Search results

It realized optimum compression of fault sample data and enhanced classification speed and precision.
MODEL STRUCTRUE BASED ON THE PCA-BP NEURAL NETWORK Principal Component Analysis and the Steps in the Algorithm PCA[5-7] could simplify data structures in the precondition of minimizing loss of original data information, the linear combination for input data had caused big variance in order to have better ability to distinguish fault model.
Fig.2 The simulation calculation of 11x11 grounding grids Sample Data Reduction When analyzing grounding grids by electric network analysis, each conductor of grounding grids could be seen as a resistance.
Lastly ,the 12 lines and 33 columns sample data after normalization were obtained.
According to principle components score vectors (the input vectors of BP neural network) ,we could gain a array of 12x11new sample data,part of the sample data was shown by the Table2: Table2.

The results showed that indigenous LAB had more predominant effective species helping the induction of sugar but reduction in pH.
Additionally, slight reductions detected for alcohol production were observed after LAB and yeasts were both inoculated in the fermentation.
LAB isolates induced the sugar contents with high numbers of cells to 107-108 CFU/ mL during the fermentation days 3-7, and after that numbers of LAB isolates were declined to 103-104 CFU/ mL (data not shown).

Based on the microhardness and tensile test data, it was possible to verify a significant change over the mechanical properties of the UNS S31803 steel.
The precipitation of the σ phase is of high interest to the austenic, ferritic and duplex stainless steels, as effects which cause not only the reduction in the mechanical properties, but also the decrease in the resistance to steel corrosion are associated to it [ESCRIBA et al., 2006].
Besides, it is known that the sigma phase precipitation causes reduction in the corrosion resistance of this kind of stainless steel.
Conclusion Based on the results obtained, it is possible to conclude that the sigma phase precipitation significantly changed the mechanical properties of the steel being studied, as there was reduction in the ductility and increase of the hardness and maximum tension.

Main findings of the present paper conclude that industrial (product/ plant) safety, in its broadened sense, does embrace the gain/loss statistical data of a product's introductory versions and represents a trade–off function of its profits (resource renovation) and its losses (resources drain).
Damages due to user implementation follow a Statistical Process Control (SPC) which can be conventionally controlled within a tolerance field described by the Upper Control Limit (UCL) and the Lower Control Limit (LCL). 2.2 The FMEA Technique as Safety Assessment for Automobile In industrial engineering, FMEA is implemented to monitor the progressive improvement in a product's quality, or equivalently the progressive reduction in a product's imperfection (deficiency).
Thus, our analysis will rely mainly on data available at the end of this phase near year 1900. 2.3.2 Modeling of Disturbed Phase It starts near year 1900 (beginning of the 20th century).
Comparison of the two volumes indicates that the overall RPN shrinks to about 11% of its volume, reflecting proper actions taken with regard to reduction of CO emission in air.
1- In SPC technique, the lower control limit is considered as a standard target for expected future performance, and helps in monitoring the direct effects of product’s environmental safety. 2- Displaying of RPN in new graphical domains helps in monitoring the gradual changes in the direct effects of product’s environmental safety which focuses on the human discretion. 3- In Markov model, most aspects of the direct and indirect effect of a product become so controllable that a compromised state can be designed to balance both useful and harmful outcomes of a proposed product. 4- Markov model application is still systematic and straightforward when it is limited to only six categorized entities, and to assess direct, and indirect effects, and most aspects of the industrial safety (plant / product) for environmental systems, especially strategic industries. 5- The designer of a prospective mechanical product has to include similar safety instructions for the whole ECS to achieve a reduction

The first is to acquire a set of several CT images per slice through the volume; the second step is to resort the images and assemble spatio-temporally coherent data sets from the data acquired, which are most certainly shuffled and out of order [5].
Manifold learning is a nonlinear dimensionality reduction (NLDR) technique which presume that the data lies on or in the vicinity of a low dimensional manifold and attempt to map the high dimensional data into a single low dimensional, global coordinate system[10].
Section 3 details the experimental procedure, data set, parameter extraction.
This is the result returned by LLE as the low dimensional embedding of the high dimensional data points Xi.
Image data can be naturally represented as points in a high dimensional data space (one dimension for each pixel), i.e. each image is a long vector with the dimension of m × n, where m and n is the number of columns and rows respectively.

Our aim, from this analysis and using primary data from a company that builds algae farms, is to estimate the output of algae biofuel and decrease of GHG emissions in the process.
The origin of used data was a life cycle analysis of a plant in the State of Pernambuco, in northeastern Brazil, and company data See algae Technology (SAT) a specialist in growing algae and producing - among other products of the refining of algae oil - biofuel.
To be able to process all the data gathered in the field, this study resorted to prior literature in order to obtain emission factors of the various activities at the plant.
Based on the values of literature and primary data obtained from the plant, the fuel consumption during agricultural operation of the process is 266,7 l/ha year.
Compared to the original value of 495.19 kg CO2 eq/ha.yr, we have a reduction of 78.4%.

Its presence translates into increase in allotropic transformation temperature as well as hardness and strength properties of alloys at the expense of reduction in its ductility.
The available literature data show that component which could improve the relationship between strength and hardness of unalloyed titanium is carbon [6].
The increase in carbon content is also reflected in the increase in strength properties and reduction in plastic properties (Table 3).

The reduction on the key inputs is important and necessary in terms of effectiveness since the number of experiment increase exponentially with respect to the number of factors.
Cause and effect diagram identifies problem areas where data can be collected and analyzed [3].
Thus, 816 data were collected from one lot whose production lasted 6 days.
Data were analyzed with MINITAB 13.0 Statistical Software.
While this plot can be used for attribute data, the correlation coefficient can be employed for continuous variables.

And it introduced the study area and test data.
Study area and data The Yanzhou coal mining area is located in the south of Shandong Province, China.
Taking No.1 interferometric pair as an example, we illustrated the data processing steps.
Secondly, we introduced background of Yanzhou coal mining region and data.
We also thank USGS for providing us the SRTM DEM data.

It would provide important theoretical basis and data support for promote the sustainable utilization of cultivated land resources.
Methodology and Data sources The change trends of arable land area from 1998 to 2009 would be analyzed.
The arable land area and socio-economic data were utilized from statistical yearbook of Anhui Province [7].
As the basic data have different dimensions and level, the data need to be standardized for equal analysis.
The data were standardized by SPSS17.0.