CN107341281A - A kind of coal unit dust concentration analysis method based on big data technology - Google Patents
A kind of coal unit dust concentration analysis method based on big data technology Download PDFInfo
- Publication number
- CN107341281A CN107341281A CN201611125089.XA CN201611125089A CN107341281A CN 107341281 A CN107341281 A CN 107341281A CN 201611125089 A CN201611125089 A CN 201611125089A CN 107341281 A CN107341281 A CN 107341281A
- Authority
- CN
- China
- Prior art keywords
- data
- mrow
- electric field
- concentration
- dust
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Abstract
The present invention relates to a kind of coal unit dust concentration analysis method based on big data technology, specifically use big data analytical technology, depth analysis is carried out in different load operating mode, not under the same electric field method of operation to coal-fired minimum discharge unit dust concentration, and then data basis is provided to assessing the confidence level of current smoke dust discharge concentration Monitoring Data and the reliability of measuring instrument.
Description
Technical field
The invention belongs to energy field, and in particular to a kind of coal unit dust concentration analysis side based on big data technology
Method.
Background technology
With the development that deepens continuously that China's energy-saving and emission-reduction work, coal unit pollutant emission standard requires more and more tighter
It is severe.2014, the department such as National Development and Reform Committee, Chinese Ministry of Environmental Protection, which combines, to formulate《The energy-saving and emission-reduction of coal electricity upgrade and transformation action plan
(2014-2020)》, it is proposed that coal unit reaches the requirement of gas turbine group pollutant emission standard, i.e., so-called ultralow
Discharge, it is desirable under the conditions of benchmark oxygen content 6%, flue dust, sulfur dioxide, discharged nitrous oxides concentration are respectively no higher than 10,35,
50mg/Nm3.Wherein, dust concentration have dropped at least 50% (note compared with 2011 editions atmosphere pollutants emission standardses：2011 editions air
In pollutant emission standard, dust concentration emission limit is 20mg/Nm for key area3)。
In response to the call of national policy, Jiangsu Province's coal unit started Efforts To Develop minimum discharge in 2014 and changed
Make, Jiangsu Electric Power Technology Co in a government office with the support of relevant policies, be directed to carrying out minimum discharge system
Online data of uniting monitoring and minimum discharge electricity price appraisal management work, coal unit minimum discharge system related data is implemented to join
Net is integrated, and mouth dust concentration, oxygen amount, temperature, humidity, pressure and dust pelletizing system are arranged including load condition parameter and chimney
The procedure parameters such as electric field secondary current, secondary voltage.
The measurement of coal unit dust concentration is all one of electricity power enterprise's focus of attention problem all the time, universal at present
The measuring method of use has the direct method of measurement and extraction-type mensuration.Because coal unit flue structure is complicated, flue gas flow field point
Often be present uneven phenomenon in cloth, interfered plus smoke components, be to impact to emission measurement result, and according to country
Promulgate《Fixed pollution source smoke discharge continuous monitoring technical specification》, the minimum discharge standard of flue dust is less than current state's domestic discipline and family rules
Fixed allowable error scope.Because coal unit Pollutant emission concentration data not only influence air quality, it is related to pollution blowdown
Fee, while the development of environmental protection subsidy electricity price assessment mode is also influenceed, therefore, how to assess dust concentration in Monitoring Data
The problem of confidence level of data and the reliability of measuring instrument become urgent need to resolve.
The content of the invention
The present invention's is directed to deficiency of the prior art, there is provided a kind of coal unit dust concentration based on big data technology
Analysis method.
To achieve the above object, the present invention uses following technical scheme：
A kind of coal unit dust concentration analysis method based on big data technology, it is characterised in that comprise the following steps：
Unit historical data is pre-processed, includes shutting down data rejecting, the calculating of electric field running status and cigarette successively
Dust concentration calculates；
Cluster analysis is carried out to the unit historical data after processing.
To optimize above-mentioned technical proposal, the concrete measure taken also includes：
The shutdown data, which are rejected, to be specifically included：The boundary condition of coal unit generator power is set, rejects to be in and stops
The data of machine state.
The electric field running status is calculated and specifically included：Based on electric field rated current and rated voltage, to electric field two
Primary current and secondary voltage are analyzed, and judge whether cleaner electric field runs.
The smoke concentration meter specifically includes：Required according to data prediction, by the original sample data conversion of unit into
Model sample data, the model sample data include time, unit load, institute once electric field running status, all secondary
Electric field running status and smoke dust discharge concentration, the smoke dust discharge concentration arrive standard state according to equation below conversion：
In formula, SOOT be conversion after smoke dust discharge concentration, SOOTActual measurementTo survey smoke dust discharge concentration, T is flue-gas temperature,
P is flue gas pressures, and X is smoke moisture, O2 actual measurementsTo survey oxygen concentration.
