Search results

In this paper, a standardized method, based on a large number of measured data or Monte Carlo experiment to get the maximum distance or minimum distance, use the following formula to normalize.
Similarly, the presence of the simulator will be distributed in a plurality of spaced apart intervals, the interval needed for the reduction processing.
There are two main methods currently, expert advice and data analysis.
Data analysis is performed according to the logic and statistical analysis of the relationship between simulation data and indicators.

The natural disturbing fields are mainly caused by earth current field, fluctuations in the ionosphere, lightning activity, groundwater flow, oxidation reduction and so on.
When we were graphically fitting lines to the observed data, one of our criteria minimized the total sum of the absolute deviations between the data points and their corresponding points on the fitted line.
This criterion can be generalized: Given some function type and a collection of data points, minimize the sum of the absolute deviations.
Now the polynomial with order () is built in the local range of the () points: (6) The so-called “Least-Square approach” is to reduce the sum of squared deviation between value of polynomial and the observing value to a minimum, that is: (7) The sum of squared deviation is function of the parameters, because that of polynomial is known and unknown data re of the parameters of polynomial.

Oxygen radicals occlusion / release behavior of nanoporous aluminosilicate, Ca12Al14-XSiXO33+0.5X (0≦X≦4), synthesized under different condition was examined by the temperature programmed reduction (TPR) in an atmosphere of hydrogen in the temperature range of 200-1000˚C and temperature programmed oxidation (TPO) measurement at 800˚C.
X-ray powder diffraction data of samples were obtained using a Rigaku, RINT2000 diffractometer with Ni-filtered Cu Kα radiation (50 kV, 100 mA).
A comparison of data reveals that the amount of oxygen released in the α, and γ for all the sample are more or less same, around 3μg-O2/mg, and about 1μg-O2/mg, respectively.

With the reduction of Mo content, Mo metal phase as the continuous network structure is dispersedly distributed in the ceramic phase zone.
In order to ascertain the reliability of the data, the experiments were repeated three times.
With the reduction of Mo content, Mo metal phase as the continuous network structure is dispersedly distributed in the ceramic phase zone.

It was found that the single-step deformation of as-cast alloy via hot rolling at 350°C with a thickness reduction of 50% refined the alloy microstructure by creating deformation bands of the Mg(α) phase with a size of the order of tenths of micrometers.
The as-cast ingot was subsequently hot rolled at 350°C with a thickness reduction of 50% using the quarto-duo DW4-L rolling mill.
(a) Plot of liquid fraction versus temperature for the Mg-3%Zn alloy, obtained from heat flow versus temperature data, recorded with DSC; (b) microstructure of the Mg-3%Zn as-cast ingot, re-heated to 630°C and quenched in water Effect of hot rolling step Plastic deformation in the solid phase is one of many methods of feedstock preparation for light alloys as well as for steels [3, 17, 19].

GM (1,1) model Grey Series GM (1,1) model prediction is a realistic and dynamic analysis prediction mehod, if given a raw data sequence ,respectively from x sequence, select a different length of the data as a sub-sequence. for the sub-sequence building the GM (1,1) model. determined any sub-sequence data as (2) To the sub-sequence data for an accumulative generation,attained (3) Where ，.
After the repeated calculation ,the equation for exponential smoothing value (14) Where,—the starting year's corresponding raw data.
According to the historical data of agriculture machinery total power in Heilongjiang province from 1995 to 2008 show in Table I,using the equation (10) ~(14) to calculate,the prediction results shown in Table II.
Combination forecasting method based on rough set made weight coefficient transformed into the attribute importance evaluation of rough set , regarded the various forecasting methods composition set as a condition attribute of decision table ,took the observation value of forecast object as decision-making attributes, calculated the importance that the various attributes (forecast method) to the decision-making set (forcast indexes) ,finally ,determining the weights of the various forecasting methods according to the importance.The method completely analyzed the importance of different forecasting methods from the data analysis,overcome the subjectivity of combination forecasting method. 1) Build the relationship data model In order to apply rough set theory to determine the weight cofficients of combined forecasting model , we first need to establish the relational data model.Took the various forecasting methods as condition attributes, then the condition attribute set,took the
Today there are many mathematical methods on discretization of continuous data, the typical practical methods mainly were: the discrete approach based on hierarchical clustering method , the discrete method based on genetic algorithm, the discrete method based on conditional information entropy and based on self-organizing neural network SOM and so on[7].

For the fixed sampling parameters used in the special test system may cause the loss or redundancy of the data; and the simple data treatment is fixed, it is not easy to expand.
Tab 1 Data Acquisition Board Parameter PXI-6251- High-Speed M Series Multifunction DAQ Resolution 16-bit Analog inputs 32 channel with Data Transfer Rate 1.25 MS/s Analog outputs 2 channel with Data Transfer rate 2.5Ms/s Range 7 programmable input ranges(±100 mV to ±10 V) per channel 4.
LabVIEW integrates almost all of the features of hardware communications such as data acquisition board.
In the central processing system, the DAQ (Data Acquisition) system could acquire the electric signal data from the transducers according to the sample parameter.
The computer receives raw data though the DAQ device and the data was read and analyzed by the program.

Studies of electrochemical data indicate that, TCPE reduces MS corrosion through adsorption using a mixed inhibition mechanism.
The iron atoms oxidation and the reduction of H+ ions are both retarded down by an increase in the Rct value [38], where an increase also follows the increase in Rct in the value of inhibition efficiency and a lower capacitance with an upgrade in the concentration of extract [3, 39].
Parameter Electrochemical Concentration (%) 0.0 0.5 1.0 1.5 2.0 2.5 Rs (Ω) 34.0 23.0 23.3 22.2 32.4 18.1 Rct (Ωm2) 505 850 1680 2435 3259 3552 Cdl (µFcm2) 0.63 0.31 0.16 0.093 0.081 0.075 IE (%) 0.0 40.9 69.94 79.29 84.50 85.78 Water molecules on the steel surface are swapped out for TCPE molecules as a result of the reduction in capacitance value [40].
Adsorption Isotherm Analysis Using formula Eq. 1, the surface degree cover data from the weight loss data acquired from the initial experiment were used to characterize TCPE adsorption on the MS surface [14].
The adsorption data indicates the amount of metal coated by the corrosion inhibitor molecule rises with the quantity of the additional inhibitor.

FFNN Modeling: Data Preprocessing.
Outlier removal is performed on voltage data.
A small outlier of data can distort the whole result[19].
MAE is the average of the absolute values between the original data and the predicted data whereas MSE is .
The mathematical equation of the MAE value can be written as follows: MAE =1/n i=1n|yi-yi| (iii) MSE =1/n i=1n(yi-yi)2 (iv) where n is number of data, yi is actual data and y is predicted data [20] .

The experimental and theoretical data processing in through-diffusion methods have been described previously [16].
The measured (round) data at low concentration boundary are in agreement with the predicted (line) data from the through diffusion experiment, whereas the measured data (square) are systematically higher than predicted data at high concentration boundary.
The measured data is not agreement with predicted one.
GMZ bentonite experiments data were solid symbol (I[20], Tc[4, 22] Se[this work] and Re[15])and Kunigel-F bentonite experiments data were hollow symbol (Se[13], I[12] and Tc[12]).
Out-diffusion results of Re(VII) and Se(IV) showed a discrepancy between measured data and predicted data due to the heterogeneous porosity distribution in clay boundaries and species changed when the diffusion occurred in GMZ bentonite.