Search results

The data sampling frequency was 200 Hz.
Results show that when no noise is added, the method based on the ARMAX model doesn’t need many data, and only 361 data are enough to get excellent results (see Fig. 7).
The data length is 361.
It means the proposed method works quite well when applied on real experimental data.
And this method is reliable when applied to real experimental data

Based on GARCH model to catch the financial data of auto correlation and volatility clustering, while the use of expert predictive value and the short term newer data using the Bootstrap method for Vary estimation.
First of all, is the financial market data is always certain timeliness, long historical data may not reflect the current and predicted future situation, so usually not take too long history data.
Below we use prior to 8 August data of SSE index to estimate the day the stock market risk.
Use (2) (3) type reduction, that is 　　　　　　　　　　　　　　　　　　 (4) (4) Type that we use Bootstrap method to estimate the confidence level for VaR.
At the same time, this method, combined with long-term historical data using the GARCH model better catch the financial data of auto correlation and volatility clustering, and then use the Bootstrap method only use a short period of a new data for VaR estimation, avoid the old data invalid information, to improve the estimation accuracy.

According to the data of the Southern power market of China, the order parameter was conformed and the order parameter evolution equation which reflects the operational situation of the market was constructed.
This paper is based on the operational data from 2007-2009 of the China Southern Power Grid.
Based on the data (2007 to 2009) of China Southern Power Grid, the paper uses the grey relational analysis to study the four variables.
By comparing the original data, we find out that when (gotten by restoring )for HHI is about 1050, southern power grid market realize effective operation.
Finally, it gives an empirical analysis combining some related data of China Southern Power Grid regional power market.

Three-dimensional modeling of the image acquisition system based on binocular computer vision scanning system scene with MCU processing the two-dimensional image data collected by the CCD sensor, and complete the image binarization processing using hardware, to improve the efficiency of the image processing; using timing delay analysis circuit the edge of the identification image characteristics, the data amount of the compression of the image processing; will be collected by the CCD sensor the two-dimensional image data transmission to a computer, thereby effectively shortening the transmission time of the data.
(4) hardware and software design of image acquisition MCU system:use of the on-site processing mode, image acquisition and control SCM system analyzes the image feature points, the pre-processing of the data, thereby improving the speed of modeling of the system
(5) software design of image data transfer: the data using the USB interface protocol transported to the computer for subsequent processing in the transmission process, the design of specialized software to simplify the transfer content, further improve the speed of the system modeling
(6) software design of computer-based 3D image generation and processing: the data for analysis, combination, it is determined to create the three-dimensional model, to generate a three-dimensional image.
The three-dimensional modeling system software divided according to their function, can be broadly divided into five parts: the initialization part of the system, the image acquisition part of the system calibration part, image processing part of the point cloud data processing section.

Environment protection ability mainly includes reduction rate of material consumption x55, reduction rate of energy consumption x55, treatment ration of industrial waste x56, recycling ratio of resource x57, and so on.
This tool establishes double feedforward neural network based on selected data, trains this network by Levenberg-Marquardt backpropagation algorithm (trainlm), finally assesses network effect through MSE and regression analysis.
If the network's performance on the original data is not satisfying, train it again or increase the number of neurons or get a larger training data set until a satisfying network is obtained.
For example, to evaluate green sustainable innovation ability of other enterprises or some sample enterprise in following years, just import input data, and get output result based on fitting neutral network system.
Because of business secrets, the paper hides original data and enterprises’ name.

Furthermore, a kind of speedy and unified CRC algorithm for read-out data of the sensor is proposed.
However, SHTxx does not like the DS18B20 which directly outputs the Binary integer and decimal complement form of measurement data, but outputs unsigned binary integer: corresponding to the measuring temperature is -41.08℃, 14 bit output data is x000H; when 100℃, in between 371FH to 3720H.
So only the conversion and non-linearity compensation of the output data are done by receiving party, the real temperature value Tture and humidity value RHture can be solved.
After 11/55/210mS (8/12/14-bit resolution ratio), it serially outputs two bytes of measurement data and one byte of CRC checksum [2, 3].
Noted two important points: 1. the command bytes (03H or 05H) must first be bring into calculation, follow by the measurement data to obtain the correct CRC checksum; 2.

Therefore, in denoising of wavelet transform, the border symmetric assumption is needed to the data in the data window.
(7) Choose the middle part data of 0B' , and abandon the border data.
Collect another m data again, and compose them as a data series, which is denoted as A2＝[s2m+1,s2m+2...s3m].
Process the data obtained subsequently with wavelet denoising to gain a data series: ...,, 1001 BBA '''' .
This algorithm of extending the data border and adopting the middle part of data after been denoised can avoid border interference caused by the data for output being in the window border in each time.

Using a Data-Model.
The data model is the data structure for storing the information during the engineering.
To keep the configuration data in sync with the life cycle of the production system, a tight coupling between configuration data and the production system is required.
Data Model.
The data model was developed using UML 2.0.

An accurate simulation can save cost by a multitude of ways like material savings, process optimization, cycle time reduction, improved quality of parts and so on.
Scan data Fig. 5(b).
CAD data Fig. 5(c).
This data is imported into Geomagic verifier a reverse engineering and inspection software in .stl format and the nominal CAD data (shown in Fig. 5(b)) which is also imported to Geomagic software in .step format is superimposed and the deviation is compared, The result of this comparison process is delivered in the form of a color map deviation (shown in Fig. 5(c) and 5(d)) which pictorially describe the differences between the scan and CAD data.
By considering the best gate location and feed system the part is produced through the injection molding process and the physical part is scanned and the scan data of the physical part is superimposed with CAD data to check the warpage in Geomagic verify and the deviation report is generated.

Finally, the results obtained from the proposed model are examined with published experimental data.
The study demonstrates that the proposed analytical model agrees with the experimental data of existing investigations.
But in this model corrosion pressure was used from experimental data.
It can be seen from the comparison, the results obtained from the analytical model are in good agreement with the experimental data.
The predicted results show good agreement with the published experimental data.