Search results for: block implicit QR algorithm

Authors: Tong He, Long Chen, Irag Mantegh, Wen-Fang Xie

Abstract:

This paper proposes an image-based obstacle avoidance and tracking target identification strategy in GPS-degraded or GPS-denied environment for an Unmanned Aerial Vehicle (UAV). The traditional force algorithm for obstacle avoidance could produce local minima area, in which UAV cannot get away obstacle effectively. In order to eliminate it, an artificial potential approach based on harmonic potential is proposed to guide the UAV to avoid the obstacle by using the vision system. And image-based visual servoing scheme (IBVS) has been adopted to implement the proposed obstacle avoidance approach. In IBVS, the pixel accuracy is a key factor to realize the obstacle avoidance. In this paper, the deep reinforcement learning framework has been applied by reducing pixel errors through constant interaction between the environment and the agent. In addition, the combination of OpenTLD and Tensorflow based on neural network is used to identify the type of tracking target. Numerical simulation in Matlab and ROS GAZEBO show the satisfactory result in target identification and obstacle avoidance.

Keywords: image-based visual servoing, obstacle avoidance, tracking target identification, deep reinforcement learning, artificial potential approach, neural network

Procedia PDF Downloads 130

Authors: Partha P. Gopmandal, Somnath Bhattacharyya, Naren Bag

Abstract:

In this article, we have studied the effect of pH-regulated surface charge on the electroosmotic flow (EOF) through nanochannel filled with binary symmetric electrolyte solution. The channel wall possesses either an acidic or a basic functional group. Going beyond the widely employed Debye-Huckel linearization, we develop a mathematical model based on Nernst-Planck equation for the charged species, Poisson equation for the induced potential, Stokes equation for fluid flow. A finite volume based numerical algorithm is adopted to study the effect of key parameters on the EOF. We have computed the coupled governing equations through the finite volume method and our results found to be in good agreement with the analytical solution obtained from the corresponding linear model based on low surface charge condition or strong electrolyte solution. The influence of the surface charge density, reaction constant of the functional groups, bulk pH, and concentration of the electrolyte solution on the overall flow rate is studied extensively. We find the effect of surface charge diminishes with the increase in electrolyte concentration. In addition for strong electrolyte, the surface charge becomes independent of pH due to complete dissociation of the functional groups.

Keywords: electroosmosis, finite volume method, functional group, surface charge

Procedia PDF Downloads 408

Authors: Avyarthana Dey, Naren P. Rao, Arpitha Jacob, Chaitra V. Hiremath, Shivarama Varambally, Ganesan Venkatasubramanian, Rose Dawn Bharath, Bangalore N. Gangadhar

Abstract:

Objective: Impaired facial emotion recognition is widely reported in schizophrenia. Neuropeptide oxytocin is known to modulate brain regions involved in facial emotion recognition, namely amygdala, in healthy volunteers. However, its effect on facial emotion recognition deficits seen in schizophrenia is not well explored. In this study, we examined the effect of intranasal OXT on processing facial emotions and its neural correlates in patients with schizophrenia. Method: 12 male patients (age= 31.08±7.61 years, education= 14.50±2.20 years) participated in this single-blind, counterbalanced functional magnetic resonance imaging (fMRI) study. All participants underwent three fMRI scans; one at baseline, one each after single dose 24IU intranasal OXT and intranasal placebo. The order of administration of OXT and placebo were counterbalanced and subject was blind to the drug administered. Participants performed a facial emotion recognition task presented in a block design with six alternating blocks of faces and shapes. The faces depicted happy, angry or fearful emotions. The images were preprocessed and analyzed using SPM 12. First level contrasts comparing recognition of emotions and shapes were modelled at individual subject level. A group level analysis was performed using the contrasts generated at the first level to compare the effects of intranasal OXT and placebo. The results were thresholded at uncorrected p < 0.001 with a cluster size of 6 voxels. Neuropeptide oxytocin is known to modulate brain regions involved in facial emotion recognition, namely amygdala, in healthy volunteers. Results: Compared to placebo, intranasal OXT attenuated activity in inferior temporal, fusiform and parahippocampal gyri (BA 20), premotor cortex (BA 6), middle frontal gyrus (BA 10) and anterior cingulate gyrus (BA 24) and enhanced activity in the middle occipital gyrus (BA 18), inferior occipital gyrus (BA 19), and superior temporal gyrus (BA 22). There were no significant differences between the conditions on the accuracy scores of emotion recognition between baseline (77.3±18.38), oxytocin (82.63 ± 10.92) or Placebo (76.62 ± 22.67). Conclusion: Our results provide further evidence to the modulatory effect of oxytocin in patients with schizophrenia. Single dose oxytocin resulted in significant changes in activity of brain regions involved in emotion processing. Future studies need to examine the effectiveness of long-term treatment with OXT for emotion recognition deficits in patients with schizophrenia.

