Search results

PCA is a method of a dimension reduction where it discards the noise by projection of the data onto the main principal components.
Methodology Our denoising step consisted of using the dimension reduction technique and find an adaptive basis in the projection data set.
This because the hidden or "latent" data structure is masked by noisy dimensions and becomes evident after the dimension reduction.
It does this by transforming the data to a coordinate system so that the greatest variance of the data by a projection of the data into the principal components.
PCA can be done via SVD of the data matrix.

Extending the range of finished product sizes from a given ingot or concast bloom or billet section is often limited by the minimum area reduction required to ensure effective central consolidation and final mechanical properties.
(a) (b) Fig. 4: (a) Initial hole and section geometry (b) Variation of relative hole size/relative section height with inverse roll gap shape factor hm/L at bloom/ingot centre (experimental data based on lead/steel samples function of roll diameter and section height) From Fig. 3d and Fig. 4b, it can be observed that using 2D plane strain indentation theory: · For hm/L = 1, homogeneous deformation prevails, the vertical stress sy is maximum in compression and uniform and closure will be greatly accelerated (see also Fig.4b) · For hm/L between 1.8 and 4.8, the vertical stress sy is compressive but all other stresses are tensile, allowing some consolidation although at a lower rate than the two previous conditions · For hm/L between 1 and 1.8, the mean or hydrostatic stress is compressive, therefore closure of porosity will be promoted at a greater rate than reduction as hm/L decreases. · For hm/L > 4.8, all principal stresses are tensile representing a regime of tensile triaxiality
(in Fig3d, sz is the mean stress), therefore consolidation is unlikely to proceed at a rate greater than reduction with low/no penetration as well as nucleation / growth of cavities.
A comprehensive DoE based 3D FEM using the ABAQUS commercial software (incompressible flow) has been run integrating range of roll diameter (305-820mm) and feedstock reduction.
Recently the consolidation factor CF at steady state was fitted to all FEM DoE data against the inverse roll gap shape factor hm/L and further normalised to account for effect of temperature (Tm: melting temperature of given steel grade), thus following analogy presented with respect to slip line field / state of stress of Figs 3-4 (equation 4).

Proper selection of cross-sectional area reduction is therefore a crucial task as the improper reduction sequence may lead to excessive straining, possibly causing cracks or even material failure.
The tube cross-sectional area reduction can be calculated as follows: (1) where: SR = S0 – S [mm2] and S0 – tube area before drawing [mm2], S – tube area after drawing [mm2]. [4] In the 1st pass the area reduction was 35 %, in the 2nd pass the area reduction was 33 % and, finally, in the 3rd pass the area reduction was 29 %.
It should be noted however, that the results obtained are merely qualitative, because the material flow stress data and the friction model used were just an approximation of the true material behaviour that would eventually need a more sophisticated identification.
This effect was caused by insufficient lubrication after several high-reduction passes.
This met our goals of a successful drawing with 70 % overall area reduction.

Isothermal compression of superalloy GH4169 has been conducted on Gleebe-1500D hot simulation at the deformation temperatures ranging from 950℃ to 1100℃，the strain rates ranging from 0.01s-1 to 10s-1, and the height reduction of 50%.
Constitutive equation of superalloy GH4169 was established by experimental data.
Error analysis showed that calculated stress values by the established constitutive equation were coincident with experimental data well, and it provided the theory basis to optimize forging processing of superalloy GH4169.
The hot compressive tests were conducted on Gleeble-1500D simulator at deformation temperature of 950℃, 1000℃, 1050℃ and 1100℃, at strain rate of 0.01s-1, 0.1s-1, 1s-1 and 10s-1 ,and the height reduction of 50%, followed by water quenching, to preserve the hot-deformed structure.
On the basis of experimental data, the hot deformation constitutive equation of superalloy GH4169 was established as following： 3．

