Search results

Data and Methods Statistical Data.
In cases where the MCF is not available, estimates were made adopting Eq. 2 along with the annual crop harvested area (Aharvest) data and fiber yield (Yfiber).
(Estimates Based on Published Statistical Data by PSA [13] and PhilFIDA [12].)
(Estimates Based on Published Statistical Data by PSA [13] and PhilFIDA [12], Assuming 5 Wt.% Residual Grade Fiber.)
Andrew, Global CO2 emissions from cement production, 1928–2018, Earth Syst Sci Data 11 (2019) 1675–1710. https://doi.org/10.5194/essd-11-1675-2019

Using the reduction method of the variable precision rough set,the hidden information in power transformer faults data is reduced , and the information which plays a major role in fault classification can be obtained.
The extended model of RST, Variable Precision Rough (VPRS) can overcome this restriction by introducing a probability value [7].Thevalue represents a bound on the conditional probability of a proportion of objects in a condition class which are classified to the same decision.The power transformer fault data was dealed with reduction, redundant information is removed to acquire decision-making rules.
Rough set theory Standard rough set theory.In fault diagnosis,due to the influence of various factors in the actual system, there are some inconsistence between in the test data and the actual value.
Such as in the oil dissolved gas analysis, test data inevitably contain the influence of interference and noise, so,the test data accompany with imprecision and inconsistency.
Pawlak, Rough Sets: Theoretical Aspects of Reasoning About Data,(Kluwer Academic Publishers, USA, 1991)

First, we got the data from the database, transformed the corresponding decision table, then got the data in decision-making table for further simplification, generated the final decision rules. and got good results, experimental results showed that the method provided some practical value.
Customer Classification Model Construction Data acquisition The initial decision table method using NS discrete discrete data: where, D = {d} is decision attribute set, said the types of customers, divided into 3 categories; 1 Gold customers, 2 for Silver customers, 3 copper Card.
Get customer attributes discretization results: Table 1 The initial decision table U a1 a2 a3 a4 a5 d X1 3 1 1 2 1 1 X2 3 1 1 2 3 2 X3 3 2 1 2 1 1 … … … … … … … X8 2 1 2 2 2 2 X9 1 1 3 1 1 1 X10 2 3 1 3 1 1 Attribute Reduction of Decision Table Improved attribute reduction algorithm based on logic operations Demand reduction algorithm： get DNF by conjunctive above expression : By logical operation converte disjunctive to CNF That is to say: For, First calculate,get， so valueis necessary.
Similarly, get overview Table of Value Reduction Table 3 Overview Table of Value Reduction UOB a2 a3 a5 d X1 — — 1 1 X2 — 1 3 2 X3 — — — 1 … … … … … X8 1 — — 2 X9 1 — 1 1 X10 — 1 — 1 Get from Table 3： (1) In terms of decision rules: , first consider for decision rules ,whether it is necessary relative to D (whether as the core values)?
References [1] Wang Xing, Niu Yugang, rough set theory based on customer relationship management, Shanghai: East China University of Information Science and Engineering, 2008 [2] David Hand, Heikki Mannila, Padhraic Smyth, Principles of Data Mining, Machine Press Society, April 2003 [3] He Rongqin, CRM principles, design, practice, Electronic Industry Press, January 2003 [4] Newell, Frederick.Loyalty.com: Customer Relationship Management in the New Era of Internet Marking.New York: McGraw-Hill, 1999 [5] Wang Biao, Duan Lun Chan, Wu Hao, Song Yonggang, rough sets and fuzzy sets and its application, Beijing: Electronic Industry Press ,2008,59-62, 1-2, 2 -6,172-173, 8-10

It can remove the redundant information in the original data, and extract the most useful information through data analysis directly.
If the condition attributes set reduction is a nonempty set .
Then is called a reduction, the set of all reduction credited as.
This paper collects hundreds of groups transformers historical fault data, and selects the 30 groups representative samples to transform the training set, 10 groups for test set.
Reduction of decision table.

Energy saving and emission reduction is a necessary way for China to develop low carbon economy.
Thus, there is still a great of potential for energy saving and emission reduction.
Data resource This paper focus on the interconnection between energy intensity and structure evolution and scientific technology progress from 1995 to 2010, we can get the GDP, total energy consumption and each type of energy use in industry from China statistical yearbook 2011 [1], as for the carbon dioxide emission, we take the energy coefficient from IPCC 2006 as reference, which means the carbon dioxide emission coefficient of coal, oil and natural gas are 2.74, 2.14 and 1.63 t/tce, respectively.
The aggregative scientific technology progress index comes from China scientific technology statistical yearbook 2003-2011 [5], because the lack of statistical data from 1995 to 2001, so we only get the data from 2002 to 2010.
Also based on the four models, we make a potential analysis about the two factors on energy saving and emission reduction in the future.

Generally, the technologies of noise control and vibration reduction in railway including isolation, absorption, sound insulation, and sound-absorbing [3].
With a condition of 60km/h speed, analysis results of the data from 2000 Hz to 22000 Hz dispalyed that the equivalent sound level would be reduced for up to 5.4 dBA, and the maximum sound level is diminished about 2.7 to 7.4 dBA.
The function of noise reduction by dampers is due mainly to the significant inhibition of the noise from wheel rail, whose frequency is rather high.
The vibration data of track and platform under the 2000 Hz~22000 Hz band demonstrated a decreasing trend, reflected as more than 2.2~9.6dB reduction in vibration acceleration level, and more than 0.7~8.6dB of the maxmium vibration acceleration level of tracks diminished.
The total noise of the train with a low speed is comprised by a large number of wheel noise contribution, hence, the reduction function by masked damper would be limited to some extent.

But the capacity of dealing with massive data and the efficiency of the mining fault rules are still not ideal.
Information entropy, a measure of the overall uncertainty, can be represented as a data statistical characteristic.
Fault Rules of Power Grid Mining Process Firstly, attribute selection and data cleansing on fault data of power grid, and then setting up fault samples for mining.
The original decision table is established by using fault data.
Conclusions This article uses data mining method which combines rough set attribute reduction theory with association rules in the power grid fault data.

Material flow stress properties for FEM inputs were extracted from SPF tensile test data [4] for Ti-64 FGS and RTI proprietary testing data for Ti-64 STD.
Data was fit using the power law flow stress equation.
The strain rate has less effect on YS reduction in Ti-64 FGS, which implies that the forming time may not significantly affect the reduction in YS of the final product.
Superplastic strain has direct effect on percent reduction of YS.
These results are in agreement with the tensile property data, where yield and tensile strength properties are comparable in both directions.

Data Mining Technology in Transformer Condition Assessment Principle of Data Mining.
Data preprocessing.
Data preprocessing is a crucial step before data mining, due to the obtained transformer parameters often contain a lot of noisy, incomplete or inaccurate data.
Test data should be dispersed firstly.
The ant colony algorithm is applied for reduction of transformer diagnostic data [6].

Reduction in area is affected by second phases and inclusions.
Tensile strength-reduction in area balance Figure 6(a) and (b) shows the variation of percentage elongation and reduction in area as a function of the volume fraction of cementite.
The test data of conventional ferrite-pearlite steel [12], tempered martensitic steel [12] and bainite steel [13] are also plotted in Fig. 7 for the purpose of comparison.
Reduction in area is a measure of formability as well as uniform elongation.
[12] NIMS data base, Materials Information Technology Station, National Institute for Materials Science, Tsukuba, Japan