Keywords: recognition, functional connectivity, oxytocin, schizophrenia, social cognition

Procedia PDF Downloads 207

Authors: Ariful Islam, Showkat Ahmad Lone

Abstract:

Statistical distributions have long been employed in the assessment of semiconductor devices and product reliability. The power function-distribution is one of the most important distributions in the modern reliability practice and can be frequently preferred over mathematically more complex distributions, such as the Weibull and the lognormal, because of its simplicity. Moreover, it may exhibit a better fit for failure data and provide more appropriate information about reliability and hazard rates in some circumstances. This study deals with estimating information about failure times of items under step-stress partially accelerated life tests for competing risk based on adoptive type-I progressive hybrid censoring criteria. The life data of the units under test is assumed to follow Mukherjee-Islam distribution. The point and interval maximum-likelihood estimations are obtained for distribution parameters and tampering coefficient. The performances of the resulting estimators of the developed model parameters are evaluated and investigated by using a simulation algorithm.

Keywords: adoptive progressive hybrid censoring, competing risk, mukherjee-islam distribution, partially accelerated life testing, simulation study

Procedia PDF Downloads 338

Authors: Violeta Damjanovic-Behrendt

Abstract:

This paper presents an approach for optimal cyber security decisions to protect instances of a federated Internet of Things (IoT) platform in the cloud. The presented solution implements the repeated Stackelberg Security Game (SSG) and a model called Stochastic Human behaviour model with AttRactiveness and Probability weighting (SHARP). SHARP employs the Subjective Utility Quantal Response (SUQR) for formulating a subjective utility function, which is based on the evaluations of alternative solutions during decision-making. We augment the repeated SSG (including SHARP and SUQR) with a reinforced learning algorithm called Naïve Q-Learning. Naïve Q-Learning belongs to the category of active and model-free Machine Learning (ML) techniques in which the agent (either the defender or the attacker) attempts to find an optimal security solution. In this way, we combine GT and ML algorithms for discovering optimal cyber security policies. The proposed security optimization components will be validated in a collaborative cloud platform that is based on the Industrial Internet Reference Architecture (IIRA) and its recently published security model.

Keywords: security, internet of things, cloud computing, stackelberg game, machine learning, naive q-learning

Procedia PDF Downloads 343

Authors: Taufik Rizkiandi, Teuku Yuri M. Zagloel, Andri Dwi Setiawan

Abstract:

Distribution and growth in the transportation and warehousing business sector decreased by 15,04%. There was a decrease in Gross Domestic Product (GDP) contribution level from rank 7 of 4,41% in 2019 to 3,81% in rank 8 in 2020. A decline in the transportation and warehousing business sector contributes to GDP, resulting in oil and gas companies implementing an efficient supply chain strategy to ensure the availability of goods, especially lubricants. Fluctuating demand for lubricants and warehouse service time limits are essential things that are taken into account in determining an efficient route. Add depots points as a solution so that demand for lubricants is fulfilled (not stock out). However, adding a depot will increase operating costs and storage costs. Therefore, it is necessary to optimize the addition of depots using the Capacitated Vehicle Routing Problem with Time Windows (CVRPTW). This research case study was conducted at an oil and gas company that produces lubricants from 2019 to 2021. The study results obtained the optimal route and the addition of a depot with a minimum additional cost. The total cost remains efficient with the addition of a depot when compared to one depot from Jakarta.

Keywords: CVRPTW, optimal route, depot, tabu search algorithm

Procedia PDF Downloads 126

Authors: Andrew R. Winters, Gregor J. Gassner

Abstract:

A high-order numerical magentohydrodynamics (MHD) solver built upon a non-linear entropy stable numerical flux function that supports eight traveling wave solutions will be described. The method is designed to treat the divergence-free constraint on the magnetic field in a similar fashion to a hyperbolic divergence cleaning technique. The solver is especially well-suited for flows involving strong discontinuities due to its strong stability without the need to enforce artificial low density or energy limits. Furthermore, a new formulation of the numerical algorithm to guarantee positivity of the pressure during the simulation is described and presented. By construction, the solver conserves mass, momentum, and energy and is entropy stable. High spatial order is obtained through the use of a third order limiting technique. High temporal order is achieved by utilizing the family of strong stability preserving (SSP) Runge-Kutta methods. Main attributes of the solver are presented as well as details on an implementation of the new solver into the multi-physics, multi-scale simulation code FLASH. The accuracy, robustness, and computational efficiency is demonstrated with a variety of numerical tests. Comparisons are also made between the new solver and existing methods already present in FLASH framework.

Keywords: entropy stability, finite volume scheme, magnetohydrodynamics, pressure positivity

Procedia PDF Downloads 333

Authors: Seyyed Masoud Seyyedoshohadaei, Borhanuddin Mohd Ali

Abstract:

To support group mobility, the NEMO Basic Support Protocol has been standardized as an extension of Mobile IP that enables an entire network to change its point of attachment to the Internet. Using NEMO in IEEE 802.16e (WiMax) networks causes latency in handover procedure and affects seamless communication of real-time applications. To decrease handover latency and service disruption time, an integrated scheme named Optimized Fast NEMO (OFNEMO) was introduced by authors of this paper. In OFNEMO a pre-establish multi tunnels concept, cross function optimization and cross layer design are used. In this paper, an analytical model is developed to evaluate total cost consisting of signaling and packet delivery costs of the OFNEMO compared with RFC3963. Results show that OFNEMO increases probability of predictive mode compared with RFC3963 due to smaller handover latency. Even though OFNEMO needs extra signalling to pre-establish multi tunnel, it has less total cost thanks to its optimized algorithm. OFNEMO can minimize handover latency for supporting real time application in moving networks.

