Search results

B.The Experimental Data Acquisition and Processing We used base oils as lubrication in the experiment, select a different variable values randomly.
In order to eliminate the possible effects of different dimension we should bring the data obtained above to standardized[5].
The above data to be converted in accordance with the following Eq.1 .
D.Experimental Verification We designed four group experiments are based on the MMW-1A vertical universal friction and wear testing machine by using control variable method, then represented the data by Figures.
[5] Fan Jincheng, Mei Changlin, Data Analysis, Peking, 2010

Dielectric constant, dielectric loss and a.c conductivity data of the NR composites are reported.
However, no a.c conductivity data of the composite was reported.
Thermal reduction of graphene oxide into graphene was carried out for 60 minutes in a furnace at 200°C.

The relationship between relative density of P/M steels and mechanical behavior is also obtained from FEM and compared with the experimental data.
The ultimate goals in gear modeling may be summarized as the stress analysis such as bending and contact stresses, reduction of surface pitting and scoring, reliability, fatigue life, transmission efficiency and radiation noise.
(1) where E0 is the Young's modulus of the fully dense steel (taken by extrapolating the experimental data to zero porosity, which yielded a value of approximately 201 GPa), and ke is a constant in terms of the Poisson's ratio of the fully dense material, ν0 [3]: ke = 2 − 3ν0
Also, according to experimental data (given in table 1), with 1.3% and 7.2% increase in average porosity, Proportional limit stress systematicaaly decreases 4% and 15.5%.
• According to experimental data (given in table 1), with 1.3% and 7.2% increase in average porosity, Proportional limit stress systematicaaly decreases 4% and 15.5%

Seismic data from the 1999 Adapazarı and Düzce earthquakes were used for the modeling.
Recorded data from the most devastating recent earthquakes that have recently affected Istanbul, namely, the Kocaeli earthquake of August 17, 1999 and the Düzce earthquake of November 12, 1999 were used for the model.
The data was obtained from the Berkeley University-PEER Strong Motion Database [6].
Consequently, the values in literature were adapted with a specific reduction rate when it was seen that the test results were not sufficient to perform a generalization for MOE.
Based on the investigation, storey drift ratios for three alternative models using data from the August 17, 1999 Kocaeli Earthquake are shown in the Table 3 and the resulting graphs are shown in the Figure 13.

Grey system theory can effectively deal with incomplete and uncertain information using only a few data points.
Grey prediction gained popularity in the past decade because of its simplicity and ability to characterize unknown system by using a few data points.
That is, only CO2 emissions data series are concerned.
MAPE for model evaluation[7] MAPE(%) Forecasting power >50 Weak and inaccurate forecasting 20~50 Reasonable forecasting 10~20 Good forecasting <10 Highly accurate forecasting Data sources CO2 emissions come from energy consumption, industry production, agriculture plant, litter emission and so on.
The forecasting data from 2011 to 2020 and carbon intensity are shown in Table 4.

The simulation results were compared to experimental data for the Sm(Co, Fe, Cu, Zr)7.5 type of alloy.
The results of the simulated experiment calculated in the FRMR program are used as input data for the Analyzer program.
Results and Discussion The demagnetization curves of the Sm(Co, Fe, Cu, Zr)7.5 alloy samples were taken to compare the simulation results and experimental data. 1.
According to the literature data of such heat treatment the alloy is a single phase solid solution with crystal structure of the hexagonal modification Sm2Co17 [4].
Conclusions The comparison of the experimental data with the results of domain structure modeling after various stages of heat treatment allows to reach the following conclusions: 1. 

Experimental results and analysis Five data sets are taken from the UCI database as experimental data.
The experimental data is divided into two parts: 2/3 of them are used for training and the other for testing.
The experimental data and result is recorded in Table 1.
Table 1 The experimental data and result Data set Total number of samples Number of sample feature Original Adaboost Improved Adaboost Original Adaboost Improved Adaboost Classification accuracy Training time(hour) Classification accuracy Training time(hour) Audiology 226 69 93.5% 0.958 93.5% 0.958 Gas Sensor 13910 128 95.8% 1.777 95.8% 1.777 Arrhythmia 452 279 94.7% 3.875 94.7% 3.875 Human Activity Recognition 10290 561 96.6% 7.792 96.6% 7.792 CNAE-9 1080 857 98.8% 11.903 98.8% 11.903 From the experimental result above, we know that, compared with the original algorithm, the improved Adaboost algorithm not only reduces the training time to a certain extent, but also reduces the classification accuracy.
The reduction of training time is related to the number of sample’s features: fewer features there are, less effective the improved algorithm is, and the final result is more close to the original algorithm.

In addition, data transfer between different processors shall be kept to a minimum as much as possible.
Data of the particles straddling more than one sub-domain and touching particles in other processers need to be shared among all sub-domains involved.
In order to make the data synchronisation simple, overlapping areas with the width identical to maximum diameter of particles are set around each sub-domain.
Hence, only the data of the particles located within the overlapping areas need to be synchronised.
As it is also observed and studied by [1], the reason was proved to be the ‘pseudo-vector processing capacity’, indicating that the reduction of the array inner loop size caused a save in runtime in the array operation for an equal amount of data.

Some researchers discuss the testing of a prototype system, as well as the analysis of the test data, and the development of the conclusions.
It can be seen from the above data, charging station in the actual operation situation of energy consumption or "load factor" varies greatly.
In the case of electricity infrastructure remains the same status, control unit of power supply and unit of data acquisition can also be used in other power supply equipments in an electric vehicle.
It connects a power grid with the inverter/charger system fits UL 1741 for safety and provides the data acquisition system.
At the same time, achieve the purpose of energy conservation and emission’s reduction.

Even though a certain number of studies have been carried out for investigating static and fatigue properties of UFG metals, very few data on fatigue crack growth resistance are available[2, 3].
Closure measurement procedure: (a) Compliance curve obtained from the gauge on the back face of the CT specimen, (b) packing per 20 data points (c) slope fitting, (d) Pop calculated after signal treatment shown schematically above.
This approach permits to reduce considerably the scatter observed when the raw data is used directly.
Fatigue crack propagation in air Figure 2 shows relationships between the fatigue crack growth rates da/dN and the stress intensity factor ΔK in the ECAPed specimen together with the reference data for conventional copper polycrystal.
Available experimental data on copper [14] are very near to these estimations where the inclination angle is reported to be 27°.