Search results

When we apply positive voltage to the LSMO surface, reduction is induced in local region and the region switch to HRS.
(a) Sample absorption image by 529 eV and (b) O 1s NEXAFS data of HRS LSMO and LRS LSMO.
The NEXAFS data corresponds to the absorption associated to the dipole-allowed transition where an O 1s electron is excited to unoccupied O 2p orbitals.
The second feature revealed by the HRS data is the increased absorption ‘b’.
The resistive switching was induced by reduction and the field-induced oxygen vacancies doped electrons to partially emptied Mn 3d e1g↑ band (doped hole by Ca2+).

Digital audio watermarking technology can realize owner identification, copy control and data authentication functions by hiding extra information into the audio products.
In order to guarantee the robustness and good imperceptibility of the algorithm, the watermark is embedded in the low frequency components of wavelet transform domain, and at the same time, the algorithm adopts the quantization scheme [3], the extraction of the watermark doesn't need original signal, which greatly enhance the practical application of the algorithm. 1 Digital Watermarking Embedding 1.1 Watermark image preprocessing The preprocessing algorithm includes scrambling encryption, dimensional reduction, and adding synchronous code sequence before the watermark is embedded.
This paper adopts the watermark with the size of , if it is embedded in the audio signal, it is needed to be converted to one dimension signal, that is, (2) V is the watermarking sequences after dimension reduction, W is the watermark image after scrambling
(3) Structure of the embedded data.
(2) Establish a SVM training samples and train, generate a fixed {0, 1} sequence and form a map relationship with part of the coefficients extracted in step 1, establish the training data of SVM: (3) Then use SVM to train samples, get classification model SVM '

After formula derivation Beltrami got the matrices of SVD: （1） Assuming the pure signal and additive noise signal are s[n] and v[n] respectively, where n=0,1,2,···N-1, so the received signal which contains the noise can be expressed as x[n]=s[n]+v[n], N is the length of the analysis of the data, in this paper that is the sampling points of stator current.
（3） SVD algorithm can be summed as [10]: (1) According to the sampling signal x[n], uses equation (2) to build Hankel matrix and takes half the length of the data, i.e, M=L=N/2 ,the remaining sample points discarded
Extended Prony Algorithm As early as 1795, Prony proposed the model using a linear combination of the exponential function to describe the equally spaced sampling data.
[8] Zhengwen Qian, Li Cheng, Yinghong Li, Signal Noise Reduction Method Using Singular Value Decomposition.
[10] Chunyu Kang, Xinhua Zhang, An Adaptive Noise Reduction Method Based on Singularity value decomposes.

The rough set theory can deal with historical load data, and then give new samples from original data.
Rough Sets method of Attributes reduction Rough Sets[9-12] can only deal with uncertainty and discrete data, so when the attribute values of the information is continuous, discreting it is necessary.
RS-SVM Algorithm Step 1: Use rough sets algorithm to deal with historical load data, and creat new samples to train. the historical warship load data contains 30 days power data , 84 equipments' measurable factors, wave height and wind speed.
Step 2: Select some related input variables, mainly including the past 30 day load data, previous 3-week load data, previous 3-day load data, previous 3-hour and the related data of the 9 factors.
The influence factors are the important electrical equipment load data, wind velocity data and the wave-height data of the sea area .

Circumstances of both aspects were researched by online questionnaires and data from Taobao, in order to check if there are some deviations between them and further analyze the detailed difference, reasons, and potential impacts which caused the deviations.
In this study, we researched Taobao official data cube to collect the equal information on "brand", "style”, “price" and “goods credit” for the present products recorded from top 45 brands in April.
Based on selling data from Taobao mall the comparison result to the survey data are illustrated in Fig.2.
Analysis on the derivation of actual styles and target styles Fig. 3 Comparison between actual styles and target styles in April Data from Taobao has shown that there is a wider category of menswear online than the traditional market.
Utility of Perceived Risk Reduction Strategies in B2C E-commerce, Soft Science, 10(04) (2006), p.131-135.

Contribution of this paper mainly includes the following two parts: First, proposes a fast clustering algorithm for complex data sets.
This method first put multidimensional feature data into a single value, and then use a variety of hash functions to remapping data.
This method can ensure that the same data is classified as a class, similar data is classified as a class has a large probability.
The main idea is to use multiple hash function hash data points, to ensure for each hash function that the similar data point, the higher to collision.
For the data set G, which contains N samples , for each sample, the data format is , represents recommendation algorithm for recommendation weights, arranged in order of descending.

These data maybe rearranged yielding up to one million diffractograms (one for each pixel of the CCD) corresponding to the different sample regions.
To handle this large amount of data various techniques for data reduction and processing have been developed or adapted from other disciplines (for example, data sets from remote sensing have a similar structure - i.e. images from a single scene but at different wavelengths).
The structural variation was, therefore, only perpendicular to this plane allowing integration in the direction of the weld leading to a significant data reduction.
All this led to rather noisy images requiring more sophisticated methods for data reduction [20].
Examples include the development of various sample environments like the above mentioned reaction cells for XAFS but also the implementation of data evaluation procedures to handle the huge amount of data arising from the various imaging techniques.

According to data from the Central Statistics Agency (BPS), Indonesia’s population reached 257.77 million in mid-22.
Data collected from sensors installed in various locations of the is transmitted to a centralized data collection center.
This center may operate on a cloud based platform, enabling remote monitoring and comprehensive data analysis.
Investment in Technological Infrastructure Build digital networks, software platforms, and secure data management systems.
The formulation of robust privacy regulations and secure data storage protocols is essential to mitigate these risks [21].

Apart from being a high k gate oxide, the presence of intrinsic oxygen vacancies as well as ability of Ce4+ state in the oxide to transform to Ce3+ (reduction) state and vice versa (oxidation) [1-4] have motivated research efforts to investigate CeO2 as a gate oxide on Si [5] and 4H-silicon carbide substrates [6-9].
Nevertheless, earlier, it was reported that reducibility of CeO2 happens at an onset temperature of 600°C [10], meaning that reduction of Ce4+ to Ce3+ does not happen at temperatures lower than 600°C.
In order to obtain a statistical significant value of the investigated Qeff, five data have been extracted from each sample and presented using error bars in inset of Fig. 1.
Therefore, as the post-deposition annealing time increases, formation of thicker SiOx IL indicates compensation of much of the oxygen vacancies and thus a reduction in Qeff is obtained.
In order to obtain a statistical significant value of the investigated Dit, five data have been extracted from each sample and presented using error bars.

The compensation cost of the interruptible load is shown in Eq. 6: (6) Where, is the unit load reductions cost of the interruptible load which set in contract; is the load reductions of the interruptible load at period; is the start variables of the interruptible load , which means the interruptible load is called, and means it was not called.
Table 1 Interruptible load data Load(Node) Interruption Capacity /MW Compensation Price /$*MWh Maximum Interruption times Interruption Duration /h 1(7) 4.2 10 2 2 2(19) 2.1 5 2 2 3(21) 3.5 8 2 4 4(30) 5.6 15 2 4 Then, calculate the traditional optimization model and the multi-objective optimization model of dispatching optimization problem, compare the results which include generation costs, losses, and wind power penetration.
Multi-objective optimization scheduling for hydrothermal power systems based on electromagnetism -like mechanism and data envelopment analysis.
Multi-objective unit commitment fuzzy modeling and optimization for energy-saving and emission reduction.