Keywords: fast mobile IPv6, handover latency, IEEE802.16e, network mobility

Procedia PDF Downloads 182

Authors: Zhe Li, Tao Ju, Liguo Zhang, Zehong Zhang, Baoshun Zhang

Abstract:

In-grow stacking-faults (IGSFs) in 4H-SiC epilayers can cause increased leakage current and reduce the blocking voltage of 4H-SiC power devices. Double Shockley stacking fault (2SSF) is a common type of IGSF with double slips on the basal planes. In this study, a 2SSF in the 4H-SiC epilayer grown by chemical vaper deposition (CVD) is characterized. The nucleation site of the 2SSF is discussed, and a model for the 2SSF nucleation is proposed. Homo-epitaxial 4H-SiC is grown on a commercial 4 degrees off-cut substrate by a home-built hot-wall CVD. Defect-selected-etching (DSE) is conducted with melted KOH at 500 degrees Celsius for 1-2 min. Room temperature cathodoluminescence (CL) is conducted at a 20 kV acceleration voltage. Low-temperature photoluminescence (LTPL) is conducted at 3.6 K with the 325 nm He-Cd laser line. In the CL image, a triangular area with bright contrast is observed. Two partial dislocations (PDs) with a 20-degree angle in between show linear dark contrast on the edges of the IGSF. CL and LTPL spectrums are conducted to verify the IGSF’s type. The CL spectrum shows the maximum photoemission at 2.431 eV and negligible bandgap emission. In the LTPL spectrum, four phonon replicas are found at 2.468 eV, 2.438 eV, 2.420 eV and 2.410 eV, respectively. The Egx is estimated to be 2.512 eV. A shoulder with a red-shift to the main peak in CL, and a slight protrude at the same wavelength in LTPL are verified as the so called Egx- lines. Based on the CL and LTPL results, the IGSF is identified as a 2SSF. Back etching by neutral loop discharge and DSE are conducted to track the origin of the 2SSF, and the nucleation site is found to be a threading screw dislocation (TSD) in this sample. A nucleation mechanism model is proposed for the formation of the 2SSF. Steps introduced by the off-cut and the TSD on the surface are both suggested to be two C-Si bilayers height. The intersections of such two types of steps are along [11-20] direction from the TSD, while a four-bilayer step at each intersection. The nucleation of the 2SSF in the growth is proposed as follows. Firstly, the upper two bilayers of the four-bilayer step grow down and block the lower two at one intersection, and an IGSF is generated. Secondly, the step-flow grows over the IGSF successively, and forms an AC/ABCABC/BA/BC stacking sequence. Then a 2SSF is formed and extends by the step-flow growth. In conclusion, a triangular IGSF is characterized by CL approach. Base on the CL and LTPL spectrums, the estimated Egx is 2.512 eV and the IGSF is identified to be a 2SSF. By back etching, the 2SSF nucleation site is found to be a TSD. A model for the 2SSF nucleation from an intersection of off-cut- and TSD- introduced steps is proposed.

Keywords: cathodoluminescence, defect-selected-etching, double Shockley stacking fault, low-temperature photoluminescence, nucleation model, silicon carbide

Procedia PDF Downloads 303

Authors: Cameron Hamilton, Walter Potter, Gerrit Hoogenboom, Ronald McClendon, Will Hobbs

Abstract:

A model was constructed to predict the amount of solar radiation that will make contact with the surface of the earth in a given location an hour into the future. This project was supported by the Southern Company to determine at what specific times during a given day of the year solar panels could be relied upon to produce energy in sufficient quantities. Due to their ability as universal function approximators, an artificial neural network was used to estimate the nonlinear pattern of solar radiation, which utilized measurements of weather conditions collected at the Griffin, Georgia weather station as inputs. A number of network configurations and training strategies were utilized, though a multilayer perceptron with a variety of hidden nodes trained with the resilient propagation algorithm consistently yielded the most accurate predictions. In addition, a modeled DNI field and adjacent weather station data were used to bolster prediction accuracy. In later trials, the solar radiation field was preprocessed with a discrete wavelet transform with the aim of removing noise from the measurements. The current model provides predictions of solar radiation with a mean square error of 0.0042, though ongoing efforts are being made to further improve the model’s accuracy.

Keywords: artificial neural networks, resilient propagation, solar radiation, time series forecasting

Procedia PDF Downloads 373

Authors: Kritiyaporn Kunsook

Abstract:

Machine learning algorithms (MLAs) such us artificial neural networks (ANNs), decision tree, support vector machines (SVMs), Naïve Bayes, and ensemble classifier by voting are powerful data driven methods that are relatively less widely used in the mapping of technique of system, and thus have not been comparatively evaluated together thoroughly in this field. The performances of a series of MLAs, ANNs, decision tree, SVMs, Naïve Bayes, and ensemble classifier by voting in technique of hydroponic systems prospectively modeling are compared based on the accuracy of each model. Classification of hydroponic systems only covers the test samples from vegetables grown with Nutrient film technique (NFT) and Deep flow technique (DFT). The feature, which are the characteristics of vegetables compose harvesting height width, temperature, require light and color. The results indicate that the classification performance of the ANNs is 98%, decision tree is 98%, SVMs is 97.33%, Naïve Bayes is 96.67%, and ensemble classifier by voting is 98.96% algorithm respectively.

