Search results

Secondly, the parameters of model were achieved by using vector fitting method based on the measured impedance data under different voltages.
Experimental data at low frequencies (Fig.3) showed that the capacitance C increased with the applied voltage within the specified voltage range.
One possible explanation for the increase in capacitance with voltage can be due to the electrolyte dielectric constant increase, or, to the reduction of the distance separating the charges at the electrode/electrolyte interface [3].
The electrochemical impedance spectroscopy (EIS) experimental data is used for parameter identification at each voltage.

Compared to the alternative MOS characterization, the non-contact method lowers the manufacturing cost and shortens the data feedback time from weeks to less than one hour.
Availability of C-V corresponding to DVlight data makes it possible to compare the present approach with the other direct method for flatband voltage described in ref. 4.
For a comparison, the corona non-contact 1/C2 data was differentiated and analyzed according to ref 4.
In the corona-Kelvin method the interfacial charge trapping DQit is manifested as a reduction in magnitude of the surface space charge change DQSC in response to the corona charge increment DQC.

Fig. 3 Model calculated data and observed data fitting curve Modeling Result Water Balance.
The head data of the shallow and deep aquifers in 2008 were shown in Fig. 5.
In all, the deep groundwater of over-exploited area varied greatly during the years 2009-2023 that reflected in levels restoration and scope reduction of the cones.

System framework General framework of the intelligent online monitoring and anti-theft system for underground cables comprises an anti-theft device, a remote monitoring subsystem and a server acting as the data center, wherein the anti-theft device, as is shown in Fig. 4, is composed of an ARM processor as the overall controlling component responsible for receiving and sending controlling instructions, a constant DC input module for generating and feeding detective signals into the cable ground loop, a signal processing module for detecting the voltage of R0 and amplifying the signals, a fault diagnosis module for determine whether a failure is occurred and a GSM alarm module for sending alarm SMS when needed to inform a cable failure.
Fig. 4 Framework of the anti-theft device Anti-theft devices, each installed in one of the distributed substations, are connected by branches of underground cables and provided with network communication interface by which the anti-theft devices can communicate with the data center, so that a local area network composed of anti-theft devices, cables and the server is built up.
The remote monitoring subsystem installed in the data center where the status information of the branch cables are aggregated, analyzed and stored can monitor the running status of each device in real time.
Loss reduction.

Furthermore, high temperature at the Neodymium-Iron-Boron (NeFeB) permanent magnets could cause the reduction of magnet’s magnetic fields strength and eventually reducing the generator’s efficiency.
Based on the data collected, the winding coils temperature is founf to settle after 9 hours and no peak temperature is detected as shown in Fig. 2.
To formulate an empirical thermal model for the winding coils temperature in a controlled environment based on the data collected, a regression analysis was used together with a curve fitting method.
Conclusion From the data presented, it can be concluded that there is a minor effect of the room temperature on the winding coils temperature.

Again, the defect of high computational cost is easily handled by the AdaBoost feature selection component of the Ada-AdaSVM. 3 Feature Reduction The boosting algorithm boosts the performance of weak binary classifiers by strengthening training on misclassified sets.
The scheme of the algorithm is to weight a set of weak classifiers in respect to a function of the classification error .The feature selection algorithm can be simplified as follows: Step 1: Calculate the probability distribution of the data according to the weights and set the initial distributions and weights to be equal to each other if prior knowledge about them is unknown.
Step 2: In each iteration, call the Weak algorithm K times to leam the N*M number of data, M = number of features) according to a single feature each time.
The statistics presented in 2 indicates that the proposed spent far lesser time to process the same amount of data than any of the three popular methods.

A plausible mechanical relationship between mat properties and injury risk has been established by experimental data and mathematical modeling.
Method Movement of salto backward stretched with 3/2 twist was captured by two high-speed video cameras, and then kinematics data were digitized by using SIMI Motion 3D analysis software.
Simulation Process Define segments and joints of human body model, and then transform kinematical data with Simi motion into motion agent of human body by programming with python language.
Standard test ability to measure impact forces reduction on mats [J].

Oxidation in air led to the formation of an α-Fe2O3 phase and a sharp reduction of magnetization down to 20 emu/cm3.
The Mossbauer data of the starting Zn/Fe2O3 bilayers shows the formation of primary α-Fe2O3 (~ 67%) and secondary γ-Fe2O3 phases (~33%) Fig. 2a.
X-ray data and the occurrence of magnetization at 250 °C, along with its subsequent growth, indicates the mixing of the Zn and Fe2O3 layers of Zn/Fe2O3 films following the solid-phase reaction (1).
This data is consistent with the results of the Mössbauer and X-ray studies indicating the formation of the ZnO-Fe3O4 nanocomposites after annealing at 400 °C.

Experiments and Results The experiment data used in our work is the simulation data coming from the University of Leicester.
A．Low SNRs As the spikes can be extracted from the raw data using the spike time and templates, the test data with different SNRs could be constructed.
In the experiment, the experimental data is processed as follows: First extract the noise from the raw data, and zoom in or out noise signals according the Eq. 4 as in [9] to synthesize different SNR experiment data.
The spike times of new test data is the same as raw data
Baseline Wander To simulate the data with baseline wander, a 0.2Hz sinusoidal wave was added to the raw data.

Before data is collected with the three-dimensional laser scanner, the shape of the teacup is analysis first, the purposes of the analysis focus on conducting the necessary scanning path planning, improving the scanning efficiency and avoiding incomplete data scanning.
(2) Filing-up of missing data
If these semi-discrete points are cut off, data will be missing in this position, and this position of the teacup will be a hole after aligning.
The data is imported into the rapid prototyping machine, and the data is reduced by half in the processing, to obtain the resin cup in Fig.1(c).
The reverse engineering of the teacup is very successful. 4 Problems in Data Processing 4.1 Aligning For a sample, in order to obtain the complete data, scanning with the three-dimensional laser scanner shall be conducted for several times, because the coordinate systems are inconsistent, only after spliced, the point clout data can compose the complete data of the sample, therefore, the data in each scanning shall has pointes overlapped with in another scanning, when the overlapped points are determined, the aligning and fitting can be computed by the software, to obtain the point cloud data under the same coordinate system. 4.2 Scanning Path Planning Because linear scanning is adopted in the three-dimensional laser scanner, the scanning path shall be rationally planned, which helps to improve the efficiency and the quality of the point cloud data. 4.3 Format Conversion The import / export formats are different in different three-dimensional reverse software, the import / export