Search results for: hyper-parameter tuning

Authors: Muhammad Rehan Usman, Muhammad Arslan Usman, Soo Young Shin

Abstract:

The new era of digital communication has brought up many challenges that network operators need to overcome. The high demand of mobile data rates require improved networks, which is a challenge for the operators in terms of maintaining the quality of experience (QoE) for their consumers. In live video transmission, there is a sheer need for live surveillance of the videos in order to maintain the quality of the network. For this purpose objective algorithms are employed to monitor the quality of the videos that are transmitted over a network. In order to test these objective algorithms, subjective quality assessment of the streamed videos is required, as the human eye is the best source of perceptual assessment. In this paper we have conducted subjective evaluation of videos with varying spatial and temporal impairments. These videos were impaired with frame freezing distortions so that the impact of frame freezing on the quality of experience could be studied. We present subjective Mean Opinion Score (MOS) for these videos that can be used for fine tuning the objective algorithms for video quality assessment.

Keywords: frame freezing, mean opinion score, objective assessment, subjective evaluation

Procedia PDF Downloads 494

Authors: Pinghe Wang

Abstract:

In this paper, we demonstrate a tunable long-cavity passive mode-locked fiber laser. The mode locker is a nonlinear amplifying loop mirror (NALM). The cavity frequency of the laser is 465 kHz because that 404m SMF is inserted in the cavity. A tunable bandpass filter with ~1nm 3dB bandwidth is inserted into the cavity to realize tunable mode locking. The passive mode-locked laser at a fixed wavelength is investigated in detail. The experimental results indicate that the laser operates in dissipative soliton resonance (DSR) region. When the pump power is 400mW, the laser generates the rectangular pulses with 10.58 ns pulse duration, 70.28nJ single-pulse energy. When the pump power is 400mW, the laser keeps stable mode locking status in the range from 1523.4nm to 1575nm. During the whole tuning range, the SNR, the pulse duration, the output power and single pulse energy have a little fluctuation because that the gain of the EDF changes with the wavelength.

Keywords: fiber laser, dissipative soliton resonance, mode locking, tunable

Procedia PDF Downloads 237

Authors: Nassim Bessaad, Qilian Bao, Zhao Jiangkang

Abstract:

The fiber optics gyroscope in the strap-down inertial navigation system (FOG-SINS) suffers from precision degradation due to the influence of random errors. In this work, an enhanced Allan variance (AV) stochastic modeling method combined with discrete wavelet transform (DWT) for signal denoising is implemented to estimate the random process in the FOG signal. Furthermore, we devise a measurement-based iterative adaptive Sage-Husa nonlinear filter with augmented states to integrate a star tracker sensor with SINS. The proposed filter adapts the measurement noise covariance matrix based on the available data. Moreover, the enhanced stochastic modeling scheme is invested in tuning the process noise covariance matrix and the augmented state Gauss-Markov process parameters. Finally, the effectiveness of the proposed filter is investigated by employing the collected data in laboratory conditions. The result shows the filter's improved accuracy in comparison with the conventional Kalman filter (CKF).

Keywords: inertial navigation, adaptive filtering, star tracker, FOG

Procedia PDF Downloads 80

Authors: Moses Emetere, Marvel Akinyemi, S. E. Sanni

Abstract:

At the moment, AERONET or AMMA database shows a large volume of data loss. With only about 47% data set available to the scientist, it is evident that accurate nowcast or forecast cannot be guaranteed. The calibration constants of most radiosonde or weather stations are not compatible with the atmospheric conditions of the West African climate. A dispersion model was developed to incorporate salient mathematical representations like a Unified number. The Unified number was derived to describe the turbulence of the aerosols transport in the frictional layer of the lower atmosphere. Fourteen years data set from Multi-angle Imaging SpectroRadiometer (MISR) was tested using the dispersion model. A yearly estimation of the atmospheric constants over Ouagadougou using the model was obtained with about 87.5% accuracy. It further revealed that the average atmospheric constant for Ouagadougou-Niger is a_1 = 0.626, a_2 = 0.7999 and the tuning constants is n_1 = 0.09835 and n_2 = 0.266. Also, the yearly atmospheric constants affirmed the lower atmosphere of Ouagadougou is very dynamic. Hence, it is recommended that radiosonde and weather station manufacturers should constantly review the atmospheric constant over a geographical location to enable about eighty percent data retrieval.

