Search results

Medium and Long Term Power Load Forecasting Model Is the essence of power load forecasting using the existing historical load data and related information, in a certain way to identify changes in the law of power load in order to predict future changes in electrical loaded.
Grey theory is based on the amount of random quantity as gray variation within a certain range, and the law is not obvious raw data generated by the accumulation to become law with a certain number of columns generated model equations constructed and bleaching process.
Vocabulary A ^ = [a ^, u ^] T, and a ^, u ^ are a and u predicted value, the parameter A ^ is determined by the least square method: (5) Where, B and Yn is a known quantity, A ^ parameters to be determined, and B and Yn obtained by the following two equations: (6) (7) To the obtained a ^, u ^, Back substituting equation (4), with (8) Available again after discrete GM (1,1) model to generate a gray sequence (9) (2) New Information replacement projections Improved optimization for GM (1,1) model, although as long-term forecasts, but the real practical significance and high accuracy predictive value is only the most recent one or two data, and other data only further reaction trend said the plan values ​​or value (long-term
planning of the forecast), so the model will predict new information obtained in each column of data into the same time, they remove one of the most stale data, namely (10) This metabolic data processing dimension and new information technologies is the approach the prediction accuracy is significantly improved.
This paper proposes a new portfolio optimization forecasting model, its basic idea is to improve the original sequence given gray prediction residuals, corrected for residual prediction model sequences were obtained before treatment and other reformers to improve the residual interest gray forecasting model which combines the advantages of improved prediction residuals gray prediction model and so on Innovation Gray predicted with a higher accuracy than a single prediction model their application as follows : 1 ) Accumulating the raw data processing ; 2 ) For processing data accumulated column GM (1,1) model to predict ; 3 ) Reduction of the prediction process ; 4 ) Experience poor after verification, such as the accuracy requirements are not satisfied, then take the local residual build residual data sequence ; 5 ) Residual data columns then create GM (1,1) model ; 6 ) Amended the original gray model with residual gray model ; 7 ) Re predictive analysis.

The gray prediction model is mainly gray system theory, data from the columns to create dynamic models, which, GM (1, 1) is the most common practical application of a gray model, is the core model of gray system theory.
GM (1, 1) model theory Grey identification system through the development trend of the different factors the degree of processing the raw data generated to find the variation of the system to generate a strong regular data sequence, and then create the appropriate differential equation model to predict things future trends.
GM (1, 1) model and model checking Data Processing In this paper, the original sequence data (million kWh conversion units) is from the China Statistical Yearbook 2001-2010, electricity consumption in Shandong Province, which used for gray prediction.
Data processing is as follows: The original data series: Accumulation of once to generate the data column is: Smoothness and exponential test The gray GM (1,1) model for the exponential equation in real terms, requiring the sample data should be used to predict exponential law, therefore, to conduct regular inspection sequence.
Which Solving was Therefore, GM (1, 1) model of the corresponding sequence of time Will get a and b are the values into the formula, Shandong Province, electricity consumption by the prediction model: According to the type, the data obtained can be reduced through the cumulative reduction the predicted value of the original data: The prediction results in Table 1.

Test results show that (1) BFRP sheet perform better than CFRP or GFRP sheets under high freeze-thaw cycles; (2) exposed hybrid FRP sheets not only show very little loss in mechanical properties, but also contribute to the stability of test data; (3) mechanical properties of rein epoxy decrease significantly with increasing freeze-thaw cycles.
Each group contains 7 specimens to insure that a minimum of five valid data is available for the statistic analysis.
Test data of GFRP sheet exposed to the same freeze-thaw cycling condition from reference [8] was added to compare the durability behavior of three different FRP types, as shown in fig. 2.
About 10% reduction was found in ultimate strain of 50 and 100 cycles exposed 1C2B specimens, but there were nearly no decreases in 150 and 200 cycles exposed specimens. 1C1B performed better than 1C2B FRP sheets under freeze-thaw cycling condition.Fig.4 summarizes COV values of different FRP sheets.
Hybrid FRP sheets not only show very little loss in mechanical properties compared with homogeneous FRP sheets, but also contribute to the stability of test data after exposure to freeze-thaw cycling. 3.

The crystal structure of both samples has been solved by powder X-ray diffraction, data in the tetragonal space group I4/m (a= b= 5.55182 Å, c =7.86955 A0) for SrLaFeNi0.5W0.5O6sample and (a=b= 5.49129Å, c= 7.82233Å) for CaLaFeNi0.5W0.5O6 sample, and shows an almost perfect ordering between Ni2+ and W5+ cations at the B-site of the perovskite structure.
Results and Discussion Crystallographic data and structural aspects The prepared SrLaFeNi0.5W0.5O6 and CaLaFeNi0.5W0.5O6 perovskites show identical powder diagrams indicating the formation of a pair of isostructural materials.
The refined unit cell parameters, together with other relevant crystallographic data, are shown in Table 1 and the complete indexed powder diagram is presented in Table 2.
On the other hand, symmetry reduction observed in numerous of the investigated materials allows predicting additional spectral complexities [19].
Conclusions The Rietveld refinement of the XRD data show that both samples SrLaFeNi0.5W0.5O6 and CaLaFeNi0.5W0.5O6 are single phase without any detected secondary phase and their crystal structure is tetragonal unit cell with space group I4/m, but in SrLaFeNi0.5W0.5O6 shows that there is small impurity of SrWO4 and We observed that there is small difference in the lattice parameters due to ionic radius of the Sr and Ca in A site [20].