Keywords: artificial neural networks, decision tree, support vector machines, naïve Bayes, ensemble classifier by voting

Procedia PDF Downloads 351

Authors: Mohammadreza Heydariazad

Abstract:

Productivity growth of wind energy as a clean source needed to achieve improved strategy in production and transmission and management of wind resources in order to increase quality of power and reduce costs. New technologies based on power converters that cause changing turbine speed to suit the wind speed blowing turbine improve extraction efficiency power from wind. This article introduces variable speed wind turbines and optimization of power, and presented methods to use superconducting inductor in the composition of power converter and is proposed the dc measurement for the wind farm and especially is considered techniques available to them. In fact, this article reviews mechanisms and function, changes of wind speed turbine according to speed control strategies of various types of wind turbines and examines power possible transmission and ac from producing location to suitable location for a strong connection integrating wind farm generators, without additional cost or equipment. It also covers main objectives of the dynamic control of wind turbines, and the methods of exploitation and the ways of using it that includes the unique process of these components. Effective algorithm is presented for power control in order to extract maximum active power and maintains power factor at the desired value.

Keywords: wind energy, generator, superconducting inductor, wind turbine power

Procedia PDF Downloads 317

Authors: Noureddine Ben Ayed, Abdelkader Soumaya, Saïd Maouche, Ali Kadri, Mongi Gueddiche, Hayet Khayati-Ammar, Ahmed Braham

Abstract:

The archaeological monument of Utica, located in north-eastern Tunisia, was founded (8th century BC) By the Phoenicians as a port installed on the trade route connecting Phoenicia and the Straits of Gibraltar in the Mediterranean Sea. The flourishment of this city as an important settlement during the Roman period was followed by a sudden abandonment, disuse and progressive oblivion in the first half of the fifth century AD. This decadence can be attributed to the destructive earthquake of 5 July 408 - 410 AD, affecting this historic city as documented in 1906 by the seismologist Fernand De Montessus De Ballore. The magnitude of the Utica earthquake was estimated at 6.8 by the Tunisian National Institute of Meteorology (INM). In order to highlight the damage caused by this earthquake, a field survey was carried out at the Utica ruins to detect and analyse the earthquake archaeological effects (EAEs) using structural geology methods. This approach allowed us to highlight several structural damages, including: (1) folded mortar pavements, (2) cracks affecting the mosaic and walls of a water basin in the "House of the Grand Oecus", (3) displaced columns, (4) block extrusion in masonry walls, (5) undulations in mosaic pavements, (6) tilted walls. The structural analysis of these EAEs and data measurements reveal a seismic cause for all evidence of deformation in the Utica monument. The maximum horizontal strain of the ground (e.g. SHmax) inferred from the building oriented damage in Utica shows a NNW-SSE direction under a compressive tectonic regime. For the seismogenic source of this earthquake, we propose the active E-W to NE-SW trending Utique - Ghar El Melh reverse fault, passing through the Utica Monument and extending towards the Ghar El Melh Lake, as the causative tectonic structure. The active fault trace is well supported by instrumental seismicity, geophysical data (e.g., gravity, seismic profiles) and geomorphological analyses. In summary, we find that the archaeoseismic records detected at Utica are similar to those observed at many other archaeological sites affected by destructive ancient earthquakes around the world. Furthermore, the calculated orientation of the average maximum horizontal stress (SHmax) closely match the state of the actual stress field, as highlighted by some earthquake focal mechanisms in this region.

Keywords: Tunisia, utica, seimogenic fault, archaeological earthquake effects

Procedia PDF Downloads 29

Authors: Nuri Caglayan, H. Kursat Celik

Abstract:

There are many factors that influence the health and productivity of the animals in livestock production fields, including temperature, humidity, carbon dioxide (CO2), ammonia (NH3), hydrogen sulfide (H2S), physical activity and particulate matter. High NH3 concentrations reduce feed consumption and cause daily weight gain. At high concentrations, H2S causes respiratory problems and CO2 displace oxygen, which can cause suffocation or asphyxiation. Good air quality in livestock facilities can have an impact on the health and well-being of animals and humans. Air quality assessment basically depends on strictly given limits without taking into account specific local conditions between harmful gases and other meteorological factors. The stated limitations may be eliminated. using controlling systems based on neural networks and fuzzy logic. This paper describes a fuzzy logic based ventilation algorithm, which can calculate different fan speeds under pre-defined boundary conditions, for removing harmful gases from the production environment. In the paper, a fuzzy logic model has been developed based on a Mamedani’s fuzzy method. The model has been built on MATLAB software. As the result, optimum fan speeds under pre-defined boundary conditions have been presented.