Keywords: aerosols retention, aerosols loading, statistics, analytical technique

Procedia PDF Downloads 315

Authors: Jalil Boudjadar

Abstract:

Dynamic Voltage and Frequency Scaling (DVFS) multicore platforms are promising execution platforms that enable high computational performance, less energy consumption and flexibility in scheduling the system processes. However, the resulting interleaving and memory interference together with per-core frequency tuning make real-time guarantees hard to be delivered. Besides, energy consumption represents a strong constraint for the deployment of such systems on energy-limited settings. Identifying the system configurations that would achieve a high performance and consume less energy while guaranteeing the system schedulability is a complex task in the design of modern embedded systems. This work studies the trade-off between energy consumption, cores utilization and memory bottleneck and their impact on the schedulability of DVFS multicore time-critical systems with a hierarchy of shared memories. We build a model-based framework using Parametrized Timed Automata of UPPAAL to analyze the mutual impact of performance, energy consumption and schedulability of DVFS multicore systems, and demonstrate the trade-off on an actual case study.

Keywords: time-critical systems, multicore systems, schedulability analysis, energy consumption, performance analysis

Procedia PDF Downloads 107

Authors: Amani Alotaibi

Abstract:

3D printing has emerged as a pivotal technique for scaffold development, particularly in the field of bone tissue regeneration, due to its ability to customize scaffolds to fit complex geometries of bone defects. Among the various methods available, fused deposition modeling (FDM) is particularly promising as it avoids the use of solvents or toxic chemicals during fabrication. This study investigates the effects of three key parameters, extrusion temperature, screw rotational speed, and deposition speed, on the crystallization and mechanical properties of polycaprolactone (PCL) scaffolds. Three extrusion temperatures (70°C, 80°C, and 90°C), three screw speeds (10 RPM, 15 RPM, and 20 RPM), and three deposition speeds (8 mm/s, 10 mm/s, and 12 mm/s) were evaluated. The scaffolds were characterized using X-ray diffraction (XRD), differential scanning calorimetry (DSC), and tensile testing to assess changes in crystallinity and mechanical properties. Additionally, the scaffolds were analyzed for crystal size and biocompatibility. The results demonstrated that increasing the extrusion temperature to 80°C, combined with a screw speed of 15 RPM and a deposition speed of 10 mm/s, significantly improved the crystallinity, compressive modulus, and thermal resistance of the PCL scaffolds. These findings suggest that by fine-tuning basic 3D printing parameters, it is possible to modulate the structural and mechanical properties of the scaffold, thereby enhancing its suitability for bone tissue regeneration.

Keywords: 3D printing, polymer, scaffolds, tissue engineering, crystallization

Procedia PDF Downloads 0

Authors: Alexander Goncharenko, Andrey Denisov, Sergey Alyamkin, Evgeny Terentev

Abstract:

The neural network quantization is highly desired procedure to perform before running neural networks on mobile devices. Quantization without fine-tuning leads to accuracy drop of the model, whereas commonly used training with quantization is done on the full set of the labeled data and therefore is both time- and resource-consuming. Real life applications require simplification and acceleration of quantization procedure that will maintain accuracy of full-precision neural network, especially for modern mobile neural network architectures like Mobilenet-v1, MobileNet-v2 and MNAS. Here we present a method to significantly optimize training with quantization procedure by introducing the trained scale factors for discretization thresholds that are separate for each filter. Using the proposed technique, we quantize the modern mobile architectures of neural networks with the set of train data of only ∼ 10% of the total ImageNet 2012 sample. Such reduction of train dataset size and small number of trainable parameters allow to fine-tune the network for several hours while maintaining the high accuracy of quantized model (accuracy drop was less than 0.5%). Ready-for-use models and code are available in the GitHub repository.

Keywords: distillation, machine learning, neural networks, quantization

Procedia PDF Downloads 325

Authors: Le Linh, Truong Nguyen Luan Vu, Le Hieu Giang

Abstract:

In this work, the IMC-PID controller cascaded filter based on Internal Model Control (IMC) scheme is systematically proposed for the simplified decoupling control system. The simplified decoupling is firstly introduced for multivariable processes by using coefficient matching to obtain a stable, proper, and causal simplified decoupler. Accordingly, transfer functions of decoupled apparent processes can be expressed as a set of n equivalent independent processes and then derived as a ratio of the original open-loop transfer function to the diagonal element of the dynamic relative gain array. The IMC-PID controller in series with filter is then directly employed to enhance the overall performance of the decoupling control system while avoiding difficulties arising from properties inherent to simplified decoupling. Some simulation studies are considered to demonstrate the simplicity and effectiveness of the proposed method. Simulations were conducted by tuning various controllers of the multivariate processes with multiple time delays. The results indicate that the proposed method consistently performs well with fast and well-balanced closed-loop time responses.

