Sort by:
Publication Type:
Open access:
Publication Date:
Periodicals:
Search results
Online since: September 2005
Authors: Keith Worden, Gareth Pierce, Graeme Manson, Daley Chetwynd
Two-class Gaussian data
For the purposes of investigating the use of machine learning techniques with uncertain-valued weights for data classification, synthetic two-class data sets were constructed.
Fig. 1 shows the training data set.
The Bayesian classification rate for the training data set was 81.6%, whilst those for the validation and testing data sets were 80.7% and 84.2% respectively.
Fig. 1. 1000 point training data set showing Fig. 2.
The goal in both cases is to improve the robustness of the classification capability of the network via a reduction in misclassification rate.
Fig. 1 shows the training data set.
The Bayesian classification rate for the training data set was 81.6%, whilst those for the validation and testing data sets were 80.7% and 84.2% respectively.
Fig. 1. 1000 point training data set showing Fig. 2.
The goal in both cases is to improve the robustness of the classification capability of the network via a reduction in misclassification rate.
Online since: March 2011
Authors: Bo Wang, Yong Wei Wu, Wei Min Zheng
Each data package possesses a header which guides the receiver to take appropriate measure.
Data transferring is affected by switches or routers which locate among web cache, web server or clients.
When cache server possesses what clients request, the clients can directly acquire data without contacting the web sever.
The directive can be expired data, cache control or content-length.
This reflects that if the data element in heap can keep close relation with the users' visiting action.
Data transferring is affected by switches or routers which locate among web cache, web server or clients.
When cache server possesses what clients request, the clients can directly acquire data without contacting the web sever.
The directive can be expired data, cache control or content-length.
This reflects that if the data element in heap can keep close relation with the users' visiting action.
Online since: September 2008
Authors: José Rodellar, Fayçal Ikhouane, Francesc Pozo, Gisela Pujol, Leonardo Acho, Josep M. Rossell
In this respect, it may be sufficient to characterize the dynamics of
some materials or devices by models in the form of differential or difference equations, transfer functions or black box operators, which are obtained through matching of experimental inputoutput data.
Practical advantages of these schemes are in the reduction of the transmission costs within the control loops, the increasing of reliability of the control system in case of failures in sensors, actuators or controllers, the reduction of the overall computational effort and the possibility for parallel realtime implementation.
From hardware point of view, integration of the data acquisition board (for sensors readings and for energy supply to the actuators) and the mathematical processor unit, was a challenge too.
Moreover, with the high capacity of modern digital processors, additional benefits can be added to the system, such as monitoring, statistical data analysis and on-line parameter identification for self adjusting controllers.
In fact, nowadays, more complex control algorithms are possible to be implemented in digital platforms with a single board integrating the data acquisition board and the processor unit.
Practical advantages of these schemes are in the reduction of the transmission costs within the control loops, the increasing of reliability of the control system in case of failures in sensors, actuators or controllers, the reduction of the overall computational effort and the possibility for parallel realtime implementation.
From hardware point of view, integration of the data acquisition board (for sensors readings and for energy supply to the actuators) and the mathematical processor unit, was a challenge too.
Moreover, with the high capacity of modern digital processors, additional benefits can be added to the system, such as monitoring, statistical data analysis and on-line parameter identification for self adjusting controllers.
In fact, nowadays, more complex control algorithms are possible to be implemented in digital platforms with a single board integrating the data acquisition board and the processor unit.
Online since: April 2016
Authors: Yue Sheng Chao, Yan Hui Zhang, Ying Jin
The results show that the structral relaxation , free volume migration and annihilation during the vacuum annealing will lead to the reduction of like-vacancy defects concentration and distortion stress.
The vabrating sample magnetometer(VSM) was used to measure the main soft properties of annealing specimens.The microstructural changes of specimens were analysed by Mössbauer spectrometer and tranmission electron microscopy(TEM).The g-radiation source is 57Co/pd, the data fitting of Mössbauer spectra adopted the MSU8 program.
Fig.2 show the Mössbauer spectra of amorphous Fe52Co34Hf7B6Cu1 specimens after magneto-pulsing and low-temperature annealing .Table 4 show the fitting results of Mössbauer spectra data.
The vabrating sample magnetometer(VSM) was used to measure the main soft properties of annealing specimens.The microstructural changes of specimens were analysed by Mössbauer spectrometer and tranmission electron microscopy(TEM).The g-radiation source is 57Co/pd, the data fitting of Mössbauer spectra adopted the MSU8 program.
Fig.2 show the Mössbauer spectra of amorphous Fe52Co34Hf7B6Cu1 specimens after magneto-pulsing and low-temperature annealing .Table 4 show the fitting results of Mössbauer spectra data.
Online since: March 2013
Authors: Yi Zhang, Jiang He
Based on analysis results of measurement data, comfortable thermal environments for arcaded pedestrian paces in the pedestrian streets of the Lingnan region were understood.
A round of field data measurement for five measurement points was required to be completed within one hour.
As can be seen from Fig. 6, points A3 and A5 each have two data where the globe temperature is higher than the air temperature.
It is discerned from the original recorded data that, these two abnormal data of point A3 were measured at 12:30 and 14:30 respectively.
It is thus clear that the abnormal data of points A3 and A5 are related to intense solar radiation at noon.
A round of field data measurement for five measurement points was required to be completed within one hour.
As can be seen from Fig. 6, points A3 and A5 each have two data where the globe temperature is higher than the air temperature.
It is discerned from the original recorded data that, these two abnormal data of point A3 were measured at 12:30 and 14:30 respectively.
