Search results

One of the primary goals of many data mining and machine learning systems is dimensionality reduction.
The goal of dimensionality reduction is to reduce the number of features of data in order to perform tasks like clustering and/or training a classifier.
Non-negative Matrix Factorization (NMF) Suppose that the data matrixconsists of m-dimensional nonnegative data vectors, each of which is a sample vector.
Data Eng, vol. 23, no.6, June (2011) 902-913
Data Mining (2006) 126-135.

This paper fixes the data of the ground settlement and moving calculation parameters by doing a back analysis of monitoring data which comes from ground settlement caused by subsurface excavation and combing with other engineering experience.
This research can reduce or avoid our country and people’s losing,meanwhile,it can provide technical support for the disaster prevention and reduction of underground engineering.The paper mainly make a numerical simulation analysis of which surface deformation caused by tunnel excavation affetes shear-wall frame at main city aera.
It can predict the settlement by using the probability integral method. the parameter is according to the local actual monitoring data’s back analysis and similar project.
Table 2 Prediction Data Of Surface Deformation X(m) -31 -24 -17 -10 -5 -2 0 3 6 11 16 25 34 W(mm) 0 1.8 2.9 5.9 8.1 9.6 11.0 9.8 8.7 7.4 3.9 2.2 0 U1(mm) 1.1 1.9 3.2 3.8 3.1 1.3 0.9 1.2 3.3 4.1 4.1 2.7 0 E1(mm/m) 0.09 0.18 0.20 0.09 0.09 0.09 0.09 0.09 0.09 0.09 0.18 0.23 0.09 T1(mm/m) 0 0.18 0.31 0.45 0.41 0.31 0.10 0.26 0.33 0.48 0.51 0.33 0.10 K1(mm-1) 0.01 0.03 0.03 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.03 0.01 ps： X means horizontal distance between measuring point and tunnel center .1 stands for calculation data of principal direction stress.
It can provide reasonable data for damage evaluation of the building on the ground.

In the ideal case test data are available for both states.
It is well known that low frequency vibration test data or static response data are not very well suited for detecting and quantifying localized small size damage.
Data acquired with scanning laser vibrometers may be used but also test data from static tests.
Examples for both types of data, modal data extracted from laser vibrometer measurements as well as static influence line data measured from a slowly moving load will be presented later in this paper.
In damage assessment it is well known that low frequency vibration test data or static response data are not very well suited for detecting and quantifying localized small size damage unless high spatial resolution of the response data is available.

Quantitative risk analysis is a method to evaluate risk and to identify areas for risk reduction.
Define the potential accident scenarios Evaluate the event consequences Estimate the potential accident frequencies Estimate & evaluate the risks Identify and prioritize potential risk reduction measures Fig. 1 CPQRA flowchart [1].
Therefore, results for t ³ 3.5 ms (last three data points in Fig. 5) are excluded and the quasi-steady fractal dimension (t = 2-3 ms) is used in the following discussion.
Together with the present experimental data, the previous experimental value [5] obtained by large-scale experiments and our previous results obtained by flame instability analysis [7] and computational fluid dynamics (CFD) calculations [8] are also shown in the same figure.
Although the present results are not inconsistent with previous data, detailed quantitative comparison is difficult because of experimental error of the present data.

Jansto1, * 1 CBMM North America, Inc, 1000 Old Pond Road, Bridgeville, PA 15017 USA Keywords: Recrystallization kinetics, Reduction schedule, Reheat conditions, Solidification, TMCP Abstract.
Niobium enables achievement of substantial grain refinement when the plate is rolled with the proper reduction and thermal schedule.
TMCP process metallurgy involves the triad of slab reheating practices, proper reduction schedules for appropriate recrystallization type during deformation and accelerated cooling rates (exceeding a minimum of 5°C/second) on the flat rolled products.
Relationship Between Recrystallization Type and Impact Toughness There is limited data relating the impact toughness to the perecentage of Type 1, 2 and/or 3 in a given microstructure.
- Superheat control 15-20°C - Maintain air-to-gas - Mould level fluctuation burner ratio of 1.10 - Control casting speed variations - Maximize adiabatic flame temperature - Homogeneity of heating through slab thickness Hot Rolling Deformation Schedule - Prior austenite conditioning for Type 1 - Appropriate reduction

Fig.2 B/S structure of operation and maintenance system of petrochemical units The development structure has three layers: Data layer: Provide data store and management by data base server.
The data visit parts is responsible for opening the data in data base to the business layer.
In the DAL, a common connection part of data base is set up to establish the connection among data bases.
The middle layer avoids the direct operation to the data base, reduces the threat of being destroyed for the data base and promotes the security of the data base.
Through the security between the network and data center, the stability of operation and complete of the data can be ensured.