A dynamometer is set under the workpiece to measure three components of the cutting forces and the data is processed by Business software MATLAB.
Finally, we compared the test data to determine the role of the textured tool rake face in improving the friction properties Experimental procedures Preparation of carbide tools with micro-texture.
Cutting test schematic Cutting performance of newly developed cutting tool in dry cutting Data processing.
Then we calculated the value of the average friction coefficient μ in the tool-chip contact area with the processed data.
In the actual process of the titanium, increasing the feed rate and the depth of cutting are not effective methods to improve the friction properties, it is because that the reduction of friction coefficient is caused by increasing the cutting area, while larger main cutting force still makes worse process.

The SPSS19.0 software was used to analyse the data of enclosure experiments last for 7 days in summer of 2011.
Statistical Analysis Method Using SPSS19.0 software to analyze the experimental data.
Firstly, using Kendall's tau_b effective method to analyze the correlation of the 36 groups of data, for analyzing the main impact factors to the change of Coscinodiscus Jonesianus biomass qualitatively.
Secondly, to identify the factors affecting changes in biomass, according to the relevant data, statistical analysis, to establish the preliminary statistical model to obtain the degree of influence to Coscinodiscus Jonesianus quantitatively.
Jewson, Size reduction, reproductive strategy and the life cycle of a centric diatom.

The magnetic properties and their dependence upon temperature data are interpreted in terms of the Herzer random anisotropy model for nanocrystalline materials.
Saturation magnetization values were obtained by extrapolation of the magnetization data to high fields.
Table 1 shows the summary of MS and HC data obtained in the present work for nanocrystalline Ni-15%Fe samples in comparison with Ni samples.
It is seen that increasing temperature from 2 K to 298 K leads to a very small reduction in magnetization, what agrees well with other studies [12].
The magnetic property data are interpreted using Herzer’s random anisotropy model for nanocrystalline materials.

To evaluate the advantages or disadvantages of the proposed methodology, it has been applied to the data collected from a semiconductor manufacturing factory.
Reducing cycle time, producing high quality products, on-time delivery, continual reduction of costs and improving efficiency were considered as the most direct and effective ways.
To evaluate the advantages or disadvantages of the proposed methodology, it has been applied to the data collected from a semiconductor manufacturing factory.
Experimental Results and Discussion The proposed collaborative yield learning planning approach has been applied to the data collected from a semiconductor manufacturing factory.
To evaluate the advantages or disadvantages of the proposed collaborative yield learning planning approach, it has been applied to the data collected from a semiconductor manufacturing factory, and led to capacity re-allocation in the semiconductor manufacturing factory.

In the paper the topic is addressed using experimental data, showing the influence of the most important variables like materials properties, strain values, and variables like thickness.
The processing of all the data had two focal points: 1.
Comparing these data for the same combination but different pressures, showed the decrease of the wrinkle dimensions and changes in the ratio A/l (i.e. dA/dl).
The ratio dA/dl was used to address point 1, the data related to the average values of A or l were used to determine the pressure for which this particular combination becomes free of wrinkles (point 2).
As figure 5 shows, the test data indicate that the curves have a shape similar as for the “absolute” limit, although at lower values.

It has been found that the method for determining thermodynamic properties from the cross-section data allows to find the contribution of short-range ordering into the thermodynamic state of an imperfect alloy.
It follows from (5) that the sum of all thermodynamic factors of the g matrix is equal to the sum of the diagonal elements of the g matrix, that is, to the matrix's spur: 1 11 () () 1 SSSl kk il l kk k lk x ggTrg x= == ≠ += − ∑∑∑ (8) In binary solutions the increase of the first-order interaction's potential energy for one component causes exactly the same reduction of the first-order interaction's potential energy for the other component [2].
As the results of calculations performed with the available data on a number of alloys like nickel-palladium [2], molybdenum-tungsten [4], gold-copper, gold-silver, gold-nickel, coppernickel [5], gallium-indium-cadmium [6], nickel-rhenium-molybdenum [7], and lead-indiumantimony [8] show, the expressions (4), (7) and (8) are satisfactory implemented within the experimental data repeatability margin.
(17) As the results of calculations performed for triple systems like gallium-indium-cadmium [6], nickel-rhenium-molybdenum [7], and lead-indium-antimony [8] show, the expressions (16) and (17) are satisfactory implemented within the experimental data repeatability margin.
Activity coefficients The results represented in [1,4,7,12,13,14] facilitate to formally generalize the data on thermodynamic coefficients to apply them to components' chemical potentials using the expression [2] , kk kl l Tp x g kT x µ∂ =  ∂ , where klg is the partial thermodynamic factor; kµ is the chemical potential of the k -th component; lx is the molar concentration.