Keywords: coefficient matching method, internal model control (IMC) scheme, PID controller cascaded filter, simplified decoupler

Procedia PDF Downloads 442

Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan

Abstract:

Gliomas are most challenging and aggressive type of tumors which appear in different sizes, locations, and scattered boundaries. CNN is most efficient deep learning approach with outstanding capability of solving image analysis problems. A fully automatic deep learning based 2D-CNN model for brain tumor segmentation is presented in this paper. We used small convolution filters (3 x 3) to make architecture deeper. We increased convolutional layers for efficient learning of complex features from large dataset. We achieved better results by pushing convolutional layers up to 16 layers for HGG model. We achieved reliable and accurate results through fine-tuning among dataset and hyper-parameters. Pre-processing of this model includes generation of brain pipeline, intensity normalization, bias correction and data augmentation. We used the BRATS-2015, and Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.81 for complete, 0.79 for core, 0.80 for enhanced tumor regions. However, these results are comparable with methods already implemented 2D CNN architecture.

Keywords: brain tumor segmentation, convolutional neural networks, deep learning, HGG

Procedia PDF Downloads 256

Authors: H. Vahdani

Abstract:

By the adsorption of polar compounds and/or the deposition of organic material, the wettability of originally water-wet reservoir rock can be altered. The degree of alteration is determined by the interaction of the oil constituents, the mineral surface, and the brine chemistry. Recently improving oil recovery by tuning wettability alteration is believed as a new recovery method. Various researchers have demonstrated that low salt water injection has a significant impact on oil recovery. It has been shown, for instance, that additional oil can be produced from reservoir rock by managing the injection water. Large wettability sensitivity has been observed, indicating that the oil/water capillary pressure profiles play a major role during low saline water injection simulation. Although the exact physics on how this alteration occurs is still a research topic; however, it has been reported that some of its effect can be captured by a relative permeability shift from an oil-wet system to a water-wet system. Modeling of low salt water injection mainly is based on the theory of wettability alteration and is hence strongly dependent on the wettability of the reservoir. In this article, combination of different wettabilities has been simulated and it is observed that the highest recoveries were from the cases were the reservoir initially was water-wet, and the lowest recoveries was from the cases were the reservoir initially was considered oil-wet. However for the cases where the reservoir initially was oil-wet, the effect of low-salinity waterflooding was the largest.

Keywords: low salt water injection, wettability alteration, modelling, relative permeability

Procedia PDF Downloads 494

Authors: Shinji Kajiwara, Tadamasa Fukuoka

Abstract:

The formula project of Kinki University has been involved in the student Formula SAE of Japan (JSAE) since the second year the competition was held. The vehicle developed in the project uses a ZX-6R engine, which has been manufactured by Kawasaki Heavy Industries for the JSAE competition for the eighth time. The limited performance of the concept vehicle was improved through the development of a power train. The supercharger loading, engine dry sump, and engine cooling management of the vehicle were also enhanced. The supercharger loading enabled the vehicle to achieve a maximum output of 59.6 kW (80.6 PS)/9000 rpm and a maximum torque of 70.6 Nm (7.2 kgf m)/8000 rpm. We successfully achieved 90% of the engine’s torque band (4000–10000 rpm) with 50% of the revolutions in regular engine use (2000–12000 rpm). Using a dry sump system, we periodically managed hydraulic pressure during engine operation. A system that controls engine stoppage when hydraulic pressure falls was also constructed. Using the dry sump system at 80 mm reduced the required engine load and the vehicle’s center of gravity. Even when engine motion was suspended by the electromotive force exerted by the water pump, the circulation of cooling water was still possible. These findings enabled us to create a cooling system in accordance with the requirements of the competition.

Keywords: engine, combustion, cooling system, numerical simulation, power, torque, mechanical super charger

Procedia PDF Downloads 300

Authors: Soma Chatterjee, I. Das

Abstract:

In the present study, we have investigated the structural, electrical and magneto-transport properties of polycrystalline and nanocrystalline Gd0.5Sr0.5MnO3 compounds. The variation of transport properties is modified by tuning the grain size of the material. In the high-temperature semiconducting region, temperature-dependent resistivity data can be well explained by the non-adiabatic small polaron hopping (SPH) mechanism. In addition, the resistivity data for all compounds in the low-temperature paramagnetic region can also be well explained by the variable range hopping (VRH) model. The parameters obtained from SPH and VRH mechanisms are found to be reasonable. In the case of nanocrystalline compounds, there is an overlapping temperature range where both SPH and VRH models are valid simultaneously, and a new conduction mechanism - variable range hopping of small polaron s(VR-SPH) is satisfactorily valid for the whole temperature range of these compounds. However, for the polycrystalline compound, the overlapping temperature region between VRH and SPH models does not exist and the VR-SPH mechanism is not valid here. Thus, polarons play a leading role in selecting different conduction mechanisms in different temperature ranges.

