Search results

Benchmark of Damage Localisation Algorithms using Mode Shape Data J.
This paper presents a comparative study of three enhanced signal processing methods to locate damage on mode shape data.
The undamaged case is estimated here using a non-linear fit algorithm between experimental data and analytical mode shape [6].
Continuous Wavelet Transform (CWT) We use the ability of CWT to extract damage from mode shape data [8].
It clearly exhibits the abilities of CWT to identify damage even with noisy data.

For the time effectiveness and cost reasons, the data, including topographic maps in 1995 and 2000, stereo aerial images in 2005, and stereo IRS-P5images in 2007, are selected as the data source in this research.
According to the existing topographic maps and mining subsidence prediction data, the additional slope schematic diagrams of the Yanzhou coal mining area are acquired.
Results of the Demarcation The subsidence information data in different times are acquired by employing the analysis of overlay and differences of the DEM raster images at different phases in Arcgis.
The damaged cultivated land of the Yanzhou coal mining area is classified, including mild yield reduction area, moderate yield reduction area and out of yield area.
Dynamic monitoring of the land subsidence in mining area from multisource remote-sensing data – a case study at Yanzhou, China.

Feature selection is to retain an effective reduction of information, the purpose is to remove redundant features, the knowledge of fault diagnosis based on attributes of attribute reduction, attribute reduction methods currently there are many, including rough sets, information entropy, Fuzzy identification and many scholars have studied and applied Rough Entropy[1] ,Attribute reduction rules[2] ,Chi-square statistics[3] and so on.
But also for machine learning, the need for a large variety of operating conditions of the data sample, but in reality, very few samples of mechanical failure, lack of adequate learning machine, in fault diagnosis and is not accurately identify the fault.
This paper based on information entropy method is applied to attribute reduction in decision table is established the method for each condition attribute values in accordance with the credibility to carry out after the information entropy, information entropy largest The redundant features; then the same for other faults, at last, the results of the outcome even if the attribute reduction, and finally the entropy as a feature vector; enter the SVM for fault identification.
Example research All data used in this article in the rotor test rig Bentley simulate a variety of turbine rotor failure of the data, the rotor speed is 3500n/min, the sampling frequency is 10000, the decision table in the value of the property as a training sample, Samples in the prediction as a prediction, classification operation.
[2] Jiayang Wang, Chao Liao, Research on the Attribute Reduction and Rule Acquisition of Time-Series Data Based on Rough Entropy [J] , Journal of Hunan University(Natural Sciences) 2005, 32(4),P.112-116(in Chinese) [3] Ming Yang, Ping Yang, Zhihui Sun, An Attribute Reduetion Algorithm Basedon Rules [J], Computer Science, 2003,30(6),P.122-125(in Chinese) [4] Lixin Wei, Chongzhao Han,A Novel Attribute Reduction Method Using Chi Square Statistics [J], Computer Simulation, 2007,24（5),P.72-74,106(in Chinese) [5] Runqiang Yin, Xiaokun Huan, Zhenliang Zhang, Decision Algorithm for Finding Reduct Based on Inter-information of Rough Set[J], Journal of Yunnan Nationalities University(Natural Sciences Edition, 2006,15(1),P.12-14(in Chinese) [6] Dongbo Zhang, Huixian Huang, Yaonan Wang, Feature selection neural network ensemble based on rough sets Reducts[J], Control and Decision, 2010,25(3),P.371-377(in Chinese) [7] Zhonghe Han, Xiaoxun Zhu, Selection of Training Sample Length in Support

Data collecting and processing According to the influence factors of slope stability, we collected 45 groups’ typical examples.
Due to space limitations, this article only lists some of the data in Table 1, the complete data please see literature [8].
Six main factors which influence the stability of arc slope are rock density (), cohesion (), internal friction angle (), slope angle (), slope height () and pore ​​pressure ratio (), safety factor () is determined by the parameters in the table, the data in Table 1 are original data.
The processed data please see Table 2.
Discrete data after normalization processing x1 1 0 2 2 0 1 0 x2 1 0 2 2 0 2 0 x3 1 0 2 2 0 0 0 x43 2 1 2 2 1 0 1 x44 2 1 2 2 1 0 1 x45 2 1 2 2 1 0 1 Attribute reduction and importance calculation After calculation, C1, C3, C4, C6 are the core attributes and C2, C5 are redundant attributes.

