Speed-up of high accuracy analog test stimulus optimization

2003 ◽  
Author(s):  
A. Khouas ◽  
A. Derieux
Keyword(s):  
Test Stimulus ◽  
High Accuracy ◽  
Analog Test ◽  
Speed Up

MOCUM Code Verification and Sensibility Study Using C5G7 Benchmark

2016 ◽  
Author(s):  
Xue Yang ◽  
Rajan Borse ◽  
Nader Satvat
Keyword(s):  
Method Of Characteristics ◽  
Polar Angle ◽  
Mesh Size ◽  
Sensitivity Study ◽  
High Accuracy ◽  
Benchmark Problem ◽  
Code Verification ◽  
Heterogeneous Effects ◽  
Speed Up ◽  
Better Than

This work uses the 2-D C5G7 benchmark to verify the accuracy of the MOCUM code, a parallel neutronics program based on the method of characteristics (MOC) for solving arbitrary core geometry. Compared to the MCNP results, MOCUM k-eff, maximum assembly and pin power percentage errors are 0.02%, −0.06%, and 0.64%, respectively. The results demonstrate the high accuracy of the MOCUM code. The calculation uses a total of 56 threads, and the runtime on dual Intel Xeon E5-2699 v3 CPUs is 26 minutes, with speed up higher than 50 times. The sensitivity study of various MOC parameters using the calculation of the C5G7 benchmark problem is also performed. The study reveals that C5G7 requires the usage of 48 or more azimuthal angles. The strong flux gradient and the heterogeneous effects need fine unstructured meshes to resolve. The simulation uses 258 million zones with an average mesh size of 0.016 cm2. The investigation of the polar angle quadrature indicates that Leonard polar angle performs slightly better than Gauss-Legendre and Tabuchi polar angles and more than three polar angles are not necessary. In addition, parameter sensitivity study shows that coarse parameters are prone to introduce error to the neutron flux but not k-eff.

Stereo Vision Based Distance Measurement and its Application

2014 ◽  
Vol 926-930 ◽  
pp. 3258-3261
Author(s):  
Gan Le Hu ◽  
Yan Chen ◽  
Zheng Guo Gu
Keyword(s):  
Computer Graphics ◽  
Stereo Vision ◽  
Stereo Matching ◽  
High Accuracy ◽  
Good Combination ◽  
Practical Applications ◽  
Time Performance ◽  
Speed Up ◽  
Left And Right ◽  
Made In

In computer graphics, Stereo Vision has been a research hotspot for many years, and it is been widely used in many areas. Stereo Vision is a technique which utilizes computer to simulate human eye system. In order to achieve this simulation, two main problems need to be solved: camera calibration and stereo matching. We focus on stereo matching in this paper. After years of development some achievements have been made in stereo matching, but some problems remain unsolved. The two most important things concerned in stereo vision trend to be contradictory: accuracy and efficiency. This paper presents a method called Advanced-Census which is a good combination of SAD and Census. When finding corresponding pixel in left and right image, SAD can get high accuracy with low speed and Census have the opposite result. Advanced-Census has advantages of both SAD and Census. It retains the speed of Census while having the accuracy of SAD. Although it has the speed of Census but not enough for practical applications, so we speed up Advanced-Census using multi-thread technique and edge detection. After speeding up, Advanced-Census gets nearly real-time performance.

Implementation of neutralizing fields for particle–particle simulations using like charges

2021 ◽  
Vol 87 (3) ◽  
Author(s):  
Yinjian Zhao ◽  
Chen Cui ◽  
Yanan Zhang ◽  
Yuan Hu
Keyword(s):  
Analytical Solution ◽  
Computational Cost ◽  
High Accuracy ◽  
Ion Temperature ◽  
Coulomb Collisions ◽  
Relaxation Problem ◽  
Temperature Relaxation ◽  
Particle Simulations ◽  
Speed Up ◽  
State Of Art

