Search results

Furthermore, a measuring system for segmented SOFC cells has been developed allowing for the spatially resolved characterisation of cells in terms of current density/voltage characteristics, impedance spectroscopy data, operating temperature and gas composition.
In order to determine the temperature at each segment allowing for correlating experimental data such as voltage, current density and impedance data with the local temperatures, thermocouples are introduced in the metallic segments.
As a consequence, the overpotential is mainly related to the cathode and further cell improvement should concentrate on improving performance of the oxygen reduction reaction.
The obtained data can be used to avoid critical temperature gradients during operation and also to optimise functional layers to homogenise current density and heat production especially in a fuel cell stack.
It seems that the increasing conversion rate at the segments 9-11 causes performance reduction at segment 12 that could result in critical conditions and hence enhanced corrosion at the fuel exit.

For this reason the method has a poor noise reduction ability, in addition the imaging system itself is disturbed by various of noise inevitably, so it is necessary to choose suitable methods to do de-blurring and de-noising [5].
The imaging results of experiment data show that the modified algorithm proposed in this paper improves the performance of RL algorithm, and it can reduces the influence of noise well.
The enhancement of the sparsity is achieved by grouping similar 2D fragments of the image into 3D data arrays, the arrays are called ‘‘groups’’.
The original image, the original image spectrum, the reduction clarity image with noise, and the reduction clarity image with spectrum noise are shown in Fig. 1(a) to Fig. 1(h).
The second experiment is an image with complex scene, and the original image, the original image spectrum, the reduction clarity image with noise, and the reduction clarity image with spectrum noise are shown in Fig. 2(a) to Fig. 2(h).

This energy reduction is a result of the avoidance of bulk melting commonly attributed with various arc welding methods.
Results 3.1 Comparison between Ceramic and Steel Tools (Aluminum 6061-T651 similar welds) The resulting tensile testing data indicates a minor strength difference between joints welded using either tool.
The similarity between microstructure properties of the specimens is supported by the hardness data shown in Fig. 6.
This data reveals that no difference in hardness exists between tool use.
Fig. 6: Tool comparison hardness data 3.2 Similar Aluminum 6061-T651 Weld The joint strength results of EAFSW of aluminum to aluminum are shown in Table 4.

In the next step, detailed material characterization of the multilayered sheet metal in combination with the short-term heat treatment is required as input data for the finite element analysis of forming processes.
Tensile tests and layer compression tests were carried out for the evaluation of material data.
A reduction of the true stress of about 75 MPa can be achieved.
In the next step, the evaluated material data is used as input data for the finite element analysis of a deep drawing process and a subsequent validation of simulation and experiments.
In this context, true stress – strain curves and yield loci determined by means of tensile tests and layer compression tests were established as input data for the finite element analysis of forming operations.

In the commonly seen dimension reduction methods, PCA[1], LDA[2], Locality Preserving Projection (LPP)[3] and Orthogonal Local Preserving Projection (OLPP)[4], etc.
LPP constructs a nearest-neighbor graph for each data point, and it remains the manifold structure information among data; OLPP constructs its feature space bases with orthogonal characteristic so that it keeps the manifold structure of the raw data more completely.
GDA is a nonlinear discriminant analysis using Kernel function, and it maps the raw data vector into high dimension feature space, and in this high dimension feature space, the data distribution is formed to be linearly divisible, meanwhile, it uses LDA conversion to find out the projection vectors having the largest ratio of between-class variation and within-class variation.
In the traditional dimension reduction method (for example, PCA, LDA and GDA), the entire structure of a human face is considered, and through linear statistical analysis, the dimension of the raw data is reduced to perform classificaiton and recognition.
Therefore, the proposed FPBR approach accompanied with traditional statistical dimension reduction algorithm can indeed enhance the correctness rate of face recognition effectively.