Keywords: electrical resistivity, manganite, small polaron hopping, variable range hopping, variable range of small polaron hopping

Procedia PDF Downloads 89

Authors: Laidi Maamar, Achwak Madani, Abdellah El Ahdj Abdellah

Abstract:

The aim of this work is to use Support Vector regression (SVR) combined with dragonfly, firefly, Bee Colony and particle swarm Optimization algorithm to predict global solar radiation on horizontal surfaces in some cities in Algeria. Combining these optimization algorithms with SVR aims principally to enhance accuracy by fine-tuning the parameters, speeding up the convergence of the SVR model, and exploring a larger search space efficiently; these parameters are the regularization parameter (C), kernel parameters, and epsilon parameter. By doing so, the aim is to improve the generalization and predictive accuracy of the SVR model. Overall, the aim is to leverage the strengths of both SVR and optimization algorithms to create a more powerful and effective regression model for various cities and under different climate conditions. Results demonstrate close agreement between predicted and measured data in terms of different metrics. In summary, SVM has proven to be a valuable tool in modeling global solar radiation, offering accurate predictions and demonstrating versatility when combined with other algorithms or used in hybrid forecasting models.

Keywords: support vector regression (SVR), optimization algorithms, global solar radiation prediction, hybrid forecasting models

Procedia PDF Downloads 35

Authors: Mandar Ghodekar, Sharad Jadhav, Sangram Jadhav

Abstract:

Speed control of rotor during startup and under varying load conditions is one of the most difficult tasks of gas turbine operation. In this paper, power plant gas turbine (GE9001E) is considered for this purpose and fuzzy and fuzzy-PI rotor speed controllers are designed. The goal of the presented controllers is to keep the turbine rotor speed within predefined limits during startup condition as well as during operating condition. The fuzzy controller and fuzzy-PI controller are designed using Takagi-Sugeno method and Mamdani method, respectively. In applying the fuzzy-PI control to a gas-turbine plant, the tuning parameters (Kp and Ki) are modified online by fuzzy logic approach. Error and rate of change of error are inputs and change in fuel flow is output for both the controllers. Hence, rotor speed of gas turbine is controlled by modifying the fuel ƒflow. The identified linear ARX model of gas turbine is considered while designing the controllers. For simulations, demand power is taken as disturbance input. It is assumed that inlet guide vane (IGV) position is fixed. In addition, the constraint on the fuel flow is taken into account. The performance of the presented controllers is compared with each other as well as with H∞ robust and MPC controllers for the same operating conditions in simulations.

Keywords: gas turbine, fuzzy controller, fuzzy PI controller, power plant

Procedia PDF Downloads 334

Authors: Amal A. Hassan, Faten H. Fahmy, Abd El-Shafy A. Nafeh, Hosam K. M. Youssef

Abstract:

Due to the growing penetration of wind energy into the power grid, it is very important to study its interactions with the power system and to provide good control technique in order to deliver high quality power. In this paper, an intelligent control methodology is proposed for optimizing the controllers’ parameters of doubly fed induction generator (DFIG) based wind turbine generation system (WTGS). The genetic algorithm (GA) and particle swarm optimization (PSO) are employed and compared for the parameters adaptive tuning of the proposed proportional integral (PI) multiple controllers of the back to back converters of the DFIG based WTGS. For this purpose, the dynamic model of WTGS with DFIG and its associated controllers is presented. Furthermore, the simulation of the system is performed using MATLAB/SIMULINK and SIMPOWERSYSTEM toolbox to illustrate the performance of the optimized controllers. Finally, this work is validated to 33-bus test radial system to show the interaction between wind distributed generation (DG) systems and the distribution network.

