Search results

In this paper, we give a fuzzy decision tree (simply FDT) induction algorithm, named FDTAmbig, to handle the classification with discrete attributes through the uncertainty reduction.
FDTAmbig selects the attribute which will cause the further reduction of uncertainty as the expanded attribute for each decision node.
We can regard this instance as a noise data.
However, FDTAmbig with q=0.95 can recognize this noise data effectively and terminates the growth opportunely when no uncertainty reduction occurs.
Liu, Multi-valued attribute and multi-labeled data decision tree algorithm, International Journal of Machine Learning and Cybernetics 2 (2) (2011) 67-74

Main part of preparation work is simplification of 3D-CAD data to decrease mesh scale and without impacting the solution accuracy.
In this study, automatic model preparation method is developed by using of geometrical and topological information of 3D-CAD data.
Several thousand bumps can exist in manufacturing 3D-CAD data.
We can easily detect it using geometrical data which is described as conic or cylindrical face in CAD data.
The purpose of this study is reduction of period to create CAE model from 3D-CAD data.

From this data, the reduction potentials were determined for each metal, finding the values of -0.0024, -1.1274 and -0.5892 V vs calomel for Cu, Ni and Zn, respectively.
Displacement in the reduction potential with the increase of the metal concentration in the leaching solution was observed.
Concentrations were calculated for each metal in order to produce the data needed in the thermodynamic study using Pourbaix type diagrams, by the Hydra/Medusa software [13] and to evaluate the possibility for the selective metal recovery by electrochemical methods.
From the above diagrams, it was possible to observe the influence of the concentration of each metal on the reduction potential.
Reduction potential values determined for the electrodeposition of leached metals.

Based on the reduction, obtain the optimal decision support system.
Introduction VPRS[1] (Variable Precision Rough Sets) is characterized by data analysis methods do not require any prior knowledge of the data itself, only using the information provided can be achieved on the data attribute reduction and have the access to the minimum expression of knowledge.
But it fault tolerance and generalization ability is weak, can only deal with quantized data.
Bearing fault diagnosis model based on variable precision rough set theory and neural network technology 2.1 The building of a fault diagnosis model Bearing fault diagnosis model based on Variable Precision Rough Set Theory has the following main steps: ① The collected bearing failure data as the domain U, determine the condition attribute set C and decision attribute set D;② Discrete attribute values ​​for continuous processing to form a decision table ;③ Calculation condition beta attribute reduction of decision table, to obtain the relative minimal set of attributes ;④ Build sub neural networks for each selected reduction, using the simplified decision table to train sub-networks ; ⑤ Merge output of each subnet, and get effectively converged network architecture;⑥ Predict new samples and achieve the final result . 2.2 Specific steps ( 1 ) determine the condition attribute set and decision attribute set Perform statistical analysis on the collected data of bearing failure ,
table 3 , in this paper, in order to identify the strong attributes and patterns in the data, β is set to become a higher value , orderβ=0.95,this step to achieve 6-reduction as shown in table 3.

The user data is divided into several data blocks which are stored in different storage nodes, based on data fragmentation method of erasure code.
The user data is divided into several data blocks which are stored in different storage nodes, based on data fragmentation method of the erasure code.
According to the principle of erasure code, the lack of a single block will not affect the data reduction, so reliability of data is improved greatly.
Reconstruction of the source data is correct, and is the same as the data before coding.
Verification of data fragmentations can ensure the reliability and integrity of the user data.

Thus, this study aims to achieve a balance between noise reduction and backpressure minimization in muffler design.
In this study noise reduction parameters use TL and minimizing backpressure use PL.
Table 2 Description of Geometrical Data and Performances Configuration x1 (mm) x2 (mm) x3 (mm) TL (dBA) PL (kPa) Best TL pareto front (solution 1) 30.00 190.00 345.34 26.06 2.27 Best PL pareto front (solution 3) 48.85 110.52 418.89 8.36 1.87 Initial design 40 140 390 15.63 2.12 Best TL (sol. 90) 30.58 134.01 416.67 19.24 2.11 Best PL (sol. 99) 30.92 110.21 418.51 15.70 2.00 Compromise (sol. 41) 30.26 121.84 418.25 17.78 2.07 From Table 2, it can be seen carefully that the TL and PL values of the initial design are surpassed by other designs.
Solution 1 is a muffler design solution when referring to user references who have the highest TL muffler desire which means the best in noise reduction.
Razavi, “Investigation of the Efficiency of Various Reactive Mufflers by Noise Reduction and Transmission Loss Analyses,” J.

The grain size of the cold-rolled specimens decreased with increase of reduction rate, c.f., as the rolling reduction increased to 90%, grain size along the direction normal to the sheet decreased to about 8μm in thick.
At a reduction rate of 50%, the elastic deformation resulted in a further downward shift and an increase in diffraction peak width at the peak position than the reduction rate of 20%.
For the reduction rate of 90%, the peak width is further increased while the angle remains unchanged.
Obviously, the transition time decreases as the reduction rate and the annealing temperature increases.
Here, the JMAK model incorporates microhardness data, instead of relying on microstructure and use a unified equation.

Many algorithms for fitting and noise-reduction of range data from single feature have been proposed.
Many researches have been taken on fitting and noise-reduction method of range data.
Fitting and Noise-reduction Algorithm for Single Standard Profile Feature Since what we got from metrological system is range data, fitting of single profile feature in the pattern is the premise.
Then new profile feature fitting process is done through the left range data, as well as the same noise reduction process.
We simulated the range data of 4 planes and 2 cylinders according to method of generating range data proposed in Geometrical product specifications (GPS) and verification (ISO 10360-6-2001), then we added different equally proposed random noise and transformed the range data to a specified location in space and used the range data as original data to test the validity and robustness of our algorithm.

Dosage in the 50 ~ 100mg/l to meet the water when the wellbore is less than 25% of the wax viscosity and single-tube process requirements gathering, can be compared to the multi-port wells, monitoring data before and after dosing showed that the crude oil after dosing freezing point decreased 5.6 ℃, the average rate of 87.4% viscosity reduction, the average rate of 91.5% paraffin.
Faced with this situation provides a well technician DRA(Drag reduction agent ) oil soluble formulation and preparation method.
Dosage in the 50 ~ 100mg/l to meet the water when the wellbore is less than 25% of the wax viscosity and single-tube process requirements gathering, can be compared to the multi-port wells, monitoring data before and after dosing showed that the crude oil after dosing freezing point decreased 5.6 ℃, the average rate of 87.4% viscosity reduction, the average rate of 91.5% paraffin.
Conclusion (1)After dosing oil solidifying point, viscosity, and a significant reduction in the amount of wax, can be compared by multi-port wells, monitoring data before and after dosing showed that the freezing point of crude oil after dosing decreased 5.6 ℃, the average rate of 87.4% viscosity reduction, The average rate of 91.5% paraffin.
(2) The oil-soluble DRA can be provided with the double functions of the drag reduction in single-tube gathering process and pipeline back pressure reduced

(3) According to the data in Fig.4, the relationship between the surface resistivity and Bi2O3 content can be described as followed equation (Eq. 4), and the linear correlation coefficient is 0.991, as shown in Fig. 5.
What’s more, the slope of experiment results is higher than data of Eq. 3, indicated that Bi2O3 has a more significant effect on reducing the surface resistivity of MCP glasses.
While an obvious diffraction peak appeared after reduction.
Fig. 7 XRD patterns of glass sample 2: (a) before reduction; (b) after reduction.
These can be confirmed by the spectral transmission curve after reduction.