These results were validated by the comparison with the data obtained from the first experimental campaign.
Results in terms of strain distribution, evaluated by a Digital Image Correlation (DIC) system, were compared with data from the first experimental campaign in order to find a correlation between the major strain values along the longitudinal axis of the specimen and the mechanical properties at different temperature levels.
In the following sections results in terms of Yield Stress (YS), Ultimate Tensile Stress (UTS) and Uniform Elongation (UE) have been shown (data are average values of two results).
A further increment of temperature leads to an early reduction of the material strength (overaging).
Figure 6: Major strain profiles at 50% of the necking strain and at the necking stage Using local data from the DIC system during tensile tests, the major strain values at both stages were correlated to the local temperature values the specimen experienced during the heat treatment by Gleeble.

Based on 33-year typhoon information of South China Sea (SCS) in 1980-2012 and NCEP/NCAR reanalysis data, taking Climatology and Persistence (CLIPER) and earlier physical quantities predictors selected by Stepwise Regression (SWR) and Multidimensional Scaling (MDS) methods as model inputs, the Genetic Algorithm-Artificial Neural Network (GA-ANN) forecast model was built for typhoon gale.
Neural network ensemble prediction modeling for typhoon gale and results analysis 3.1 Data and prediction objects The 33-year tropical cyclone tracks dataset (CMA-STI) in 1980-2012 were provided by Shanghai Institute of Typhoon, and reanalysis data of element fields and physical diagnose data for each day were from NCEP/NCAR(the spatial resolution is 2.5°×2.5°and temporal resolution is six hours).
Taking tropical cyclone centers as the datum marks, surface wind velocities were located at 6×6 grid points around the datum marks (as shown in Fig.1, the positions of wind velocities at 36 lattice points move as tropical cyclone centers move).
The input nodes of GA-ANN were the same with regression models, except for adding 2 manifold learning factors by MDS dimension reduction process. 36 GA-ANN 24h prediction models for wind velocities were built.
Nonlinear Dimensionality Reduction by Locally Linear Embedding[J]．Science, 2000, 290,(5500):2323-2326

The experimental results indicate that a proper controlled reduction of the Ce-TZP phase allows a modification of the toughness of the Ce-TZP phase.
An interesting phenomenon for Ce-TZP ceramics is the possible reduction of Ce +4 to Ce +3 in a non-oxidizing atmosphere at elevated temperatures.
In most cases however, the reduction causes destabilization of the tetragonal zirconia (t-ZrO2) phase [17,19,22].
The Young's modulus, needed for the calculation of toughness from the indent-induced crack length, was obtained by the linear rule of mixtures with a data setting of 200 GPa for Ce-TZP and 380 GPa for Al2O3.
In this context, an optimum toughness increase of a Ce-TZP without spontaneous reduction was reached after 20 minutes of sintering or annealing at 1450°C [26].

At present, the most of studies on resource allocation mainly focus on improving the overall performance by balancing the load of data center.
At present, the most of studies on resource allocation mainly focus on improving the overall performance by balancing the load of data center, but they didn’t give full consideration to how to reduce energy consumption.
By estimation, China Data Center consumed 7 million KWh power in 2011 which takes up 5% of China’s total power consumption.
The increase of resource utilization rate both saves energy and reduces emission of CO2 such that it contributes to reduction of greenhouse effect, be significant to protection of the whole ecological environment and conforms to present idea of “Green IT” [5,6].
The performance of proposed method is verified by the mode of integrating theoretical, simulation and test analysis and further optimized on the basis of analysis simulation and test data for providing high resource utilization rate and energy-efficient data center and laying basis for design of allocation of resources at upper level.

The annealing results reveal that annealing temperatures up to 598 K make the carrier concentration lower due to the reduction of donor-type defects caused by ion implantation and the acceptor nature of nitrogen.
The activity E0 can be calculated from the data in Fig. 1 and it is about 0.26 ±0.1 eV which is similar to that of intrinsic InSb [19].
The Hall mobility data of the 3 samples are shown in Fig. 3.
The mobility data for the sample after being annealed at different temperatures are shown in Fig. 5.
Annealing at temperature up to 598 K results in lower carrier concentration due to the reduction of donor-type defects and the acceptor nature of nitrogen.