Keywords: DFIG wind turine, intelligent control, distributed generation, particle swarm optimization, genetic algorithm

Procedia PDF Downloads 267

Authors: Aicha Bouhlala, Sabah Chettibi

Abstract:

Doping CeO₂ with transition metals is an effective way of tuning its properties. In the present work, we have performed self-consistent ab-initio calculation using the full-potential linearized augmented plane-wave method (FP-LAPW), based on the density functional theory (DFT) as implemented in the Wien2k simulation code to study the structural, electronic, and magnetic properties of the compound Ce₁₋ₓCuₓO₂ (x=12.5%) fluorite type oxide and to explore the effects of dopant Cu in ceria. The exchange correlation potential has been treated using the Perdew-Burke-Eenzerhof revised of solid (PBEsol). In structural properties, the equilibrium lattice constant is observed for the compound, which exists within the value of 5.382 A°. In electronic properties, the spin-polarized electronic bandstructure elucidates the semiconductor nature of the material in both spin channels, with the compound was observed to have a narrow bandgap on the spin-down configuration (0.162 EV) and bandgap on the spin-up (2.067 EV). Hence, the doped atom Cu plays a vital role in increasing the magnetic moments of the supercell, and the value of the total magnetic moment is found to be 2.99438 μB. Therefore, the compound Cu-doped CeO₂ shows a strong ferromagnetic behavior. The predicted results propose the compound could be a good candidate for spintronics applications.

Keywords: Cu-doped CeO₂, DFT, Wien2k, properties

Procedia PDF Downloads 255

Authors: Elham Jelodari Mamaghani, Christian Prins, Haoxun Chen

Abstract:

In this paper, there is concentration on collaborative transportation planning (CTP) among multiple carriers with pickup and delivery requests and time windows. This problem is a vehicle routing problem with constraints from standard vehicle routing problems and new constraints from a real-world application. In the problem, each carrier has a finite number of vehicles, and each request is a pickup and delivery request with time window. Moreover, each carrier has reserved requests, which must be served by itself, whereas its exchangeable requests can be outsourced to and served by other carriers. This collaboration among carriers can help them to reduce total transportation costs. A mixed integer programming model is proposed to the problem. To solve the model, a hybrid algorithm that combines Genetic Algorithm and Simulated Annealing (GASA) is proposed. This algorithm takes advantages of GASA at the same time. After tuning the parameters of the algorithm with the Taguchi method, the experiments are conducted and experimental results are provided for the hybrid algorithm. The results are compared with those obtained by a commercial solver. The comparison indicates that the GASA significantly outperforms the commercial solver.

Keywords: centralized collaborative transportation, collaborative transportation with pickup and delivery, collaborative transportation with time windows, hybrid algorithm of GA and SA

Procedia PDF Downloads 392

Authors: Nicanor Garcia Alvarez, Belarmino Adenso-Diaz, Laura Calzada Infante

Abstract:

In recent years, researchers have shown an interest in modelling maritime traffic as a complex network. In this paper, we propose a bipartite weighted network to model maritime traffic and detect port maritime communities. The bipartite weighted network considers two different types of nodes. The first one represents Spanish ports, while the second one represents the countries with which there is major import/export activity. The flow among both types of nodes is modeled by weighting the volume of product transported. To illustrate the model, the data is segmented by each type of traffic. This will allow fine tuning and the creation of communities for each type of traffic and therefore finding similar ports for a specific type of traffic, which will provide decision-makers with tools to search for alliances or identify their competitors. The traffic with the greatest impact on the Spanish gross domestic product is selected, and the evolution of the communities formed by the most important ports and their differences between 2019 and 2009 will be analyzed. Finally, the set of communities formed by the ports of the Spanish port system will be inspected to determine global similarities between them, analyzing the sum of the membership of the different ports in communities formed for each type of traffic in particular.

Keywords: bipartite networks, competition, infomap, maritime traffic, port communities

Procedia PDF Downloads 148

Authors: R. Simutis, V. Galvanauskas, D. Levisauskas, J. Repsyte, V. Grincas

Abstract:

This paper deals with advanced state estimation algorithms for estimation of biomass concentration and specific growth rate in a typical fed-batch biotechnological process. This biotechnological process was represented by a nonlinear mass-balance based process model. Extended Kalman Filter (EKF) and Particle Filter (PF) was used to estimate the unmeasured state variables from oxygen uptake rate (OUR) and base consumption (BC) measurements. To obtain more general results, a simplified process model was involved in EKF and PF estimation algorithms. This model doesn’t require any special growth kinetic equations and could be applied for state estimation in various bioprocesses. The focus of this investigation was concentrated on the comparison of the estimation quality of the EKF and PF estimators by applying different measurement noises. The simulation results show that Particle Filter algorithm requires significantly more computation time for state estimation but gives lower estimation errors both for biomass concentration and specific growth rate. Also the tuning procedure for Particle Filter is simpler than for EKF. Consequently, Particle Filter should be preferred in real applications, especially for monitoring of industrial bioprocesses where the simplified implementation procedures are always desirable.

