Search results

Supervisory Control and Data Acquisition (SCADA) systems data from a Turkish wind turbine were leveraged to develop a predictive model using the eXtreme gradient boosting (XGBoost) algorithm.
The use of SCADA data as a basis for model training is crucial in wind power prediction, as shown in the study utilizing deep learning models with high-resolution SCADA data [7].
The model development process begins with data preprocessing, and then normalization of the data.
The results demonstrate that data performed well in predicting power output.
SCADA data for wind turbine data-driven condition/performance monitoring: A review on state-of-art, challenges and future trends.

However, these methods are only for the data reduction and streamlining, no method is adopted to fit the data more effective.
The data measured by using csvread function was imported in MATLAB, and was denoised by wavelet-transformed, saved as .ibl format.
Then the noise data was imported into Pro/E software, small plane characteristics was formed by envelope processing and small plane processing.
Conclusions The experimental measurement was used to data extract the surface with different roughness in turning aluminum products.
Small plane characteristics were formed by data preprocessing, then the double three B-spline was used to surface fitting, the solid model and surface contact model were established finally.

Statistical data analysis.
(3) The values of the parameters β, γ1 and γ2 can be determined by a MLE performed on the experimental data.
For each data set, black filled markers have been used for the case of failed specimens, while empty markers have been employed for run out specimens.
Compressive loads lead to a considerably reduction on fatigue life.
Rossetto: Comparison of fatigue data using the maximum likelihood method.

Properties “Kfoam” “Pocofoam” Pore diameter (mm) 0.5 0.31 Porosity (%) 78 75 Bulk thermal conductivity (W/m×K) 55 135 Density (g/cm3) 0.34 0.58 Data Reduction and Experimental Uncertainties.
All temperature and pressure signals were acquired by a PC-based data acquisition unit (Yokogawa MW100).
To minimise data reduction uncertainty, the time-averaging method was employed to reduce the data derived from the experiments.
The present experimental data show that a larger pore diameter may help bubble generation and detachment at the pore level.

The identity of the simulation results with experimental data on the example of a complex traffic pattern is shown in Fig. 2.
This situation consists in non-specificity of the existing criteria for the efficient priority driving of route transport, the lack of data on admissible volume of passengers flow, combined traffic volume on the main and conflicting directions.
Mironchuk, Investigating Intermittent Bus Lanes using Simulation Data.
Dailey, A prescription for transit arrival/departure prediction using automatic vehicle location data.
Callas, Analysis of Transit Signal Priority Using Archived Tri-Met Bus Dispatch System Data.

The main topic is to build and test a software tool with Visual Basic Excel to predict the PEMFC performances starting from operating conditions and with different technical data.
Also called mass transport losses, they relates to the reduction of the fuel’s concentration in the gas channels.
However, we have considered studying the performances of a PEM fuel cell with this tool [8]; the data input are shown in Table 2.
The data input of the PEM fuel cell.
The FCvb tool, by referring to the technical data in Table 2 taken from [10, 11], gives the graphical results presented in Fig. 5, and Fig. 6.

The data display an Arrhenius behavior with a relatively high degree of linear correlation (correlation coefficient ~ 0.965) and an apparent formation energy of ~8.6±1.2 eV is deduced from the slope of the Arrhenius plot, which is indeed higher than that deduced for VC formation under thermodynamic equilibrium [4,5].
No evidence is reported for the influence of the D-center on the minority charge carrier lifetime, but such data are strongly needed in order to assess the significance and challenge of thermal formation of D-center during high temperature processing.
The data are compared to SIMS depth profiles of B in the same samples (SIMS detection limit ≈ 1014 cm-3). .
An apparent formation enthalpy of ~8.6 eV is extracted from the slope of the data.
Summary The formation of D-center after high temperature processing in the range 1700 to 1950 °C was experimentally demonstrated, and the data obey an Arrhenius behavior yielding an effective formation enthalpy of ~8.6±1.2 eV.

For a high level scientific exploitation of experimental data obtained from the XRD analysis, a procedure which determines the network parameters of the phases, the crystallite size and the microstrain at network level has been used.
PEAKFIT defines a diffraction line hidden as if it is not responsible for a maximum in data flow.This does not mean that a hidden diffraction line is not visible in the collector design.
As a result of the fitting process, PEAKFIT report the amplitude (intensity), the area, the center and the width of data for each peak.
As for the diffraction patterns discussed above, the diffraction angle chosen for data processing with PEAKFIT program was in the range of 30º- 80º, whereas the Voigt function was chosen.
Fitting was performed for each sample and the resulted data are presented in tables for maximum diffraction angle (2θ) and width at half peak height - β.

Introduction The honeycomb sandwich structures are widely used in aerospace industries where weight reduction and high flexural rigidity are the major concerns in designing the composites and they also possess high load carrying capacity, stiffness and high energy-absorption capability.
Now the specimen was hit by the instrumented impact hammer, and the response was measured by the accelerometer and the suitable data acquisition card used for the analysis.
These data were then transferred to a computer using the FFT analysis, where the natural frequencies were estimated.
The microwave oven with data logger was used to heat the temperature of the specimens.
(1) PC Data Logger oven Data acquisition card Accelerometer Impact hammer Sandwich Specimen Clamped end Fig. 2.

Using software systems Python and TensorFlow, an artificial neural network was first set up to determine mechanical material parameters (output data) from punching force curves (input data).
Further Python libraries used for the presented investigations are Numpy for data preparation purposes, Matplotlib for plotting and pandas for reading data from measuring protocols.
Data augmentation represents a commonly used procedure to generate data with high diversity without special experimental effort and thus to improve the training process of neural networks [22].
Therefore the amount of data or the number of measurement curves was virtually expanded.
The remaining 5% finally were used as an evaluation data set.