the corresponding GNMs except for the mENM. Thus, pfANM and the modified mANM, especially the former, have an excellent performance in dynamical cross-correlation calculation. Compared with GNMs (except for mGNM), the corresponding ANMs can capture quite a number of positive correlations for the residue pairs nearly largest distances apart, which is maybe due to the anisotropy consideration in ANMs. Furtherly, encouragingly the modified mANM displays the best performance in capturing the functional motional modes, followed by pfANM and traditional ANM models, while mANM fails in all the cases. This suggests that the consideration of long-range interactions is critical for ANM models to produce protein functional motions. Based on the analyses, the modified mENM is a promising method in capturing multiple dynamical characteristics encoded in protein structures. This work is helpful for strengthening the understanding of the elastic network model and provides a valuable guide for researchers to utilize the model to explore protein dynamics.

Keywords: elastic network model, ENM, multiscale ENM, molecular dynamics, parameter-free ENM, protein structure

Procedia PDF Downloads 100

Authors: Ning Chang, Zelong Yuan, Yunpeng Wang, Jianchun Wang

Abstract:

The utilization of Large Eddy Simulation (LES) has been extensive in turbulence research. LES concentrates on resolving the significant grid-scale motions while representing smaller scales through subfilter-scale (SFS) models. The deconvolution model, among the available SFS models, has proven successful in LES of engineering and geophysical flows. Nevertheless, the thorough investigation of how sub-filter scale dynamics and filter anisotropy affect SFS modeling accuracy remains lacking. The outcomes of LES are significantly influenced by filter selection and grid anisotropy, factors that have not been adequately addressed in earlier studies. This study examines two crucial aspects of LES: Firstly, the accuracy of direct deconvolution models (DDM) is evaluated concerning sub-filter scale (SFS) dynamics across varying filter-to-grid ratios (FGR) in isotropic turbulence. Various invertible filters are employed, including Gaussian, Helmholtz I and II, Butterworth, Chebyshev I and II, Cauchy, Pao, and rapidly decaying filters. The importance of FGR becomes evident as it plays a critical role in controlling errors for precise SFS stress prediction. When FGR is set to 1, the DDM models struggle to faithfully reconstruct SFS stress due to inadequate resolution of SFS dynamics. Notably, prediction accuracy improves when FGR is set to 2, leading to accurate reconstruction of SFS stress, except for cases involving Helmholtz I and II filters. Remarkably high precision, nearly 100%, is achieved at an FGR of 4 for all DDM models. Furthermore, the study extends to filter anisotropy and its impact on SFS dynamics and LES accuracy. By utilizing the dynamic Smagorinsky model (DSM), dynamic mixed model (DMM), and direct deconvolution model (DDM) with anisotropic filters, aspect ratios (AR) ranging from 1 to 16 are examined in LES filters. The results emphasize the DDM’s proficiency in accurately predicting SFS stresses under highly anisotropic filtering conditions. Notably high correlation coefficients exceeding 90% are observed in the a priori study for the DDM’s reconstructed SFS stresses, surpassing those of the DSM and DMM models. However, these correlations tend to decrease as filter anisotropy increases. In the a posteriori analysis, the DDM model consistently outperforms the DSM and DMM models across various turbulence statistics, including velocity spectra, probability density functions related to vorticity, SFS energy flux, velocity increments, strainrate tensors, and SFS stress. It is evident that as filter anisotropy intensifies, the results of DSM and DMM deteriorate, while the DDM consistently delivers satisfactory outcomes across all filter-anisotropy scenarios. These findings underscore the potential of the DDM framework as a valuable tool for advancing the development of sophisticated SFS models for LES in turbulence research.

Keywords: deconvolution model, large eddy simulation, subfilter scale modeling, turbulence

Procedia PDF Downloads 42

Authors: A. Chatel, I. S. Torreguitart, T. Verstraete

Abstract:

The paper deals with a single point optimization of the LS89 turbine using an adjoint optimization and defining the design variables within a CAD system. The advantage of including the CAD model in the design system is that higher level constraints can be imposed on the shape, allowing the optimized model or component to be manufactured. However, CAD-based approaches restrict the design space compared to node-based approaches where every node is free to move. In order to preserve a rich design space, we develop a methodology to refine the CAD model during the optimization and to create the best parameterization to use at each time. This study presents a methodology to progressively refine the design space, which combines parametric effectiveness with a differential evolutionary algorithm in order to create an optimal parameterization. In this manuscript, we show that by doing the parameterization at the CAD level, we can impose higher level constraints on the shape, such as the axial chord length, the trailing edge radius and G2 geometric continuity between the suction side and pressure side at the leading edge. Additionally, the adjoint sensitivities are filtered out and only smooth shapes are produced during the optimization process. The use of algorithmic differentiation for the CAD kernel and grid generator allows computing the grid sensitivities to machine accuracy and avoid the limited arithmetic precision and the truncation error of finite differences. Then, the parametric effectiveness is computed to rate the ability of a set of CAD design parameters to produce the design shape change dictated by the adjoint sensitivities. During the optimization process, the design space is progressively enlarged using the knot insertion algorithm which allows introducing new control points whilst preserving the initial shape. The position of the inserted knots is generally assumed. However, this assumption can hinder the creation of better parameterizations that would allow producing more localized shape changes where the adjoint sensitivities dictate. To address this, we propose using a differential evolutionary algorithm to maximize the parametric effectiveness by optimizing the location of the inserted knots. This allows the optimizer to gradually explore larger design spaces and to use an optimal CAD-based parameterization during the course of the optimization. The method is tested on the LS89 turbine cascade and large aerodynamic improvements in the entropy generation are achieved whilst keeping the exit flow angle fixed. The trailing edge and axial chord length, which are kept fixed as manufacturing constraints. The optimization results show that the multilevel optimizations were more efficient than the single level optimization, even though they used the same number of design variables at the end of the multilevel optimizations. Furthermore, the multilevel optimization where the parameterization is created using the optimal knot positions results in a more efficient strategy to reach a better optimum than the multilevel optimization where the position of the knots is arbitrarily assumed.

Keywords: adjoint, CAD, knots, multilevel, optimization, parametric effectiveness

Procedia PDF Downloads 89