Search results

These are: - PZT transducers network divided into several measuring nodes; - Remote Monitoring Unit (RMU) – based on DSP architecture CPU; - Data Storage Unit (DSU); - Graphical User Interface (GUI).
Furthermore, the most of data classification models are sensitive to outlying observations; therefore an efficient sensor self-diagnostic prior to the structure evaluation is crucial for proper system operation, e.g. false calls avoidance.
The number of well separated groups of data, corresponding to different extent of the crack agrees with the number of sensing paths intertwined with the crack.
Therefore, the separation of the 2nd (5–15 mm) and the 3rd (>20 mm) group of the data (Fig. 5), corresponding to crossing those diagonal paths by the crack, is smaller than in the case of “W.15” node where intersection of such paths is shifted (Fig. 3).
The results of the tests are promising for further development of the concept of the system, especially from the point of view related to sensors technology integration with the structure as well as the methods for data classification.

(a) (b) (c) (d) Fig.1 Sketch map of tuyere 2) Experiment data of the tuyere.
So that we can improve steam recycling times in the oven, achieve the goal of energy conservation and emissions reduction, as shown in Fig.2 (b).
In order to verify performance of drying system after adding LEL sensor, on the premise of meet the print quality of drying, the experimental method is as follows: 1) According to the hot air circulation path of the FR300ELS gravure press, choosing the best installation location of LEL sensor, and install LEL sensor; 2) Choose the appropriate position of wind pressure test (as is shown in figure 3) in two kinds of drying system; 3) On the premise of meet the print quality of drying, develop the hot air experiment and record the related data; 4) According to the data from the LEL sensor, we prove that under the control of the LEL，the production process will be more security; 5) According to the data measured in step (2), calculate the minimum air consumption, and prove that the improved gravure press drying system have a better energy saving effect.
Develop the hot air experiments for the drying system of FR300ELS gravure press within and without LEL sensor respectively. 1) Hot Air Experiment Data On the resulting values, to calculate and get the minimum air volume, as is shown in Tab.2 and Tab.3.
Tab.2 Common consume of steam in 30℃ Printing speed Printing area Measured condition Conversion condition Actual temperature difference Consume of steam Conversion temperature difference Consume of steam 100m/min 0.85m2 49℃ 1kg/min 30℃ 0.61kg/min 150m/min 0.64m2 48℃ 0.9kg/min 30℃ 0.5625kg/min 180m/min 0.485m2 43℃ 1kg/min 30℃ 0.6976kg/min Tab.3 Data of experiment added LEL Printing speed Printing area Measured condition Conversion condition Actual temperature difference Consume of steam Conversion temperature difference Consume of steam 100m/min 0.89m2 30℃ 0.25kg/min 30℃ 0.25kg/min 130m/min 0.89m2 30℃ 0.25kg/min 30℃ 0.25kg/min 150m/min 0.9m2 27℃ 0.2kg/min 30℃ 0.22kg/min 180m/min 0.697m2 30℃ 0.29kg/min 30℃ 0.29kg/min 200m/min 0.846m2 37℃ 0.75kg/min 30℃ 0.60kg/min 260m/min 0.846m2 37℃ 0.66kg/min 30℃ 0.528kg/min 300m/min 0.95m2 30℃ 0.25kg/min 30℃ 0.25kg/min It can be seen from the chart above, the steam consumption is smaller in improved experiment than pre-improved.

In order to realize the dynamic measurement of the thickness, three difficult problems which shown as follow are needed to solve;(1) The measuring accuracy must reach to Micron-grade, but the vibration of the machine and some processing environment factors can affect the measurement result, so the reasonable struck of the machine and correct data processing method are needed to reduce the error. (2) With the wear of the grinding wheel, the change of the measuring basis could exceed the normal measuring scope of the sensor, the device that can realize two-dimension motion must be designed to adjust the measuring basis of the sensor dynamically;(3) the voltage between the electrodes could affect accuracy of the eddy current sensor, we can take advantage of the pulse characteristic of the voltage added on the wheel to reduce the influence.
The data co- electing card (USB-6212) is used in system to collect the signal and the signal is transferred to com- puter by USB.
The data is processed in the data-collected system based on the LabVIEW.At the end; we can obtain the thickness value.
Results comparison of rust thickness in different measurement methods number of sample points 1 2 3 4 5 6 Results of sensor[μm] 20.59 22.95 22.15 31.02 27.52 34.38 Results of microscope[μm] 25.084 27.563 26.674 35.266 31.920 38.771 The measurement results obtained from different method is shown as Figure 4 Measurement results by sensor and electron microscope 0 10 20 30 40 50 1 2 3 4 5 6 The number of sample points Results of sensor Results of microscope Thickness [μm] [μm] Fig.4 Results comparison of rust thickness in different measurement methods Summary The comparison result of the measuring point shows that the systematic error is a stable value (4.5μm).The influence of the error can be eliminated by means of correct data processing method.
A study on wear mechanism and wear reduction strategies in grinding wheels used for ELID grinding [J].