Keywords: air quality, fuzzy logic model, livestock housing, fan speed

Procedia PDF Downloads 359

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 113

Authors: Rizwan Rizwan

Abstract:

This paper provides an overview of Federated Learning (FL) and its application in enhancing data security, privacy, and efficiency. FL utilizes three distinct architectures to ensure privacy is never compromised. It involves training individual edge devices and aggregating their models on a server without sharing raw data. This approach not only provides secure models without data sharing but also offers a highly efficient privacy--preserving solution with improved security and data access. Also we discusses various frameworks used in FL and its integration with machine learning, deep learning, and data mining. In order to address the challenges of multi--party collaborative modeling scenarios, a brief review FL scheme combined with an adaptive gradient descent strategy and differential privacy mechanism. The adaptive learning rate algorithm adjusts the gradient descent process to avoid issues such as model overfitting and fluctuations, thereby enhancing modeling efficiency and performance in multi-party computation scenarios. Additionally, to cater to ultra-large-scale distributed secure computing, the research introduces a differential privacy mechanism that defends against various background knowledge attacks.

Keywords: federated learning, differential privacy, gradient descent strategy, convergence, stability, threats

Procedia PDF Downloads 12

Authors: Erica Elice Uy, Toshihiro Noda, Kentaro Nakai, Jonathan Dungca

Abstract:

Deformation in a conventional cyclic triaxial test is normally measured by using point-wise measuring device. In this study, non-contact measurement technique was applied to be able to monitor and measure the occurrence of non-homogeneous behavior of the soil under cyclic loading. Non-contact measurement is executed through image processing. Two-dimensional measurements were performed using Lucas and Kanade optical flow algorithm and it was implemented Labview. In this technique, the non-homogeneous deformation was monitored using a mirrorless camera. A mirrorless camera was used because it is economical and it has the capacity to take pictures at a fast rate. The camera was first calibrated to remove the distortion brought about the lens and the testing environment as well. Calibration was divided into 2 phases. The first phase was the calibration of the camera parameters and distortion caused by the lens. The second phase was to for eliminating the distortion brought about the triaxial plexiglass. A correction factor was established from this phase. A series of consolidated undrained cyclic triaxial test was performed using a coarse soil. The results from the non-contact measurement technique were compared to the measured deformation from the linear variable displacement transducer. It was observed that deformation was higher at the area where failure occurs.

Keywords: cyclic loading, non-contact measurement, non-homogeneous, optical flow

Procedia PDF Downloads 291

Authors: Soofiyan Atar, Adil Shaikh, Sahil Rajpurkar, Pragnesh Bhalala, Aniket Desai, Irfan Siddavatam

Abstract:

Deep reinforcement learning (deep RL) algorithms leverage the symbolic power of complex controllers by automating it by mapping sensory inputs to low-level actions. Deep RL eliminates the complex robot dynamics with minimal engineering. Deep RL provides high-risk involvement by directly implementing it in real-world scenarios and also high sensitivity towards hyperparameters. Tuning of hyperparameters on a pneumatic quadruped robot becomes very expensive through trial-and-error learning. This paper presents an automated learning control for a pneumatic quadruped robot using sample efficient deep Q learning, enabling minimal tuning and very few trials to learn the neural network. Long training hours may degrade the pneumatic cylinder due to jerk actions originated through stochastic weights. We applied this method to the pneumatic quadruped robot, which resulted in a hopping gait. In our process, we eliminated the use of a simulator and acquired a stable gait. This approach evolves so that the resultant gait matures more sturdy towards any stochastic changes in the environment. We further show that our algorithm performed very well as compared to programmed gait using robot dynamics.

Keywords: model-based reinforcement learning, gait stability, supervised learning, pneumatic quadruped

Procedia PDF Downloads 301

Authors: Maryam Kalkatechi

Abstract:

The application of algorithms in architecture has been realized as geometric forms which are increasingly being used by architecture firms. The abstraction of ideas in a formulated algorithm is not possible. There is still a gap between design innovation and final built in prescribed formulas, even the most aesthetical realizations. This paper presents the application of erudite design process to conceptualize biomimetic paradigms in architecture. The process is customized to material and tectonics. The first part of the paper outlines the design process elements within four biomimetic pre-concepts. The pre-concepts are chosen from plants family. These include the pine leaf, the dandelion flower; the cactus flower and the sun flower. The choice of these are related to material qualities and natural pattern of the tectonics of these plants. It then focuses on four versions of tectonic comprehension of one of the biomimetic pre-concepts. The next part of the paper discusses the implementation of STEAM in higher education in architecture. This is shown by the relations within the design process and the manifestation of the thinking processes. The A in the SETAM, in this case, is only achieved by the design process, an engaging event as a performing arts, in which the conceptualization and development is realized in final built.