The particle–particle (PP) model has a growing number of applications in plasma simulations, because of its high accuracy of solving Coulomb collisions. One of the main issues restricting the practical use of the PP model is its large computational cost, which is now becoming acceptable thanks to state-of-art parallel computing techniques. Another issue is the singularity that occurs when two particles are too close. The most effective approach of avoiding the singularity would be to simulate particles with only like charges plus a neutralizing field, such that the short-range collisions are equivalent to those of using unlike charges. In this paper, we introduce a way of adding the neutralizing field by using the analytical solution of the electric field in the domain filled with uniformly distributed charges, for applications with homogeneous and quasi-neutral plasmas under a reflective boundary condition. Two most common Cartesian domain geometries, cubic and spherical, are considered. The model is verified by comparing simulation results with an analytical solution of an electron–ion temperature relaxation problem, and a corresponding simulation using unlike charges. In addition, it is found that a PP simulation using like charges can achieve a significant speed-up of 100 compared with a corresponding simulation using unlike charges, due to the capability of using larger time steps while maintaining the same energy conservation.

The Performance Analysis and Comparison of the Simplified Method of the Doubly-Fed Wind Generator Model

2013 ◽  
Vol 411-414 ◽  
pp. 1831-1835 ◽  
Author(s):  
Zhi Yan Li ◽  
Qi Hui Liu ◽  
Xiao Ming Wang
Keyword(s):  
Large Scale ◽  
Operating Performance ◽  
High Accuracy ◽  
Doubly Fed Induction Generator ◽  
Simulation Efficiency ◽  
Generator Model ◽  
Simplified Method ◽  
Speed Up ◽  
Doubly Fed ◽  
Control Principle

This article puts forward a simplified method of the doubly-fed induction generator (DFIG for short) model innovatively based on the DFIG control principle. This paper establishes two different degrees of simplification models with equivalence of the converter model and the generator model respectively, and compares the operating performance of each model under the same operating condition. The simulation results show that, the two simplified models have high accuracy, and speed up the simulation efficiency effectively, which is suitable for the study of large-scale wind power generation in power system.

Research and Implementation on Face Recognition Algorithm

2012 ◽  
Vol 151 ◽  
pp. 657-660
Author(s):  
Xiao Mei Ning ◽  
Yan Shi
Keyword(s):  
Face Recognition ◽  
Linear Transformation ◽  
Euclidean Distance ◽  
Recognition System ◽  
High Accuracy ◽  
Recognition Algorithm ◽  
Face Recognition System ◽  
Speed Up ◽  
Simulation Results ◽  
Low Dimensional

This paper used the PCA based method to build a face recognition system. At first, PAC provided a linear transformation matrix between the high and low dimensional spaces, then multi-dimensional Euclidean distance were used to rebuild residual to reduce the number of the obtained dimensions, speed up and high accuracy, which simplified the process of face recognition. Simulation results proved this algorithm is efficient.

scholarly journals ESPECIALLY GEODETIC MONITORING WHEN SURVEILLANCE OF BUILDINGS AND CONSTRUCTIONS OF ENERGY FACILITIES DURING THE CONSTRUCTION PERIOD

2019 ◽  
Vol 1 (2) ◽  
pp. 39-47
Author(s):  
Serik Seksembaev ◽  
Natalia Kobeleva ◽  
Anton Nikonov
Keyword(s):  
Power Plants ◽  
Thermal Power ◽  
High Accuracy ◽  
Safe Operation ◽  
Construction Period ◽  
Environmental Disasters ◽  
Industrial Enterprises ◽  
Field Control ◽  
Speed Up ◽  
Under Construction

Observations of the precipitation of buildings and structures of industrial enterprises are part of geodetic monitoring, which is carried out both during the construction of the facility and during the operational period. Systematic geodetic measurements in the framework of monitoring is of great importance to ensure the safe operation of the engineering facility and in some cases can pre-vent man-made accidents, environmental disasters. The article presents the results of determining the absolute values of sediment foundations of buildings, structures and equipment of one of the thermal power plants under construction. The peculiarity of the measurements on the object is the use of fan-shaped trigonometric leveling, which allowed to significantly speed up the implementa-tion of the field stage of work. Leveling was performed by electronic total station Leica TS-06 with the accuracy of measuring the vertical angles of 2ʹʹ. Residuals of closed polygons and field control measurements at the station showed high accuracy of trigonometric leveling, comparable with geo-metric leveling class II.