Keywords: biomass concentration, extended Kalman filter, particle filter, state estimation, specific growth rate

Procedia PDF Downloads 428

Authors: Ayad Al-Mahturi, Herman Wahid

Abstract:

This paper presents an optimal state feedback controller based on Linear Quadratic Regulator (LQR) for a two-rotor aero-dynamical system (TRAS). TRAS is a highly nonlinear multi-input multi-output (MIMO) system with two degrees of freedom and cross coupling. There are two parameters that define the behavior of LQR controller: state weighting matrix and control weighting matrix. The two parameters influence the performance of LQR. Particle Swarm Optimization (PSO) is proposed to optimally tune weighting matrices of LQR. The major concern of using LQR controller is to stabilize the TRAS by making the beam move quickly and accurately for tracking a trajectory or to reach a desired altitude. The simulation results were carried out in MATLAB/Simulink. The system is decoupled into two single-input single-output (SISO) systems. Comparing the performance of the optimized proportional, integral and derivative (PID) controller provided by INTECO, results depict that LQR controller gives a better performance in terms of both transient and steady state responses when PSO is performed.

Keywords: LQR controller, optimal control, particle swarm optimization (PSO), two rotor aero-dynamical system (TRAS)

Procedia PDF Downloads 322

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

Abstract:

Artificial neural networks have been used in different fields of artificial intelligence, and more specifically in machine learning. Although, other machine learning options are feasible in most situations, but the ease with which neural networks lend themselves to different problems which include pattern recognition, image compression, classification, computer vision, regression etc. has earned it a remarkable place in the machine learning field. This research exploits neural networks as a data mining tool in predicting the number of times a student repeats a course, considering some attributes relating to the course itself, the teacher, and the particular student. Neural networks were used in this work to map the relationship between some attributes related to students’ course assessment and the number of times a student will possibly repeat a course before he passes. It is the hope that the possibility to predict students’ performance from such complex relationships can help facilitate the fine-tuning of academic systems and policies implemented in learning environments. To validate the power of neural networks in data mining, Turkish students’ performance database has been used; feedforward and radial basis function networks were trained for this task; and the performances obtained from these networks evaluated in consideration of achieved recognition rates and training time.

Keywords: artificial neural network, data mining, classification, students’ evaluation

Procedia PDF Downloads 613

Authors: Yoshie Asahara, Hidekuni Takao

Abstract:

Non-invasive positive pressure ventilation (NPPV), a type of ventilation therapy, is a treatment in which a mask is attached to the patient's face and delivers gas into the mask to support breathing. The NPPV mask uses a strap, which is necessary to attach and secure the mask in the appropriate facial position, but the tensile strength of the strap is adjusted by the sensation of the hands. The strap uniformity and fine-tuning strap tension are judged by the skill of the operator and the amount felt by the finger. In the future, additional strap operation and adjustment methods will be required to meet the needs for reducing the burden on the patient’s face. In this study, we fabricated a mechanism that can measure, adjust and fix the tension of the straps. A small amount of strap tension can be adjusted by rotating the shaft. This makes it possible to control the slight strap tension that is difficult to grasp with the sense of the operator's hand. In addition, this mechanism allows the operator to control the strap while controlling the movement of the mask body. This leads to the establishment of a suitable mask fitting method for each patient. The developed mechanism enables the operation and fine reproducible adjustment of the strap tension and the mask balance, reducing the burden on the face.

Keywords: balance of the mask strap, fine adjustment, film sensor, mask fitting technique, mask strap tension

Procedia PDF Downloads 238

Authors: Atilla Bayram

Abstract:

This paper presents the trajectory tracking control of a spatial redundant hybrid manipulator. This manipulator consists of two parallel manipulators which are a variable geometry truss (VGT) module. In fact, each VGT module with 3-degress of freedom (DOF) is a planar parallel manipulator and their operational planes of these VGT modules are arranged to be orthogonal to each other. Also, the manipulator contains a twist motion part attached to the top of the second VGT module to supply the missing orientation of the endeffector. These three modules constitute totally 7-DOF hybrid (parallel-parallel) redundant spatial manipulator. The forward kinematics equations of this manipulator are obtained, then, according to these equations, the inverse kinematics is solved based on an optimization with the joint limit avoidance. The dynamic equations are formed by using virtual work method. In order to test the performance of the redundant manipulator and the controllers presented, two different desired trajectories are followed by using the computed force control method and a switching control method. The switching control method is combined with the computed force control method and genetic algorithm. In the switching control method, the genetic algorithm is only used for fine tuning in the compensation of the trajectory tracking errors.