Keywords: biomimetic paradigm, erudite design process, tectonic, STEAM (Science, Technology, Engineering, Arts, Mathematic)

Procedia PDF Downloads 197

Authors: Jae-Young Lee, Woo-Young Jung, Myeong-Jun Nam

Abstract:

Water hammer is a hydraulic transient problem which is commonly encountered in the penstocks of hydropower plants. The numerical model was developed to estimate the transient behavior of pressure waves in pipe systems. The computational algorithm was proposed to model the water hammer phenomenon in a pipe system with pump shutdown at midstream and sudden valve closure at downstream. To predict the pressure head and flow velocity as a function of time as a result of rapidly closing a valve and pump shutdown, two boundary conditions at the ends considering pump operation and valve control can be implemented as specified equations of the pressure head and flow velocity based on the characteristics method. It was shown that the effects of transient flow make it determine the needs for protection devices, such as surge tanks, surge relief valves, or air valves, at various points in the system against overpressure and low pressure. It produced reasonably good performance with the results of the proposed transient model for pipeline systems. The proposed numerical model can be used as an efficient tool for the safety assessment of hydropower plants due to water hammer.

Keywords: water hammer, hydraulic transient, pipe systems, characteristics method

Procedia PDF Downloads 124

Authors: Rabiaa Gammoudi, Najet Rebei, Othman Hasnaoui

Abstract:

Photovoltaic systems use different techniques for tracking the Maximum Power Point (MPPT) to provide the highest possible power to the load regardless of the climatic conditions variation. In this paper, the proposed method is the Open Circuit (OC) method with sudden and random variations of insolation. The simulation results of the water pumping system controlled by OC method are validated by an experimental experience in real-time using a test bench composed by a centrifugal pump powered by a PVG via a boost chopper for the adaptation between the source and the load. The output of the DC/DC converter supplies the motor pump LOWARA type, assembly by means of a DC/AC inverter. The control part is provided by a computer incorporating a card DS1104 running environment Matlab/Simulink for visualization and data acquisition. These results show clearly the effectiveness of our control with a very good performance. The results obtained show the usefulness of the developed algorithm in solving the problem of degradation of PVG performance depending on the variation of climatic factors with a very good yield.

Keywords: PVWPS (PV Water Pumping System), maximum power point tracking (MPPT), open circuit method (OC), boost converter, DC/AC inverter

Procedia PDF Downloads 440

Authors: Ishapathik Das

Abstract:

The purpose of this article is to find a method of comparing designs for ordinal regression models using quantile dispersion graphs in the presence of linear predictor misspecification. The true relationship between response variable and the corresponding control variables are usually unknown. Experimenter assumes certain form of the linear predictor of the ordinal regression models. The assumed form of the linear predictor may not be correct always. Thus, the maximum likelihood estimates (MLE) of the unknown parameters of the model may be biased due to misspecification of the linear predictor. In this article, the uncertainty in the linear predictor is represented by an unknown function. An algorithm is provided to estimate the unknown function at the design points where observations are available. The unknown function is estimated at all points in the design region using multivariate parametric kriging. The comparison of the designs are based on a scalar valued function of the mean squared error of prediction (MSEP) matrix, which incorporates both variance and bias of the prediction caused by the misspecification in the linear predictor. The designs are compared using quantile dispersion graphs approach. The graphs also visually depict the robustness of the designs on the changes in the parameter values. Numerical examples are presented to illustrate the proposed methodology.

Keywords: model misspecification, multivariate kriging, multivariate logistic link, ordinal response models, quantile dispersion graphs

Procedia PDF Downloads 377

Authors: Tayyibah Khanam, Mohammad Saad Alam, Sanchari Deb, Yasser Rafat

Abstract:

Electric vehicles are considered as viable replacements to gasoline cars since they help in reducing harmful emissions and stimulate power generation through renewable energy sources, hence contributing to sustainability. However, one of the significant obstacles in the mass deployment of electric vehicles is the charging time anxiety among users and, thus, the subsequent large waiting times for available chargers at charging stations. Data analytics, on the other hand, has revolutionized the decision-making tasks of management and operating systems since its arrival. In this paper, we attempt to optimize the choice of EV charging stations for users in their vicinity by minimizing the time taken to reach the charging stations and the waiting times for available chargers. Time taken to travel to the charging station is calculated by the Google Maps API and the waiting times are predicted by polynomial regression of the historical data stored. The proposed framework utilizes real-time data and historical data from all operating charging stations in the city and assists the user in finding the best suitable charging station for their current situation and can be implemented in a mobile phone application. The algorithm successfully predicts the most optimal choice of a charging station and the minimum required time for various sample data sets.

Keywords: charging data, electric vehicles, machine learning, waiting times

Procedia PDF Downloads 176

Authors: Linhang Zhu, Hongyu Zhu, Jiahe Liu

Abstract:

This paper investigated a method of indoor real-time positioning and mapping based on the Manhattan world assumption. In indoor environments, relying solely on feature matching techniques or other geometric algorithms for sensor pose estimation inevitably resulted in cumulative errors, posing a significant challenge to indoor positioning. To address this issue, we adopt the Manhattan world hypothesis to optimize the camera pose algorithm based on feature matching, which improves the accuracy of camera pose estimation. A special processing method was applied to image data frames that conformed to the Manhattan world assumption. When similar data frames appeared subsequently, this could be used to eliminate drift in sensor pose estimation, thereby reducing cumulative errors in estimation and optimizing mapping and positioning. Through experimental verification, it is found that our method achieves high-precision real-time positioning in indoor environments and successfully generates maps of indoor environments. This provides effective technical support for applications such as indoor navigation and robot control.

