Search results

Section 2 briefly describes the previous CAD system, data used and experimental setup in this paper.
Data used To test the approach we used a real-world problem.
(5) Use bootstrap method to sample from the data set and build base classifiers with NN, then choose base classifiers which meet the accuracy requirement to vote for the final decision.
In addition, in forward stepwise feature selection procedure, we may also use a validation set for evaluating the performance of each combination of features for computation time reduction and reduction of overfitting.
Motoda, Feature Selection for Knowledge Discovery and Data Mining, Kluwer Academic Publishers, 1998, p.62-68.

The equilibrium sorption data are well demonstrated by Langmuir model.
Various methods for the removal of heavy metal ions from wastewater were developed, including chemical precipitation, membrane technology, electrolytic reduction, ion exchange and adsorption [2].
The equilibrium adsorption isotherm is one of the most useful data to understand the mechanism of the adsorption.
The Langmuir isotherm model proved an excellent fit to the isotherm data, giving R2 values of 0.9990 , which indicates that a monolayer adsorption process plays an important role in the whole process.
The Langmuir model gives an excellent fit of adsorption isotherm equilibrium data, which suggests that the most Cu2+ and Pb2+ adsorbed are monolayer arranged on the surface of distillers' grains.

Such data belong to mechanical properties essential for an assessment of residual life of components and structures containing cracks of length more than several millimetres.
Such data are important for an assessment of residual life of components and structures containing cracks of length more than several millimetres.
The measurement sensitivity in the region of short cracks up to 1 mm is not unfortunately increased by the reduction of specimen width.
Data of cut depth (i.e. crack length) and corresponding potentials were gradually recorded – Fig. 2.
The measurement provides an important knowledge and data for assessment of crack arrest conditions or fatigue life extension with existing short crack.

The model has been validated by numerical/experimental comparisons in terms of stiffness reduction over the number of loading cycles, considering three different applied loads.
Salkind [8] suggested to present the fatigue data in the form of S-N curves.
The calibration has been performed for both loads P1 and P2, to minimize the total error (meant as the sum of the errors for both curves) between numerical and experimental data.

More recently, studies on the building energy reduction technology are actively in progress.
Aimed to collect each element technology and weather data, temperature and humidity measuring sensors were installed in a total of points and portable equipments are used to collect and saved data.

B-Cells B-cells are considered to have the same structure as the antigen and will be used to recognize the data classes in the end.
Ground truth of Indiana image (b). legend of ground truth image with 9 selected classes (c) In order to classify data, 9 chosen class elements are extracted from the image. 70% of selected pixels are used as training data and the remaining are considered as test data to estimate accuracy.
Alizadeh, "Evaluation of the Potential of Swarm Clustering Techniques in Hyperspectral Data," University of Tehran, GIS Ostrava, January 2010 [2] M.
Feng, ”Ensemble Classification Algorithm for Hyperspectral Remote Sensing Data ,” IEEE Trans.
Fauvel, "Ensemble Methods for Classification of Hyperspectral Data," University of Iceland,© IEEE IGARSS 2008, ISBN 978-1-4244-2808, 2008 [4] A.

In the paper monotonic, cyclic and fatigue data for En295 steel were presented.
The elongation to failure and reduction in area were A5 =28,9 % and Z=32 % respectively.
It can been observed better agreement between experimental data and predictions based on Golos-Ellyin hypothesis.
The detailed monotonic, cyclic and fatigue data for En295 steel were presented.
In the case of three-step loading, smaller differences of ∑ni/Nfi =1 from experimental data were observed.

(1) Where is the desired output for training data i, is the network output for training data i, and n is the number of data in the training data set.
System description Network training and test data were obtained by extracting recent process information from the factory temperature database; a total of 1600 data patterns.
The training patterns were filtered from the data patterns such that a uniform distribution over the data range was obtained.
It is important when training neural networks that each element of the training data be sufficiently represented over the data range for that element.
This is shown in Figures 5 for the training data and test data sets.

The demand for weight reduction in automotive industry has generated great interest in research and development of magnesium applications, especially for powertrain components.
A chromel-alumel (K-type) thermocouple protected by a thin steel sheath was positioned at a distance of 0.02 m from the bottom of the crucible center, and was connected to a computer-based data acquisition system to measure the temperature variation.
In cooling analysis, the temperature of the solidifying AM60 alloy and grain-refined sample was recorded by the data acquisition system at a regular interval of 500 ms as they cooled (25 ℃/min) from the completely liquid state to solid.
The melt temperature (T) vs. time (t) data was processed and cooling curves (T vs. t) were plotted using the Microsoft Excel spreadsheet software.

Cost-effective and reliable damage detection is critical for the utilization of composite materials due to the relatively localised nature of damage formation and the resultant reduction in structural integrity.
Purpose of this work was to study the interaction between embedded PVDF (polyvinylidene fluoride) transducers and composite samples as well as detect and characterize the failure mechanisms in aramid/epoxy flexural test specimens using acoustic emission data obtained by embedded PVDF film sensors.
Furthermore, it has been realized a comparison with surface mounted PVDF data.