Elastic-plastic indentation was simulated using the ABAQUS/STANDARD finite element code that can analyze a large strain with uniaxial stressstrain input data.
Young's Modulus and hardness were calculated from the load displacement data obtained by nano-indentation on each sample at five different indentation loads ranging from 30mN to 260mN.
As shown in Fig. 2, the fracture strength, of 5.3 GPa best fit the experimental data.
In Figure 3, during the unloading process, the experimental data show that at an indentation depth of around 580nm, a large recovery occurs without a change of the load, a phenomenon known as pop-out or time dependent recovery.
References [1] Domnich, V., Gogotsi, Y., "High pressure surface science", Handbook of Surfaces and Iterface of Materials(2001), H.S.Nalwa,Academic Press, New York, p195-237 [2] Hu, J.Z., Merkle, L.D., Menoni, C.S., Spain, L.L.; Crystal data for high-pressure phases of silicon, Physical Review, B34 (1986), pp4679-4684 [3] Pfrommer, B.G., Cote, M., Louie, N.G., Allan, D.R.; Ab initio study of silicon in the R8 phase, Physical Review, B56 (1997), pp6662-6668 [4] T.

The diffraction data for title complex were collected at 293 K on a Bruker Smart CCD diffractometer with Mo-Ka radiation (l=0.071073 nm), and the data reduction was performed using Bruker SAINT.[14] The structure was solved using a direct method, which yielded the positions of all or most of the non-hydrogen atoms.
All calculations were performed using the SHELXTL programs.[15] The crystallographic data are summarized in Table 1.
CCDC-798273 Table 1 Crystallographic data and structure refinement for title complex.
Empirical formula C90H66I6N24O2Zn3 α [º] 103.364(2) Formula weight 2473.2 β [º] 91.508(2) Temperature [K] 293(2) γ [º] 101.476(2) Crystal system triclinic V [nm3] 4.5412(8) Space group P-1 Dc [g·cm-3] 1.809 a [nm] 0.99234(10) m [cm1] 2.888 b [nm] 2.1284(2) S on F2 0.997 c [nm] 2.2615(2) Z 2 R1a, wR2b [I>2σ (I)] 0.0788, 0.1861 R1a, wR2b [all data] 0.1543, 0.2268 R1a = å||Fo|-|Fc||/å|Fo|, wR2b = [åw(Fo2-Fc2)2/åw(Fo2)2]1/2.

Those data are of great importance for large scale numerical simulation of the explosion effects from pyrotechnic mixtures.
With those known data, the new set of mole numbers xi can be found from Eqs. 25 26.
Their molar standard thermochemical data are obtained from JANAF databank [8].
From these data as initial state, the explosion effects can be estimated via hydrodynamic code to for the calculation of shock wave propagation.
[8] Data from http://kinetics.nist.gov/janaf.

Combining theoretical considerations, including extensive ray tracing, and experimental data, a new pentaprism adjustment procedure was developed [6].
The resulting calibration data are used to correct a given angle reading of the autocollimator to achieve an accurate and traceable angle measurement by means of the device.
The measurements were performed by applying an optimized shear combination of 4 & 35 data points (=140 data points per scan, physical shears of 2.3 mm & 20.0 mm for the scan length of 80 mm).
The measurement uncertainty (calculated in accordance with [12]) for the ESAD topography scans is a function of the position of the data point in the scan.
Summary The following points summarize the basic features of ESAD shearing deflectometry in general and the measuring capabilities of the ESAD device at PTB: • Absolute measurement method (basis: straight propagation of light) • Standard measurement uncertainty of topography < 1 nm up to 500 mm scan length • Optimized measurand (angle and length) traceability to SI units • Near-constant measuring conditions independent of surface dimensions • No basic limitation of the dimensions of the surface under test to be measured • Optimized adjustment (automated and in situ) of pentaprism and surface under test • Reduction of influences of prism guiding errors on deflection angle (by factor 1:10000) • Elimination of any whole-body tilting of the surface (due to shifting of the specimen) Acknowledgement The author would like to thank Dipl.

Modern neural networks are non-linear statistical data modeling tools.
They are usually used to model complex relationships between inputs and outputs or to find patterns in data.
Clustering analysis is a technique to generate groups of subsets of data called clusters within which the distance between subsets is small, whereas a distance between clusters is greater in that two subsets from different clusters are distant.
The measurement data of the T1, T7, T25, T32 are used to train the prediction model, the output of the model is the Y axis thermal error and Z axis thermal error.
Reduction and Compensation of Thermal Errors in Machine Tools.

Amorphous Computation Material, or ACM, is a concept of an active material that has the capacity to sense its immediate environment and process that data for intelligent actuation–actuation where the applied force is based on the type of material handled.
ACM poly mer sensors and actuators, and wireless data communication are part of our long-term vision, but are not addressed in here.
Because of our finding that only the points closest to the end of the valve have a noticeable effect on the direction of propulsion (section comparison of Simulation and experimental Data )a minimally-sized, yet functional architecture for the network – was determined to be 4 input neurons, 4 hidden neurons, and a single output neuron.
The data used for this training was a rough approximation of the shapes generated for water hammer experiment (section Experimental Setup ).
Our approach is based on a much simpler design, motivated by a reduction of the complexity of both the individual neurons and the entire network.

For example, face image data is two-dimensional, the two dimensional face images must be transformed into one dimensional vectors, the size of the obtained vector space is very big.
Finally, paper conclusions are drawn in section 4 Face recognition algorithm DCT Algorithm 2D DCT is defined as[9][10]: (1) Where (2) The corresponding inverse transform is defined as: (3) For face image data is two dimensional, apply 2D-DCT to face image, then select some coefficients to restructure image.
It indicates that the part coefficients of DCT contain the primary information of the original image, and DCT can realize the effective data reduction.
In the experiments, we select some images from face images data to construct the training data set, the remaining images to be used as the test images.