Keywords: Manhattan world hypothesis, real-time positioning and mapping, feature matching, loopback detection

Procedia PDF Downloads 49

Authors: Hui-Ling Yang, Kuei-Ru Chou

Abstract:

Background: Studies show that cognitive training can effectively delay cognitive failure. However, there are several gaps in the previous studies of cognitive training in mild cognitive impairment: 1) previous studies enrolled mostly healthy older adults, with few recruiting older adults with cognitive impairment; 2) they also had limited generalizability and lacked long-term follow-up data and measurements of the activities of daily living functional impact. Moreover, only 37% were randomized controlled trials (RCT). 3) Limited cognitive training has been specifically developed for mild cognitive impairment. Objective: This study sought to investigate the changes in cognitive function, activities of daily living and degree of depressive symptoms in older adults with mild cognitive impairment after cognitive training. Methods: This double-blind randomized controlled study has a 2-arm parallel group design. Study subjects are older adults diagnosed with mild cognitive impairment in residential care facilities. 124 subjects will be randomized by the permuted block randomization, into intervention group (Cognitive training, CT), or active control group (Passive information activities, PIA). Therapeutic adherence, sample attrition rate, medication compliance and adverse events will be monitored during the study period, and missing data analyzed using intent-to-treat analysis (ITT). Results: Training sessions of the CT group are 45 minutes/day, 3 days/week, for 12 weeks (36 sessions each). The training of active control group is the same as CT group (45min/day, 3days/week, for 12 weeks, for a total of 36 sessions). The primary outcome is cognitive function, using the Mini-Mental Status Examination (MMSE); the secondary outcome indicators are: 1) activities of daily living, using the Lawton’s Instrumental Activities of Daily Living (IADLs) and 2) degree of depressive symptoms, using the Geriatric Depression Scale-Short form (GDS-SF). Latent growth curve modeling will be used in the repeated measures statistical analysis to estimate the trajectory of improvement by examining the rate and pattern of change in cognitive functions, activities of daily living and degree of depressive symptoms for intervention efficacy over time, and the effects will be evaluated immediate post-test, 3 months, 6 months and one year after the last session. Conclusions: We constructed a rigorous CT program adhering to the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines. We expect to determine the improvement in cognitive function, activities of daily living and degree of depressive symptoms of older adults with mild cognitive impairment after using the CT.

Keywords: mild cognitive impairment, cognitive training, randomized controlled study

Procedia PDF Downloads 433

Authors: Mmatladi D. Mnguni, Mustapha Mohammed, George Y. Mahama, Alhassan L. Abdulai, Felix D. Dakora

Abstract:

Industrial-based nitrogen (N) fertilizers are justifiably credited for the current state of food production across the globe, but their continued use is not sustainable and has an adverse effect on the environment. The search for greener and sustainable technologies has led to an increase in exploiting biological systems such as legumes and organic amendments for plant growth promotion in cropping systems. Although the benefits of legume rotation with cereal crops have been documented, the full benefits of soybean-sorghum rotation systems have not been properly evaluated in Africa. This study explored the benefits of soybean-sorghum rotation through assessing N₂ fixation and water-use efficiency of soybean in rotation with sorghum with and without organic and inorganic amendments. The field trials were conducted from 2017 to 2020. Sorghum was grown on plots previously cultivated to soybean and vice versa. The succeeding sorghum crop received fertilizer amendments [organic fertilizer (5 tons/ha as poultry litter, OF); inorganic fertilizer (80N-60P-60K) IF; organic + inorganic fertilizer (OF+IF); half organic + inorganic fertilizer (HIF+OF); organic + half inorganic fertilizer (OF+HIF); half organic + half inorganic (HOF+HIF) and control] and was arranged in a randomized complete block design. The soybean crop succeeding fertilized sorghum received a blanket application of triple superphosphate at 26 kg P ha⁻¹. Nitrogen fixation and water-use efficiency were respectively assessed at the flowering stage using the ¹⁵N and ¹³C natural abundance techniques. The results showed that the shoot dry matter of soybean plants supplied with HOF+HIF was much higher (43.20 g plant-1), followed by OF+HIF (36.45 g plant⁻¹), and HOF+IF (33.50 g plant⁻¹). Shoot N concentration ranged from 1.60 to 1.66%, and total N content from 339 to 691 mg N plant⁻¹. The δ¹⁵N values of soybean shoots ranged from -1.17‰ to -0.64‰, with plants growing on plots previously treated to HOF+HIF exhibiting much higher δ¹⁵N values, and hence lower percent N derived from N₂ fixation (%Ndfa). Shoot %Ndfa values varied from 70 to 82%. The high %Ndfa values obtained in this study suggest that the previous year’s organic and inorganic fertilizer amendments to sorghum did not inhibit N₂ fixation in the following soybean crop. The amount of N-fixed by soybean ranged from 106 to 197 kg N ha⁻¹. The treatments showed marked variations in carbon (C) content, with HOF+HIF treatment recording the highest C content. Although water-use efficiency varied from -29.32‰ to -27.85‰, shoot water-use efficiency, C concentration, and C:N ratio were not altered by previous fertilizer application to sorghum. This study provides strong evidence that previous HOF+HIF sorghum residues can enhance N nutrition and water-use efficiency in nodulated soybean.