(d) shows that there is a certain error between RMS from simulation and calculation,this error becomes most evident when elevation is 15°,after that the difference decreases when elevation rises.But because magnitude of difference is 10-3mm which is too small,so simulation results and calculation results are consistent and it can prove correctness of vector superposition based on Eq.6.Data of RMS obtained by simulation is optimal.
(8) After obtaining error data by simulation,Eq.7 and Eq.8 can calculate efficiency and gain loss of 25m antenna,then obtain the influence of elevation on efficiency and gain loss,at last evaluate the electrical performance when working in different band.
After transformation for 25m antenna,it is possible to cover all nine bands.Since the change of gravity deformation is caused by changes of elevation,elevation of antenna is the parameter which can be directly displayed from computer,and RMS is difficult to measure.So for creating reference conditions for the compensation of antenna.This paper obtains data about under different bands electrical performance parameters changing with elevation,Table 4,Table 5 and Table 6 shows those data.
According to Eq.7 and Eq.8,model about change of gain loss/efficiency with elevation angle can be established.First step is doing data fitting and obtains a model by using Table 3. and this model describes the correspondence between RMS and elevation angle.Result is shown in Eq.9.Unit of error is meter
[4] Hu Kaiyu,Aili Yusup,Xu Xuelin,Xiang Binbin and Liu Qi,Accurate data fitting for adjustments of focus position coordinates applied to Cassegrain antenna’s sub-reflector compensation[J].

Cameras capture image information from video monitoring site, through compression combined with the wireless network technology, to send data to the monitoring center.
The control center decodes video data the received video data display, realizing the wireless network video monitoring.
Video Compression Module.In the digital video communication, without a huge video data compression, it is not convenient to transmit on the network.
large amount of data and allowing a certain error rate, compressed video uses the UDP protocol to transmit high-efficiency.
The testing data is based on 802.11b wireless network, and with one HUB.

Another well-established reduction method is the solvothermal reduction that combines simplicity and effectiveness [25].
Reduction of graphene oxide.
Diffraction data were collected by using a PANalytical X´Pert Pro diffractometer operating in θ-θ configuration, with CuKα radiation (45 kV-40 mA), in the angular range of 5° < 2θ < 80° with a 0.017° step size.
Regarding Tour method is evident that the combination of reduction methods increases the percentage of carbon and therefore the degree of reduction.
Evolution of Raman spectra at the different reduction methods.

The system combines advanced 3G data transmission,professional engineer support in the remote background with collecting information ,it is greatly improved.
Result : b 1/ (L+1) (5) Weighted average method： Result = ((int)(0.7 * r) + (int)(0.2 * g) + (int)(0.1 * b)) (6) 3.4 The design of 3G communication and data transmission 3G transmission mode is adopted without orders, direct data transmission as set of parameters of SerialNet model in this system. 3G transmission mode is adopted without orders, direct data transmission as set of parameters of SerialNet model in this system, its advantage is that the connection is established once as the parameters, there is needn't change the SOCKET connection by command, the image data can be transmitted through the serial data communication with stable direct connection.
Communication parameters of this system: SerialNet mode, 8 data bits, no parity bit, one stop bit, no hardware flow control, communication baud rate is 115200bit/s.
[5] Yang Liuchun..Intelligent data acquisition system based on ARM designs[J].Automation & Instrumentation, 2011,(01)
[6] Li Juan,Tang Lijun,Wen Yongjun.The Design Of Handy Data Acquisition System Based on ARM[J].Microcomputer Information,2008,(25)

Hashemian , there are 3 sources of Data for predictive maintenance which are existing sensor-based maintenance techniques, test sensor-based maintenance techniques and test signal based maintenance techniques (Refer Fig. 4).
CONDITION MONITORING IMPLEMENTATION The equipment stores all the process sensors readings inside a data log-file.
A Microsoft Access programmed with macro on a server and then linked to all the 8 diffusion equipment was developed to automatically extract the required data at a specified time.
Each Bubbler Tube Pressure and its level readings are stored in the computer system inside a data log file.
These data are being extracted using Microsoft Access twice a day and exported into a SQL database.

As mentioned above, experimental data were taken from the specimens tested at 5, 9, 14, 23, 35, and 70 days for the groups of specimens with w/c ratios of 0.60 and 0.72 as shown in Fig.4.
First, to make a comparison of results between theoretical and experimental data, an equivalent ratio of 1 week in the accelerated carbonation chamber at 4% CO2 concentration to 1 year in the theoretical model was assumed.
These data are based on that proposed by [20], whose research proposes a correlation of one week of carbonation in the laboratory equivalent to one year of natural carbonation.
When comparing the theoretical and experimental data shown in Figure 6, it can be observed that both have a very similar trend.
- The modified CEB-FIB model has an accuracy of ±0.1 mm according to experimental and theoretical data for concrete with a w/c ratio of 0.60 and 0.72