Unit historical data after described pair of processing carries out cluster analysis and specifically included：Using clustering algorithm to different operations
Operating mode is divided, and analyzes the data distribution of the method for operation different under every kind of operating mode.
For ready-portioned operating mode, data are drawn according to the different running method of dust pelletizing system electric field under operating mode
Point, obtain the distribution map of different running method data under specific operation.
Using Gaussian Profile to the dust concentration data set P={ p under each operating mode1,p2,...,pnBe described, wherein
N is sample number, mean μPAnd varianceCalculate according to the following formula：
The beneficial effects of the invention are as follows：Depth digging is carried out to unit load operating mode and dust removal installation related system mass data
Pick analysis, extracts the dust concentration distribution situation under specific load operating mode, specific dust removal installation running status, realizes dense to flue dust
The confidence level and measuring instrument reliability for spending Monitoring Data are assessed, and are advantageous to supervision department and are grasped the whole province's minimum discharge unit
Flue dust data run true and false situation, level monitoring and capability of fast response are lifted, ensure that the normal of minimum discharge management work has
Sequence is carried out.
Brief description of the drawings
Fig. 1 is ready-portioned operating mode schematic diagram.
Fig. 2 is the distribution map of different running method data under specific operation.
Embodiment
In conjunction with the accompanying drawings, the present invention is further explained in detail.
The present invention, which adopts the following technical scheme that, to be realized.
The first step, data prediction, unit historical data is pre-processed, include shutting down data rejecting, electric field successively
Running status calculates and dust concentration calculates three parts part.
Data are shut down to reject：Coal unit generator power boundary condition is set, rejects the data in stopped status.
Electric field running status calculates：Based on electric field rated current and rated voltage, to electric field secondary current and secondary
Voltage is analyzed, and judges whether cleaner electric field runs, and wherein secondary current criterion is rated current more than 5%,
Secondary voltage criterion is rated voltage more than 25%.
Dust concentration calculates：Require according to data prediction and (shut down data rejecting, electric field running status calculates), will be as former state
Notebook data is converted into model sample data.Model sample file include the time, unit load, institute once electric field running status,
All secondary electrical field running statuses, smoke dust discharge concentration.Wherein, smoke dust discharge concentration arrives standard shape according to equation below conversion
State：
In formula, SOOT be conversion after smoke dust discharge concentration, SOOTActual measurementTo survey smoke dust discharge concentration, T is flue-gas temperature,
P is flue gas pressures, and X is smoke moisture, O2 actual measurementsTo survey oxygen concentration.
Second step, historical data cluster analysis.
Operating mode division is come the running situation to equipment according to the different running statuses residing for the operational factor of equipment and its
Classified.Due to operating mode division be according to the characteristics of data itself to historical data carry out a kind of dividing mode, in advance simultaneously
Each operating mode particular situation is not known, therefore the use of the clustering algorithm in non-supervisory study is that a kind of extraordinary solution is thought
Road.The present invention carries out operating mode division using the hierarchical clustering algorithm AGNES of cohesion to unit historical data, and its basic step is：
1) each object is classified as one kind, N classes is obtained, only include an object per class, the distance between class and class are exactly their institutes
Comprising the distance between object；2) immediate two classes are found and are merged into one kind；3) new class is recalculated with owning
The distance between old class；2) and 3) 4) repeat, untill being to the last merged into a class.
For ready-portioned operating mode, data can be entered according to the different running method of dust pelletizing system electric field under the operating mode
Row division, obtains the distribution map of different running method data under specific operation.Using Gaussian Profile to the flue dust under each operating mode
Concentration data collection P={ p1,p2,...,pn(n is sample number) be described, its mean μPAnd varianceCalculate according to the following formula：
3rd step, dust concentration reliability assessment.
When whether the actual flue dust data of evaluation are reasonable, operating mode and electric field operational mode according to residing for the flue dust data
Data distribution the characteristics of, two kinds of situations can be divided into：
1) when its corresponding data distribution is Gaussian Profile, we can directly use the relevant nature of Gaussian Profile,
It is (μ -1.96 σ, μ+1.96 σ) to directly obtain the rational distributed area of confidence level data in the case of 95%, and μ is average, and σ is
Variance；
2) if normal distribution is not presented for residing data distribution, it can be examined using wilcoxon to test new sample
Whether notebook data meets overall distribution in the case where confidence level is 95%.
Make specific introduction, certain 600MW rank fire coal minimum discharge machine to the present invention below in conjunction with the implementation process of method
Group, three months service datas are extracted, about 120,000 datas, its dust concentration distribution situation are analyzed as follows.
Cluster analysis is carried out to unit load operating mode first, and the concentration distribution classified corresponding to operating mode is divided,
Concrete condition is as shown in table 1 and Fig. 1.