The measuring principle of building lifecycle carbon emissions The existing problems and data base.
Much more data which can be used to measure carbon emissions have been accumulated in the course of building carbon emissions research.
We also found that the resource consumptions cannot be quantified in other parts of the jobs, but we can get the data of their cost.
However, depending on the existing data and the method in this paper, the amount of carbon emissions for single buildings is not accurate.
In order to improve the accuracy of the model, we suggest that the emphatic aspects for further study should be focused on enriching the data.

The good vibration reduction efficiency of active solutions, however, comes at a cost of high energy demands.
The work [9] explores the potential applications of a platform mechanism to vibration reduction in several DOFs.
Directly measured data yield the error signal to be used in the control process.
The vital stage of the design process involves the computer simulation of the ASC, using the pertinent data.
Another constraint acting upon the hydraulic drive is the maximal delivery of the pump supporting the drive 4, which allows for finding the maximal velocity to be implemented by the drive once the piston area has been established. 10o Simulation data yield the velocity pattern in the cylinder 4 for the maximal inputs from the road.

It analysed the change regulation about the two parameters which are rate of wheel load reduction and overturning coefficient during running of the train.
Two dynamic parameters is selected to analyze which are rate of wheel load reduction and overturning coefficient.
From the figure 3, we can see that when the vehicle runs at the speed of 50km/h cross the curves, the rate of wheel load reduction of the 1st wheelset is 0.0038.
From the data and figure 6, we can get the result that the impact of the lower speed passed the surplus superelevation is better than the higher speed passed the deficient superelevation.
For the absolute value of these data, the overturning coefficient of the deficient superelevation is the biggest.

The CO emission purification efficiency data of the DOC+CDPF+SCR system shows in Fig.4 Fig.4 The purification efficiency of CO (DOC+CDPF+SCR) As is showed in Fig.4, the percentage of the CO purification efficiency is about 60% and it increases with the increase of the temperature.Also the purification efficiency is more than 50% in most working conditions.
The NOx emission purification efficiency data of the DOC+CDPF+SCR system shows in Fig.6 Fig.6 The purification efficiency of NOx (DOC+CDPF+SCR) The conversion efficiency of SCR exists from 30% to 80%,when the exhaust gas temperature is over 250℃，the NOx purification efficiency of the SCR system is more than 50%.
Fig.9 shows the data curve of the equilibrium temperature of CDPF.
Fig.9 The data curve of the equilibrium temperature of CDPF Conclusion The purification performance of the diesel exhausts four-effect catalytic purification system which consists of DOC, CDPF and SCR is tested through the methods given in this paper.
[2] “A kind of low temperature ammonia selective catalytic reduction of denitration catalyst preparation method”, Application number: 201010539716.0[P]

However, the network needs experimental data to be trained and any configuration outside the experimental range makes the prediction to be not reliable.
For these reasons, here we propose a different approach based on model reduction, particularly in the Proper Generalized Decomposition (PGD) method [3].
And it can be shown (e.g. [6]) that solving a nonlinear problem with standard model reduction techniques and standard linearization schemes is intrinsically inefficient because the complexity of evaluating the nonlinear term remains comparable to that of the original problem.
Cueto: A short review in Model Order Reduction based on Proper Generalized Decomposition, Archives of Computational Methods in Engineering, 18, 395-404, 2011
Sorensen: Nonlinear model order reduction via discrete empirical interpolation, SIAM J.

After the estimation of kinetic parameters, the model showed good agreement with experimental data for different exhaust temperatures.
Urea-SCR systems possess an outstanding performance in NOX reduction, which can serve as a technology platform for further improvement.
According to the experimental data at different exhaust temperature, the kinetic parameters of urea thermolysis rate expression were obtained from literature [4] and then estimated by using genetic algorithm.
The standard SCR reaction plays an important role in the NOX reduction, so its activation energy has an obvious effect on the NOX conversion efficiency.
Conclusions An accurate urea-SCR model consisting of 1d/3d adblue injection model and 1d catalytic converter model was presented in this work, which showed good agreement with experimental data for different exhaust temperatures after the parameters estimation.