Keywords: ¹³C and ¹⁵N natural abundance, N-fixed, organic and inorganic fertilizer amendments, shoot %Ndfa

Procedia PDF Downloads 157

Authors: Xiaohua Zou, Yongxin Su

Abstract:

The reasonable operation of heating, ventilation, and air conditioning (HVAC) is of great significance in improving the security, stability, and economy of power system operation. However, the uncertainty of the operating environment, thermal comfort varies by users and rapid decision-making pose challenges for HVAC demand response optimization. In this regard, this paper proposes a reinforcement learning-based method for HVAC demand response optimization considering few-shot personalized thermal comfort (PTC). First, an HVAC DR optimization framework based on few-shot PTC model and DRL is designed, in which the output of few-shot PTC model is regarded as the input of DRL. Then, a few-shot PTC model that distinguishes between awake and asleep states is established, which has excellent engineering usability. Next, based on soft actor criticism, an HVAC DR optimization algorithm considering the user’s PTC is designed to deal with uncertainty and make decisions rapidly. Experiment results show that the proposed method can efficiently obtain use’s PTC temperature, reduce energy cost while ensuring user’s PTC, and achieve rapid decision-making under uncertainty.

Keywords: HVAC, few-shot personalized thermal comfort, deep reinforcement learning, demand response

Procedia PDF Downloads 66

Authors: N. Boutaghane, F. Z. Tounsi

Abstract:

Many physical and technological factors degrade the SPECT images, both qualitatively and quantitatively. For this, it is not always put into leading technological advances to improve the performance of tomographic gamma camera in terms of detection, collimation, reconstruction and correction of tomographic images methods. We have to master firstly the choice of various acquisition and reconstruction parameters, accessible to clinical cases and using the attenuation and scatter correction methods to always optimize quality image and minimized to the maximum dose received by the patient. In this work, an evaluation of qualitative and quantitative tomographic images is performed based on the acquisition parameters (counts per projection) and reconstruction parameters (filter type, associated cutoff frequency). In addition, methods for correcting physical effects such as attenuation and scatter degrading the image quality and preventing precise quantitative of the reconstructed slices are also presented. Two approaches of attenuation and scatter correction are implemented: the attenuation correction by CHANG method with a filtered back projection reconstruction algorithm and scatter correction by the subtraction JASZCZAK method. Our results are considered as such recommandation, which permits to determine the origin of the different artifacts observed both in quality control tests and in clinical images.

Keywords: attenuation, scatter, reconstruction filter, image quality, acquisition and reconstruction parameters, SPECT

Procedia PDF Downloads 433

Authors: Rafi Ullah, Shah Muhammad Emaduddin, Taha Jilani

Abstract:

Vehicle Adhoc Networks (VANET) belongs to a special class of Mobile Adhoc Network (MANET) with high mobility. Network is created by road side vehicles equipped with communication devices like GPS and Wifi etc. Since the environment is highly dynamic due to difference in speed and high mobility of vehicles and weak stability of the network connection, it is a challenging task to design an efficient routing protocol for such an unstable environment. Our proposed algorithm uses heuristic for the calculation of optimal path for routing the packet efficiently in collaboration with several other parameters like geographical location, speed, priority, the distance among the vehicles, communication range, and networks congestion. We have incorporated probabilistic, heuristic and machine learning based approach inconsistency with the relay function of the memory buffer to keep the packet moving towards the destination. These parameters when used in collaboration provide us a very strong and admissible heuristics. We have mathematically proved that the proposed technique is efficient for the routing of packets, especially in a medical emergency situation. These networks can be used for medical emergency, security, entertainment and routing purposes.

Keywords: heuristics routing, intelligent routing, VANET, route optimization

Procedia PDF Downloads 162

Authors: Mahtab Davari, Charles Edward Okon, Somayeh Aghanavesi

Abstract:

Every institute of learning is usually interested in the performance of enrolled students. The level of these performances determines the approach an institute of study may adopt in rendering academic services. The focus of this paper is to evaluate students' academic performance in given courses of study using machine learning methods. This study evaluated various supervised machine learning classification algorithms such as Logistic Regression (LR), Support Vector Machine, Random Forest, Decision Tree, K-Nearest Neighbors, Linear Discriminant Analysis, and Quadratic Discriminant Analysis, using selected features to predict study performance. The accuracy, precision, recall, and F1 score obtained from a 5-Fold Cross-Validation were used to determine the best classification algorithm to predict students’ performances. SVM (using a linear kernel), LDA, and LR were identified as the best-performing machine learning methods. Also, using the LR model, this study identified students' educational habits such as reading and paying attention in class as strong determinants for a student to have an above-average performance. Other important features include the academic history of the student and work. Demographic factors such as age, gender, high school graduation, etc., had no significant effect on a student's performance.

Keywords: student performance, supervised machine learning, classification, cross-validation, prediction

Procedia PDF Downloads 112