Search results

PCA is the second order statistical properties for data.
In order to use all of the information, this article analysis the B-scan data (B-scan is composed of many A-scan data).
Let us assume the B-scan data is given in Eq. 1
Therefore, dimensionality reduction can be implemented
Lastly, the B-scan data can be reconstructed by Eq.6

These days, the rough set is usually used to deal with mass decision-making data.
It can make attribute reduction for large numbers of data to find the foundation.
If B is an attribute reduction set, then.
Experiments and discussion The data sheet in literature [4] is a table for fault decision-making for a certain weapon system.
There are about six decision-making reduction sets. .

The high reduction ratio caused the dynamic recrystallization during annealing of the rolled sheet.
At the 30% reduction ratio, many twins and dynamic recrystallized (DRXed) grains were observed.
On the other hand, the numbers of twins were greatly reduced, when the sheet were rolled at the 50% reduction ratio, although twins and DRXed grains still co-existed similar to the microstructure at the 30% reduction ratio.
At the 30 and 50% reduction ratios after annealing, the existing twins were fully recrystallized.
(3) In the verification of the role of strain rate in the plane strain sheets using the commercial FEM package DEFORMTM-2D, the result of the FEM simulation calculation was similar to the experimental data.

The experimental data show that: The yield ratio of the Q460C was found to increase with the elongation at break A and the reduction of area Z decreasing while the plasticity and toughness decrease as the testing temperature drops.
The Comparison of Strength Index of Experimental Data with Parameter Fitting.
But it is great deviated from fitting data below -20°C.
The experimental data is as following fig.8.
(2) In comparison with the data of fitting formula, the experimental data is more adverse to the application of Q460C steel.

The solid state reduction of hematite to Fe3O4 and FeO has been observed after 120 hours of MA by a planetary ball mill.
In this way, we can point out what is the uniqueness in the solid state reduction by MA in the Fe2O3-C system.
Furthermore, it should be noticed that the magnetite is already produced after 20 h of milling as is clear from the saturation magnetization, which is invisible by X-ray diffraction data shown in Fig. 3.
Hence, it turns out that the magnetic data can supply a better understanding of the amount of magnetite phase produced by ball milling.
Conclusions The present work evidently shows that a solid state reduction is occurred by MA using a high energy mill in Fe2O3-C system, despite the fact that the chemical reduction of Fe2O3 to Fe3O4 and FeO by graphite is characterized by a positive reaction heat.

The whole process includes data capturing, noise reduction, sub-sampling and surface reconstruction.
The whole process includes data capturing, noise reduction, sub-sampling and surface reconstruction.
Data Preprocessing Data Registering.
Therefore, data filtering should be indispensable in point cloud data processing, which can improve work efficiency greatly by reducing a large number of redundant data.
Experiment Data Acquisition and pre-processing.

The material properties of core material shown in Table 2 have been taken from manufacturer's data.
The data shows a little initial rise in force with increasing impact energy before reaching a plateau and is independent of core thickness.
The trends in the maximum force data, Pmax in sandwich structures, Fig. 4, are similar to those of incipient load.
The residual strength data, normalized by the strength of the unimpacted sandwich panels, are plotted against impact energy in Fig. 8.
And the strength reduction behavior is influeced by thickness of core material.

Metal thermic reduction of oxides is an ideal way of producing non-oxide or oxide/non-oxide refractory.
TiN/Al2O3 composite material prepared by aluminothermic reduction in coke bed is a new method.
The transformation of main phases in the products reveals that aluminothermic reduction and nitridation reaction occur, it also indicates that temperature can improve aluminothermic reduction and nitridation reaction.
Lattice parameters of the synthesized TiN were calculated from the measured d-spacing data derived from XRD peaks.
It means that aluminothermic reduction and nitridation reaction can occur and gain TiN products.

Next, taking the achievements of students who are in the first semester of the freshman year (data of four-term students with the same course types and numbers; the corresponding relation of course and credit is shown in Table 1) and studying in mathematics major of Beijing University of Science and Technology (USTB) as an example, the simulation is carried out with the above method.
Taking the achievements of students in first semester of the freshman year and USTB mathematics major as an example, this paper evaluates the mean vector and covariance matrix of  multi-dimensional normal distribution with historical data (see Table 2 and 3).
Finally, the total credits achieved by students, the distribution law and cumulative distribution law, and the deviation with actual data can be worked out, as shown in Table 6, Fig. 2 andFig. 3 are the comparison of two distribution law curves and two cumulative distribution law curves.
Therefore, this paper adopts control variable method, which has higher efficiency in variance reduction techniques, to conduct variance reduction so as to reduce the random fluctuation.
The credit value found by several groups of simulation has relatively big fluctuation, for which variance reduction can be carried out through adopting the variable control method in variance reduction techniques.

The data outside intrinsic dimension space will be regarded as noise signal to be eliminated.
Finanly the process of global array by affine transformation will be replaced by mean reconstruction,the data after denoise may be obtained in the inverse process of the phase space reconstruction.
The data outside intrinsic dimension space will be regarded as noise signal to be eliminated.
Finanly the process of global array by affine transformation will be replaced by mean reconstruction,the data after denoise may be obtained in the inverse process of the phase space reconstruction.
In this experiment, vibration acceleration data with color noise is collected from cantilever fan in an aluminum company.