Keywords: computed force method, genetic algorithm, hybrid manipulator, inverse kinematics of redundant manipulators, variable geometry truss

Procedia PDF Downloads 347

Authors: Shuen-Tai Wang, Fang-An Kuo, Chau-Yi Chou, Yu-Bin Fang

Abstract:

2016 has become the year of the Artificial Intelligence explosion. AI technologies are getting more and more matured that most world well-known tech giants are making large investment to increase the capabilities in AI. Machine learning is the science of getting computers to act without being explicitly programmed, and deep learning is a subset of machine learning that uses deep neural network to train a machine to learn  features directly from data. Deep learning realizes many machine learning applications which expand the field of AI. At the present time, deep learning frameworks have been widely deployed on servers for deep learning applications in both academia and industry. In training deep neural networks, there are many standard processes or algorithms, but the performance of different frameworks might be different. In this paper we evaluate the running performance of two state-of-the-art distributed deep learning frameworks that are running training calculation in parallel over multi GPU and multi nodes in our cloud environment. We evaluate the training performance of the frameworks with ResNet-50 convolutional neural network, and we analyze what factors that result in the performance among both distributed frameworks as well. Through the experimental analysis, we identify the overheads which could be further optimized. The main contribution is that the evaluation results provide further optimization directions in both performance tuning and algorithmic design.

Keywords: artificial intelligence, machine learning, deep learning, convolutional neural networks

Procedia PDF Downloads 211

Authors: Manoj M. Sarode, Sharad P. Jadhav, Mukesh D. Patil, Pushparaj S. Suryawanshi

Abstract:

Even though the PID controller is widely used in industrial process, tuning of PID parameters are not easy. It is a time consuming and requires expert people. Another drawback of PID controller is that process dynamics might change over time. This can happen due to variation of the process load, normal wear and tear etc. To compensate for process behavior change over time, expert users are required to recalibrate the PID gains. Implementation of model based controllers usually needs a process model. Identification of process model is time consuming job and no guaranty of model accuracy. If the identified model is not accurate, performance of the controller may degrade. Model based controllers are quite expensive and the whole procedure for the implementation is sometimes tedious. To eliminate such issues Autotuning PID controller becomes vital element. Software based Relay Feedback Autotuning Controller proves to be efficient, upgradable and maintenance free controller. In Relay Feedback Autotune controller PID parameters can be achieved with a very short span of time. This paper presents the real time implementation of LabVIEW based Relay Feedback Autotuning PID controller. It is successfully developed and implemented to control level of a laboratory setup. Its performance is analyzed for different setpoints and found satisfactorily.

Keywords: autotuning, PID, liquid level control, recalibrate, labview, controller

Procedia PDF Downloads 394

Authors: Mohamed I. Mahrous, Mohamed G. Ashmawy

Abstract:

Hybrid genetic algorithm (HGA) is proposed in this paper to determine the economic scheduling of electric power generation over a fixed time period under various system and operational constraints. The proposed technique can outperform conventional genetic algorithms (CGAs) in the sense that HGA make it possible to improve both the quality of the solution and reduce the computing expenses. In contrast, any carefully designed GA is only able to balance the exploration and the exploitation of the search effort, which means that an increase in the accuracy of a solution can only occure at the sacrifice of convergent speed, and vice visa. It is unlikely that both of them can be improved simultaneously. The proposed hybrid scheme is developed in such a way that a simple GA is acting as a base level search, which makes a quick decision to direct the search towards the optimal region, and a local search method (pattern search technique) is next employed to do the fine tuning. The aim of the strategy is to achieve the cost reduction within a reasonable computing time. The effectiveness of the proposed hybrid technique is verified on two real public electricity supply systems with 13 and 40 generator units respectively. The simulation results obtained with the HGA for the two real systems are very encouraging with regard to the computational expenses and the cost reduction of power generation.

Keywords: genetic algorithms, economic dispatch, pattern search

Procedia PDF Downloads 444

Authors: Ma. Gracia Corazon Cayanan, Kai Yuen Cheong, Li Sha

Abstract:

Training a language model for a minority language has been a challenging task. The lack of available corpora to train and fine-tune state-of-the-art language models is still a challenge in the area of Natural Language Processing (NLP). Moreover, the need for high computational resources and bulk data limit the attainment of this task. In this paper, we presented the following contributions: (1) we introduce and used a translation pair set of Tagalog and English (TL-EN) in pre-training a language model to a minority language resource; (2) we fine-tuned and evaluated top-ranking and pre-trained semantic textual similarity binary task (STSB) models, to both TL-EN and STS dataset pairs. (3) then, we reduced the size of the model to offset the need for high computational resources. Based on our results, the models that were pre-trained to translation pairs and STS pairs can perform well for STSB task. Also, having it reduced to a smaller dimension has no negative effect on the performance but rather has a notable increase on the similarity scores. Moreover, models that were pre-trained to a similar dataset have a tremendous effect on the model’s performance scores.

