Search results

The data in Figure 3 shows contact resistances in n+ and p+ contact chains while held continuously at 300°C for over 9000 hours.
The data shows a high level of stability with a total change in resistance of below 10%.
With a capacity of 5 off 16 pin DIL devices in a thermal cavity the system is used to collect significant amounts of ageing data.
The well-known SiC MOS transistor instability [2], due to near interface trapping is evident in this data.
The data in Figure 5 shows the leakage current in the offstate as a function of temperature between room 23°C and 400°C.

Several wrappers for the post-processing and data sorting are also made to perform the iterative analysis automatically caused by geometrical changes of transom support bracket attached in the bogie frame.
Data and output files are managed by EDM (Engineering Data Management) in this system.
Whole data used in the implementation are managed in the EDM as three schemes, i.e. project, job and task.
The NominalPost task converts the extracted values to specified input data for the task of NominalBFAP.
I-DEAS, ANSYS and BFAP, iterative analysis and data management can be performed automatically.

Of course, comprehensive data on weld residual stress profiles also deepen our understanding of welding processes, and so improve our capability to achieve high quality welds.
Basically, employment of the contour method involves four steps: specimen cut, contour measurement, data reduction and finite element analysis.
The remainder was valid for further data reduction.
A spline function was employed again to extrapolate the deleted measurements, aiming to generate full data covering the whole plane.
The results in the shaded area in Figure 5 cannot be relied on since it is derived from extrapolated data in the recut region, rather than actual measurements.

The data analysis varies depending on the excitation source and excitation type.
In combination with advanced data analysis methods of classical pulse thermography, the signal to noise ratio can be significantly increased.
Data Processing.
Further data reduction was realized by summarizing the amplitude information for defined time periods.
Figure 4: Data reduction using FFT and sum up of amplitude information to one single frame, which shows the total damaging map.

Section 2 describes model and data.
Methods and Data Preparation SDA Model.
Data Preperation.
Energy consumption data by sectors in 1997 and 2002 is obtained from [10], and we use the same method and net caloric value (NCV), combing with related data from China Energy Statistical Yearbook of 2008, to estimate the each sector’s energy consumption of 2007 in China.
The large reduction in energy consumption was mainly contributed by the reduction of energy intensity from 1997 to 2007, and it contributes -279.16%.

However, it could be used to achieve precise navigation (partially realized by precise data given from magnetometry), as well as the new space vision concerning magnetic mapping of the planet surface for mines-ores and the subsequent finding and exploitation of important sources of ores for the needs of Earth.
So, distance scaling could be implemented, by defining different techniques to use measurement data from multiple distances and extrapolate the magnetic field in larger distances.
Using the simulated data for the induced magnetic field and its gradients, also an inverse problem can be set, in order to identify the status of the magnetic sources as well as the positioning of the magnetometers.
In reality these field measurements, instead of the simulated data, can be used as input in inverse engineering techniques, which could assist in estimating the positioning of a magnetic mass.
For the verification of methods, modelling will be followed by magnetic signature studies, so as to allow the consideration of measurement data of magnetic field and its gradients as a primary data input.

XRD was used to identify the phases present after each calcination experiment (Philips ARD 1700, Cu Kα radiation) and the data was analysed using two types of software, 'PANalytical B.V., X'Pert Highscore', and 'Philips Analytical X- Ray ProFit'.
For 'standard' powders there was a progressive reduction in α from 1.32 at 850 o C to 0.53 at 950 °C.
Table 1 - ICP (powder) elemental analysis compared to calculated data.
The analytical data shows that control of lithium content poses the greatest processing problem at > 850 °C.
Analysis of the SEM data showed that both 'standard' and 'excess' powders displayed the cuboidal particles typical of many other niobate-based ceramics.

“Trial” is stand for the data from the wind tunnel, “protective” and “un-protective” is stand for the data from CFD in having and not having protective devices.
Fig. 6 Curves of Cd Fig. 7 Curves of Cl Fig. 8 Curves of Cmz We can see from the figure that the data from CFD tally well with the data from wind tunnel.
The drag coefficient’s reduction is because of the reduction of windward area.
“Trial” is stand for the data from the wind tunnel, “protective” and “un-protective” is stand for the data from CFD in having and not having protective devices.

The blanks were cold rolled with a reduction rate of 30%.
The blanks were cold rolled with a reduction ratio of 30% on a laboratory rolling mill "KVARTO" model K220/75-300.
On the basis of the obtained data, a plot of the coherent scattering areas size’ dependence on the annealing temperature was constructed (Fig. 3) Fig. 3.
According to the recrystallization diagrams [14], the grain size in the AlMg1 alloy after rolling with reduction of 30% in the cold state and subsequent annealing varies from 30 to 50 μm, for the AlMg5 alloy from 40 to 70 μm, for the AlMg6 alloy from 70 to 120 μm.
The annealing leads to a reduction in the tensile strength, yield stress of the investigated alloys AlMg1, AlMg5 and AlMg6 in the interval 380 ... 420 ° C 2.

Such kind of data shows a different pattern that the majority of data shows, so these data should be excluded from regular processing of data mining.
First, through the dimensionality reduction of the original data set, the execution time of detection algorithm could be lower.
Let d(o,o1) represent the distance between data o and data o1.
Otherwise, the data is considered as a normal data.
With a threshold T, the data set can be divided into two parts: normal data set and candidate outlier data set.