An agreement of simulation results with experimental data is shown.
To ensure equivalent strain at 2.0-3.5 during asymmetric sheet rolling with a 75% reduction, shear angle of metal layers should be 65-80°.
According to experimental data of [10] dislocation density after asymmetric rolling of Al 7075 is (5.5±0.5)×1014.
Thus, simulation results for = 2 agree well with available experimental data [10].
A good agreement of simulation results with experimental data is shown.

The experimental data for the stress-strain response of crystals [111] and [123] was taken from Hosford et al. [9] and for the crystal [5 -4 19], it was taken from Dumoulin and Tabourot [10].
Maps of von Mises strain at ~ 3% and ~ 8% thickness reduction are shown in Fig. 2.
The von Mises strain map in these experiments is determined using the data of displacement in the thickness reduction.
Experimental data from literature (discrete data points) vs. simulation (solid lines).
Von Mises strain determined using the image correlation data in the experiments in some locations shows, different trends from that predicted by the simulations (cf. grains 9, 10, 11, and 17 in Fig. 2b and Fig. 2d).

This trend was in agreement with the results obtained from the XRD data, indicating that the smaller crystallite size gave higher surface area.
Temperature programmed reduction in H2 Figure 3 shows the TPR in H2 profile for bulk VPS catalysts.
The first and second peaks corresponded to the reduction of V5+ phase while the third peak was corresponded to the reduction of V4+ phase.
Increasing the activation duration in 1% O2/N2 shifted all the three reduction peaks to higher temperature.
Table 2 : Total amount of oxygen removed from the VPS catalysts by reduction in H2 Sample Temperature Maxima, Tm [K] Reduction Activation Energy, Er [kJ mol-1] Amount of Oxygen Removed, [mol/g] Amount of Oxygen Removed, [atom/g] VPS6 717 120 5.37 X10-4 3.23 X1020 859 143 4.72 X10-4 2.84 X1020 999 167 1.61 X10-3 9.69 X1020 Total 2.62 X10-3 1.58 X1021 VPS18 719 120 3.89 X10-4 2.34 X1020 839 140 6.52 X10-4 3.93 X1020 986 165 1.62 X10-3 9.75 X1020 Total 2.66 X10-3 1.60 X1021 VPS30 711 119 7.48 X10-4 4.50 X1020 857 143 3.89 X10-4 2.34 X1020 1012 169 1.81 X10-3 1.09 X1021 Total 2.95 X 10-3 1.78 X1021 VPS75 758 127 1.75 X10-3 1.05 X1021 863 144 5.74 X10-4 3.45 X1020 1000 167 1.94 X10-3 1.17 X1021 Total 4.26 X10-3 2.57 X1021 Selective oxidation of n-butane to maleic anhydride The details of the catalytic performance data for VPS catalysts were shown in Table 4.

Selective catalytic reduction (SCR) had significant effects on mercury removal and Hg0 re-emission rates across WFGD.
Selective catalyst reduction(SCR) systems used with wet FGD scrubbers may enhance mercury capture in the scrubber.
The reduction reaction of Hg2+ by sulfite ion such as HSO3- in weak acid condition was proposed as follows: Eq(1) Sulfite ion was from the absorption of SO2 in the absorber and its concentrations were related to absorbent type, liquid to gas ratio (L/G) in the scrubber, SO2 concentration at FGD inlet and oxidation modes etc [5].
OHM was performed at interval, and the OHM data are shown in Figure 4.
The OHM data shows that the additive has slight control capability in Hg0 re-emission.