Keywords: semantic matching, semantic textual similarity binary task, low resource minority language, fine-tuning, dimension reduction, transformer models

Procedia PDF Downloads 211

Authors: Sandy Black, Calum McLaughlin, Alessandro Pranzitelli, Marc Laing

Abstract:

Multiphase flow structures of two-phase multicomponent fluids were experimentally investigated in a large diameter high-pressure pipeline up to 130 bar at TÜV SÜD’s National Engineering Laboratory Advanced Multiphase Facility. One of the main objectives of the experimental test campaign was to evaluate the impact of pressure on multiphase flow patterns as much of the existing information is based on low-pressure measurements. The experiments were performed in a horizontal and vertical orientation in both 4-inch and 6-inch pipework using nitrogen, ExxsolTM D140 oil, and a 6% aqueous solution of NaCl at incremental pressures from 10 bar to 130 bar. To visualise the detailed structure of the flow of the entire cross-section of the pipe, a fast response X-ray tomography system was used. A wide range of superficial velocities from 0.6 m/s to 24.0 m/s for gas and 0.04 m/s and 6.48 m/s for liquid was examined to evaluate different flow regimes. The results illustrated the suppression of instabilities between the gas and the liquid at the measurement location and that intermittent or slug flow was observed less frequently as the pressure was increased. CFD modellings of low and high-pressure simulations were able to successfully predict the likelihood of intermittent flow; however, further tuning is necessary to predict the slugging frequency. The dataset generated is unique as limited datasets exist above 100 bar and is of considerable value to multiphase flow specialists and numerical modellers.

Keywords: computational fluid dynamics, high pressure, multiphase, X-ray tomography

Procedia PDF Downloads 142

Authors: Emily Awuor, Helga Huszar, Zoltan Bozoki

Abstract:

Isotope analysis has found numerous applications in the environmental science discipline, most common being the tracing of environmental contaminants on both regional and global scales. Many environmental contaminants contain ammonia (NH₃) since it is the most abundant gas in the atmosphere and its largest sources are from agricultural and industrial activities. NH₃ isotopes (¹⁴NH₃ and ¹⁵NH₃) are therefore important and can be used in the traceability studies of these atmospheric pollutants. The goal of the project is the construction of a photoacoustic spectroscopy system that is capable of measuring ¹⁵NH₃ isotope selectively in terms of its concentration. A further objective is for the system to be robust, easy-to-use, and automated. This is provided by using two telecommunication type near-infrared distributed feedback (DFB) diode lasers and a laser coupler as the light source in the photoacoustic measurement system. The central wavelength of the lasers in use was 1532 nm, with the tuning range of ± 1 nm. In this range, strong absorption lines can be found for both ¹⁴NH₃ and ¹⁵NH₃. For the selective measurement of ¹⁵NH₃, wavelengths were chosen where the cross effect of ¹⁴NH₃ and water vapor is negligible. We completed the calibration of the photoacoustic system, and as a result, the lowest detectable concentration was 3.32 ppm (3Ϭ) in the case of ¹⁵NH₃ and 0.44 ppm (3Ϭ) in the case of ¹⁴NH₃. The results are most useful in the environmental pollution measurement and analysis.

Keywords: ammonia isotope, near-infrared DFB diode laser, photoacoustic spectroscopy, environmental monitoring

Procedia PDF Downloads 148

Authors: Jessica Lopez Espejel, Mahaman Sanoussi Yahaya Alassan, Walid Dahhane, El Hassane Ettifouri

Abstract:

Pretrained transformer-based models have shown high performance in natural language generation tasks. However, a new wave of interest has surged: automatic programming language code generation. This task consists of translating natural language instructions to a source code. Despite the fact that well-known pre-trained models on language generation have achieved good performance in learning programming languages, effort is still needed in automatic code generation. In this paper, we introduce JaCoText, a model based on Transformer neural network. It aims to generate java source code from natural language text. JaCoText leverages the advantages of both natural language and code generation models. More specifically, we study some findings from state of the art and use them to (1) initialize our model from powerful pre-trained models, (2) explore additional pretraining on our java dataset, (3) lead experiments combining the unimodal and bimodal data in training, and (4) scale the input and output length during the fine-tuning of the model. Conducted experiments on CONCODE dataset show that JaCoText achieves new state-of-the-art results.

Keywords: java code generation, natural language processing, sequence-to-sequence models, transformer neural networks

Procedia PDF Downloads 284