Search results

The oxidation kinetic data were experimentally obtained at elevated temperature in dry O2/Ar gas and the temperature dependence of the reaction rates were determined with the Arrhenius-type equation.
In addition, the kinetic data, composition, and structure of oxide scales were examined by scanning electron microscopy(SEM), energy-dispersive X-ray spectroscopy(EDX), and X-ray diffraction(XRD) in an attempt to define the basic mechanism of Fe-36Ni oxidation and reduction. 2.
The specimen was reduced using pure hydrogen gas at each temperature of 450°C, 800°C, and 1000°C shown in Fig. 4 to fabricate metallic alloy powder and the reductive reaction might be taken place with 3 stage reactions as follows: The reduction ratio of the alloy powder increased with increasing temperature and finally reached up to 97% at 600°C for 80 minutes, and the measn particle size was found to be about 200μm.
Fig. 4 The reduction raio of Fe-36Ni oxide powder with time using pure hydrogen gas. 4.
The reduction ratio of metallic powder reached up to 97% and its mean particle size was obsered to be about 200μm.

Toward Sub-20 nm Magnetic Tunnel Junction for Embedded Cache Memory Toshihiro Sugii a *, Hideyuki Noshiro, Yuichi Yamazaki, Chikako Yoshida, Yoshihisa Iba Fujitsu Limited, 10-1 Morinosato-Wakamiya, Atsugi 243-0197, Japan asugii@jp.fujitsu.com Keywords: STT-MRAM, Magnetic Tunnel Junction, MTJ, Perpendicular, Cache memory, MgO, Shrink, Oxidation, Reduction, Plasma treatment Abstract.
From this figure, we found 1) longer O2 plasma treatment time shifts the cumulative probability data to higher resistance, which means further reduction in the electrical MTJ size, 2) the slopes in the cumulative probability data do not change between them, which means the variation in the MTJ size does not increase during the shrink process, 3) the shift in 1/(Rp)1/2 is proportional to the treatment time, which means the possibility in further shrink, 4) NH3 plasma treatment results in enlargement in the electrical MTJ size on the contrary.
From these results, NH3 plasma treatment results in the reduction of the sidewall surface and O2 plasma treatment results in the oxidation of it.
Owing to the size reduction, the current required for MTJ switching decreases with O2 plasma treatment time.
We clarified the mechanism of the shrink process as oxidation and reduction on the sidewall surface in MTJs, and optimized the process for higher density embedded MRAMs.

Fig. 5(b) shows the process to generate the function from the test data.
BOP shearing test data.
Illustrations of using shearing test data to develop the damage parameter function, f.
Illustrations of using shearing test data to develop the damage parameter function, f.
The predicted shearing pressure agrees with the test data, and max error is 11%.

Discretization of decision table is the important step for pretreatment of data mining and machine learning, which related to the effect of learning.
However, there are a lot of continuous database data, in order to obtain from the database which contains the continuous attribute gets better data sample and concise and effective decision rules, often need to continuous attribute discretization.
Therefore, decision-making system of continuous attributes discretization for data mining has the extremely vital significance.
Therefore, this follow-up work will study the algorithm for which the distribution of the data set to meet.
New Algorithm for Data Discretization based on Rough set.

Finally, the historical data is archived in order to access it at all times.
On the basis of this platform, the underlying data source of line loss management system comes from SCADA system, OMS system, electric energy management system, substation reading system and electric energy data acquire system.
The line loss management system established data association through a standard interface design, in order to clean and transform the data, transfer it to the Enterprise Service Bus, and supply subsystem extraction, selection.
(7) Data Integration It completes the grid management model, system management, system maintenance, data query and other various functions.
During the operation, the condition is good and the data is accurate.

The results showed reduction in the number of bubbles (as well as in void content) at frequencies between 10Hz and 50Hz for 30 minutes of vibrations.
After counting voids all data was categorized into 4 groups, depending on the size (radius) of the bubble cross-section.
Once again this showed no significant improvement in void reduction; see Fig. 2 (right).
The results presented in this paper show reduction in the number of bubbles (50-500 µm) at frequencies between 10Hz and 50Hz for 30 minutes.
The mechanical data acquisition and development program is continuing and further work on the vibration parameters and theirs impact on the process will be reported in our next publication.

The hardware consists of two linear and one rotary encoder, all sending the data to a counter/timer data acquisition board.
The goal of the work described in this paper was to develop a data acquisition system and the corresponding data processing software for measuring the standardised ship propeller’s parameters.
Data Processing Aligning the coordinates systems.
Scanned data processing.
Even if there are small chances for useful data to be gathered from small angular step measurements and one can think that these affected data may be ignored, further software developments will have to bring some improvements of this aspect for reducing void data propagation.

Physical and thermal properties of epoxy resin materials (Adapted from manufacturer data sheet).
Physical and thermal properties of reinforcing fiber (Adapted from manufacturer data sheet).
As specified in the data sheet, the individual cast of polymer composite was allowed to cure under a load of 40kg at room temperature for 24 hours before it is removed from the mould.
Statistical analysis shows that G10E30 offers 2.2% weight reduction when compared with G10E despite the fact that they contain same fibre contents.
Conclusion This experimental study has been able to offer weight reduction as evident in the panel weight analysis.

Currently, there are various methodologies for helicopter vibration reduction.
During the learning process, the thresholds and decay as follows: (24) where is the total number of the learning data. defines the desired accuracy of an output of GD-FNN, represents the predefined maximum error. and represent the smallest length of interest and the largest length of the input space, respectively.
For the th observation , calculate the out of GD-FNN of all the existing data , where and .
At each frequency, 500 observation data points are selected to train the networks and parameters of the GD-FNN.
Helicopter Vibration Reduction with Trailing-Edge Flaps.

In dealing with spatially incomplete situations, one usually applies either model reduction or modal expansion schemes.
Implementing the iterative Guyan model reduction with 5 iterations, one obtains Fig.3 (b).
In general, applying the CMCM method, together with the iterative Guyan reduction approach, could yield a better damage detection.
Baruch: Optimization Procedure to Correct Stiffness and Flexibility Matrices using Vibration Data.
Nagy: Improvement of a Large Analytical Model using Test Data.