Search results

Its deployment of distributed energy management services in three-tier web services environment, and proposed optimization algorithm to optimize the coordination of each dynamic voltage scaling technology settings to maximize the reduction of the overall data center power consumption.
Virtualization energy-saving technology has brought lot solutions to cloud computing data center energy management.
In the data center level, server virtualization technology can be integrated into multiple loads on the same physical machine.
The virtual machines are made as a unit to allocate resources application for the user in cloud data center.
Deployment and migration of virtual machines is an important issue in cloud energy optimization data center.

Its main steps include feature extraction, unlabeled data category prediction and action classification with nearest neighbor method.
The MHI of different actions of daria in Weizamman data set is shown in Fig.2 (b).
The detailed steps for unlabeled data category prediction with co-training based semi-supervised are as follows: Input: Set of two views of labeled data, i.e.
MEI and MHI of labeled actions, and unlabeled data, i.e.
Then, the co-training based learning algorithm is used to predict the class label of unlabeled data.

Generally speaking, these existing CFO estimation approaches can be categorized into two groups: data-aided and non data-aided estimators.
Non data-aided approaches, which do not require training sequences or retransmission of data symbols and also known as the blind approaches, are thus preferable.
Denote as the kth block of data to be transmitted.
Using one OFDM blocks of noiseless data, the null spectrum of in (19) is illustrated in Fig. 2.
Finally, Complexity comparisons of two approaches using one OFDM block of data and four OFDM blocks of data in Fig. 3 are shown in Table 2.

In breaking pile head, lashing crown steel beams to the attention of the sensor data transmission line protection, especially not damage or even tearing the data transmission lines.
Experimental stress data collection carried out according to the progress of excavation.
After the bottom of the pit dug collect data once every other day, in case of precipitation or other special circumstances to increase the data acquisition times.
Data collection work has been completed to backfilling, stress stable end.
Test data processing.

This marks the second important paradigm shift in manufacturing metrology towards holistic measurement data acquisition.
Today there are two different possibilities for these evaluations: One is the analysis of the entire volumetric data set and the other is the analysis of a surface model extracted from the original data set. [3] All evaluation methods and software tools for measurement data evaluation known from conventional coordinate metrology rely on surface data sets.
So these data has to be extracted from the volumetric data sets.
That means that certain features are present in several point clouds and others appear only in a single data set.
The aim of sensor data fusion is to combine the different point clouds and calculate a new data set containing all information of the different measurements and their interferences to a more detailed view of the workpiece as a base for function-oriented analyses.

In this aspect, data mining technology [9] can play a vital role.
By identifying and eliminating possible information risks with the help of data mining technology and data processing technology, the whole work of data analysis can be made simple and intuitive. 2) Information transmission stage.
The system interacts with EMS advanced application to get required data for calculation.
Data between the two parts is transferred by unified data format, as is shown in Fig.3(b).
Data mining technology for failure prognostic of avionics[J].

The article analyzes the stability of agro-ecosystem’s productivity taking sustainable development theory and wave theory as guideline, conducted field study of the research area taking ** as research case based on analyzing ** yearly grain output, and establishes relatively complete, clear and easy accessible evaluation target system for agro-ecosystem productivity stability through analyzing the constraints of the stability in the area and making full use of current statistical data with combination with actual situation.
It analyzes the stability of agro-ecosystem productivity from the aspect of both time and space with current indices and standardized data using synthetic index method and volatility index calculation method.
Land is thus more utilized for plants with economic profits or even for non-agricultural purpose, which leads to reduction of grain planting area and decrease of total grain output.
The systematic research and analysis includes the elements, relation between the elements, situation of ** county and villagers data. 17 indices are selected as per the characteristics of stability of production and the research outcomes on stability to reflect the overall condition of production stability in ** county.
Conclusion On the basis of yearly statistical data and field study data, the article analyzes the natural and social economic conditions of ** county systematically, looks for the factors restricting stability of agro-ecosystem productivity in ** county through variations of the years, establishes database for researching stability of agro-ecosystem productivity, and gives scientific basis for the establishment of evaluation guideline system for stability of ** county agro-ecosystem productivity.

Spectral images contain a large volume of data and the development of multispectral imaging systems places considerable demands on computer hardware and software compared with standard three-component or trichromatic image storage and processing.
Introduction Spectral images contain a huge volume of data and the development of multispectral imaging systems places considerable demands on computer hardware and software compared with standard three-component or trichromatic image storage and processing.
Many studies [5,6] have been done to study the number of basis functions required for low-dimensional linear models, which are all derived from the spectra data sets themselves, and few works are concerned with human visual system expect one previous work using human L, M, S cone fundamentals to get the first three components Lum(l), rg(l), by(l), which represents one luminance and two chromatic channel respectively [7].
When comparing the results in Table 1 which contains reconstruction errors for the same group of images using both proposed representation method and conventional linear model, it can be found that when only 3 basis functions are used, the spectral reflectance reconstructed by conventional linear modelling presents much smaller reconstruction errors spectrally than the spectral reflectance reconstructed by linear modelling with the first three basis functions derived from color matching functions, but the color difference DE*ab is not necessarily smaller, and it is dependent on the image data sets.
While by using the conventional linear model, where the basis functions are derived directly from the data set its own, mathematically, the spectral fitting should be the best, but it may create spectra that has relative large error at the spectrum where human eyes are very sensitive, and therefore presents large reconstruction error in color difference.

The collection of the data of the existing production line was conducted at one of the automotive company in Malaysia.
The data required in the analysis are the customer demand per day, the available working time per day, the number of manpower used, the number of workstations includes and the dimension of all workstations.
By using this data, the cycle time, takt time, minimum number of workstation and efficiency were calculated.
Results and Discussions According to the data provided by the company, the customer demand per day is 16 units.
Based on the data collected which is the cycle time from each workstations on the Trim line, the cycle time on each workstations are shown in the Table 1 and can be visualize using Yamazumi in Figure 1.

The obtained results of simulations are in good agreement with experimental data. 1.
Therefore, an ARAMIS digital image correlation (DIC) system was used to obtain the full-field in-plane displacement and strain data.
The empty regions in the picture represent the flaking of paint, which was speckled, and thus the data at these regions was not collected.
The material’s elasto-plastic behaviour was defined by inputting the true stress-strain data obtained from the nominal stress-strain test data for both types of specimens.
If mechanical gauges were used, these data had required several tests with different gauges.