It is thus clear that the abnormal data of points A3 and A5 are related to intense solar radiation at noon.
Online since: May 2011
Authors: Rong Yao, Dong Ling Peng, Hong Yuan Huang
The initial temperature rises rapidly, and along with fuel reduction, but drops gradually.
Table 3 Test sample compressive strength orthogonal data analysis K1 74.75 39.79 31.93 K2 -21.79 39.37 23.83 K3 -19.99 -13.45 -2.83 K4 -0.03 -32.77 -19.99 k1 18.69 9.95 7.98 k2 -5.45 9.84 5.96 k3 -4.10 -3.36 -0.71 k4 -0.01 -8.19 -4.10 range 24.14 18.14 12.98 scheme A2 B4 C4 Table 3 for the tentative data orthogonal analytical table, from the test result analysis, selected the 2nd level, the 4th level, the 4th level for the polypropylene textile fiber, the silica flour and the carbon fiber, the ratio of compressive strength loss is in a small scope.
Table 3 Test sample compressive strength orthogonal data analysis K1 74.75 39.79 31.93 K2 -21.79 39.37 23.83 K3 -19.99 -13.45 -2.83 K4 -0.03 -32.77 -19.99 k1 18.69 9.95 7.98 k2 -5.45 9.84 5.96 k3 -4.10 -3.36 -0.71 k4 -0.01 -8.19 -4.10 range 24.14 18.14 12.98 scheme A2 B4 C4 Table 3 for the tentative data orthogonal analytical table, from the test result analysis, selected the 2nd level, the 4th level, the 4th level for the polypropylene textile fiber, the silica flour and the carbon fiber, the ratio of compressive strength loss is in a small scope.
Online since: October 2011
Authors: Xiao Ming Fu
Using high-purity cobalt oxide or cobalt oxalate by hydrogen reduction and electrolytic cobalt powder to prepare metallic cobalt powder are two main methods for preparing cobalt powder, the processes of which have a wide range of applications in practice [5].
The data of the mass loss of the pyolytic decomposition of CoC2O4 . 2H2O in the argon gas is presented in Table 1 according to its TG-DSC curves.
Fig.1 The curve of TG-DSC of CoC2O4 . 2H2O from the room temperature to 500 °C in the argon gas (heating rate of 10 °C. min-1) Table 1 The data of the weight loss of the pyolytic decomposition of CoC2O4 . 2H2O in the argon gas Temperature / °C Thermal decomposition process Mass loss rate / % Real mass loss rate Theoretical mass loss rate 150~275 CoC2O4 . 2H2O→CoC2O4+2H2O 19.07 19.67 300~450 CoC2O4→Co+2CO2 47.60 48.09 Analysis of the morphologies of the pyrolysates of CoC2O4 . 2H2O SEM images of the pyrolysate of CoC2O4 . 2H2O at the different temperature for 10 min in the argon gas (heating rate of 10 °C. min-1) are shown in Fig. 2.
The data of the mass loss of the pyolytic decomposition of CoC2O4 . 2H2O in the argon gas is presented in Table 1 according to its TG-DSC curves.
Fig.1 The curve of TG-DSC of CoC2O4 . 2H2O from the room temperature to 500 °C in the argon gas (heating rate of 10 °C. min-1) Table 1 The data of the weight loss of the pyolytic decomposition of CoC2O4 . 2H2O in the argon gas Temperature / °C Thermal decomposition process Mass loss rate / % Real mass loss rate Theoretical mass loss rate 150~275 CoC2O4 . 2H2O→CoC2O4+2H2O 19.07 19.67 300~450 CoC2O4→Co+2CO2 47.60 48.09 Analysis of the morphologies of the pyrolysates of CoC2O4 . 2H2O SEM images of the pyrolysate of CoC2O4 . 2H2O at the different temperature for 10 min in the argon gas (heating rate of 10 °C. min-1) are shown in Fig. 2.
Online since: September 2013
Authors: Xin Le Zhang, Hai Cao, Xiao Hui Guo
Load cells and displacement gauges were used to measure the vertical load and displacement, and the data were recorded automatically by a data acquisition system.
The following conclusions can be drawn from Fig.1 - Fig.6: 1)At the same ingredient ratio and strain rate, the reduction of polymer addition can lead to a lower peak stress value and elastic modulus, but the residual loads are almost the same. 2)At the same strain rate, polymer-modified concretes (NC1, NC3andNC4) have obvious brittleness, especially NC1.
The following conclusions can be drawn from Fig.1 - Fig.6: 1)At the same ingredient ratio and strain rate, the reduction of polymer addition can lead to a lower peak stress value and elastic modulus, but the residual loads are almost the same. 2)At the same strain rate, polymer-modified concretes (NC1, NC3andNC4) have obvious brittleness, especially NC1.
Online since: February 2007
Authors: Sabine Begand, Thomas Oberbach, Wilfried Glien
The physicochemical data, mechanical strength and phase composition of this
ceramic were analysed.
Also in comparison to published data for Y-TZP or common ZTA-ceramic showed that our results concerning the initial monoclinic ZrO2 are in the same range below < 2 % [4,5].
After the test the remaining static strength was 70 kN, which corresponds to a reduction of approximately 18 %.
Also in comparison to published data for Y-TZP or common ZTA-ceramic showed that our results concerning the initial monoclinic ZrO2 are in the same range below < 2 % [4,5].
After the test the remaining static strength was 70 kN, which corresponds to a reduction of approximately 18 %.