Design and Experiment of Ultrasonic Anemometer Using TDC-GP2 Time-to-Digital Converter

2014 ◽  
Vol 609-610 ◽  
pp. 1002-1007
Author(s):  
Han Yu Du ◽  
Ming Qin
Keyword(s):  
Measurement Error ◽  
Wind Speed ◽  
Wind Tunnel ◽  
High Accuracy ◽  
Time Measurement ◽  
Digital Converter ◽  
Ultrasonic Anemometer ◽  
Speed Up ◽  
Software And Hardware ◽  
Test Result

An ultrasonic anemometer using TDC-GP2 high-accuracy time measuring chip is studied in this paper, the design of software and hardware parts and principle of flying time measurement using TDC-GP2 chip are also discussed in detail. Under this scheme, a prototype has been fabricated, with simpler circuit architecture. A wind tunnel experiment of the prototype was conducted, the test result shows that the ultrasonic anemometer can measure wind speed up to 39.9m/s, and the measurement error does not exceed 0.3m/s when wind speed is below 10m/s, or 3 percentage when its above 10m/s. Wind directions vary from 0 to 359 degree can be detected with accuracy of 3 degree.

A New 3D Medical Data Field Segmentation Algorithm Based on Improved K_Means Clustering

2010 ◽  
Vol 108-111 ◽  
pp. 69-73
Author(s):  
Xin Wu Li
Keyword(s):  
High Accuracy ◽  
Medical Data ◽  
Segmentation Algorithm ◽  
Seed Selection ◽  
Clustering Techniques ◽  
Research Fields ◽  
Data Field ◽  
Speed Up ◽  
3D Volume ◽  
Process Speed

Direct 3D volume segmentation is one of the difficult and hot research fields in 3D medical data field processing. Using K-means clustering techniques, a new clustering segmentation algorithm is presented. Firstly, According to the physical means of the medical data, the data field is preprocessed to speed up succeed processing. Secondly, the paper deduces and analyzes the clustering and segmentation algorithm and presents some methods to increase the process speed, including improving cluster seed selection, improving calculation flow, and amending pixel processing and operational principle of algorithm. Finally, the experimental results show that the algorithm has high accuracy when used to segment 3D medical tissue and can improve process speed greatly.

scholarly journals Deep Learning for Detection of Visible Land Boundaries from UAV Imagery

2021 ◽  
Vol 13 (11) ◽  
pp. 2077
Author(s):  
Bujar Fetai ◽  
Matej Račič ◽  
Anka Lisec
Keyword(s):  
Deep Learning ◽  
High Accuracy ◽  
Commercial Software ◽  
Advantages And Disadvantages ◽  
Cadastral Maps ◽  
First Case ◽  
Complex Approach ◽  
Aerial Vehicle ◽  
Speed Up ◽  
Boundary Mapping

Current efforts aim to accelerate cadastral mapping through innovative and automated approaches and can be used to both create and update cadastral maps. This research aims to automate the detection of visible land boundaries from unmanned aerial vehicle (UAV) imagery using deep learning. In addition, we wanted to evaluate the advantages and disadvantages of programming-based deep learning compared to commercial software-based deep learning. For the first case, we used the convolutional neural network U-Net, implemented in Keras, written in Python using the TensorFlow library. For commercial software-based deep learning, we used ENVINet5. UAV imageries from different areas were used to train the U-Net model, which was performed in Google Collaboratory and tested in the study area in Odranci, Slovenia. The results were compared with the results of ENVINet5 using the same datasets. The results showed that both models achieved an overall accuracy of over 95 %. The high accuracy is due to the problem of unbalanced classes, which is usually present in boundary detection tasks. U-Net provided a recall of 0.35 and a precision of 0.68 when the threshold was set to 0.5. A threshold can be viewed as a tool for filtering predicted boundary maps and balancing recall and precision. For equitable comparison with ENVINet5, the threshold was increased. U-Net provided more balanced results, a recall of 0.65 and a precision of 0.41, compared to ENVINet5 recall of 0.84 and a precision of 0.35. Programming-based deep learning provides a more flexible yet complex approach to boundary mapping than software-based, which is rigid and does not require programming. The predicted visible land boundaries can be used both to speed up the creation of cadastral maps and to automate the revision of existing cadastral maps and define areas where updates are needed. The predicted boundaries cannot be considered final at this stage but can be used as preliminary cadastral boundaries.

Sign in / Sign up

Export Citation Format

Share Document