The concentration distribution that table 1 is classified corresponding to operating mode
For ready-portioned operating mode, data can be entered according to the different running method of dust pelletizing system electric field under the operating mode
Row division, obtains the distribution map of different running method data under specific operation as shown in Figure 2.
It can be obtained according to law of great number, when sample size reaches certain amount, its data distribution shows as approximate Gaussian
Distribution, therefore the sample size under different situations, can be divided into two ways by data distribution, when data volume is sufficiently large, I
It is considered that its be distributed Gaussian distributed；And when sample size is smaller, one can be examined by shapiro inspections
Whether distribution meets Gaussian Profile.The inspection is a kind of algorithm based on correlation, can be calculated a coefficient correlation, it gets over
More show that data and Gauss Distribution Fitting must be better close to 1.
Due to big data analysis method be based on the result that unit actual operating data is clustered, extraction and analysis obtains,
Therefore, when verifying current dust concentration value confidence level with higher reproducibility.And as minimum discharge unit constantly increases
More, method of operation type constantly expands, minimum discharge system operation time constantly extends, and its analysis result can obtain further
Adjustment, application effect will more accurately and reliably.
By above method, depth excavation point is carried out to unit load operating mode and dust removal installation related system mass data
Analysis, the dust concentration distribution situation under specific load operating mode, specific dust removal installation running status is extracted, to carry out to actual flue dust
The reliability assessment of concentration, it can technically judge the data influence brought due to monitoring smoke dust instrument measurement, be advantageous to supervise
Department grasps the whole province's fire coal minimum discharge unit dust concentration quality of data situation, ensures that minimum discharge management work is normally orderly
Carry out.
The above is only the preferred embodiment of the present invention, protection scope of the present invention is not limited merely to above-described embodiment,
All technical schemes belonged under thinking of the present invention belong to protection scope of the present invention.It should be pointed out that for the art
For those of ordinary skill, some improvements and modifications without departing from the principles of the present invention, the protection of the present invention should be regarded as
Scope.
Claims (7)
1. a kind of coal unit dust concentration analysis method based on big data technology, it is characterised in that comprise the following steps：
Unit historical data is pre-processed, includes the rejecting of shutdown data successively, electric field running status calculates and flue dust is dense
Degree calculates；
Cluster analysis is carried out to the unit historical data after processing.
2. a kind of coal unit dust concentration analysis method based on big data technology as claimed in claim 1, its feature exist
In：The shutdown data, which are rejected, to be specifically included：The boundary condition of coal unit generator power is set, and rejecting is in stopped status
Data.
3. a kind of coal unit dust concentration analysis method based on big data technology as claimed in claim 1, its feature exist
In：The electric field running status is calculated and specifically included：Based on electric field rated current and rated voltage, to electric field secondary current
Analyzed with secondary voltage, judge whether cleaner electric field runs.
4. a kind of coal unit dust concentration analysis method based on big data technology as claimed in claim 1, its feature exist
In：The smoke concentration meter specifically includes：Required according to data prediction, by the original sample data conversion of unit into model sample
Notebook data, the model sample data include time, unit load, institute once electric field running status, all secondary electrical fields fortune
Row state and smoke dust discharge concentration, the smoke dust discharge concentration arrive standard state according to equation below conversion：
In formula, SOOT be conversion after smoke dust discharge concentration, SOOTActual measurementTo survey smoke dust discharge concentration, T is flue-gas temperature, and P is
Flue gas pressures, X are smoke moisture, O2 actual measurementsTo survey oxygen concentration.
5. a kind of coal unit dust concentration analysis method based on big data technology as claimed in claim 1, its feature exist
In：Unit historical data after described pair of processing carries out cluster analysis and specifically included：Using clustering algorithm to different operating conditions
Divided, analyze the data distribution of the method for operation different under every kind of operating mode.
6. a kind of coal unit dust concentration analysis method based on big data technology as claimed in claim 5, its feature exist
In：For ready-portioned operating mode, data are divided according to the different running method of dust pelletizing system electric field under operating mode, obtained
The distribution map of different running method data under specific operation.
7. a kind of coal unit dust concentration analysis method based on big data technology as claimed in claim 6, its feature exist
In：Using Gaussian Profile to the dust concentration data set P={ p under each operating mode1,p2,...,pnBe described, wherein n is sample
This number, mean μPAnd varianceCalculate according to the following formula：
<mrow>
<msub>
<mi>&mu;</mi>
<mi>P</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
</mrow>
<mi>n</mi>
</mfrac>
<mo>,</mo>
<msubsup>
<mi>&sigma;</mi>
<mi>P</mi>
<mn>2</mn>
</msubsup>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msup>
<mrow>
<mo>(</mo>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
<mo>-</mo>
<msub>
<mi>&mu;</mi>
<mi>P</mi>
</msub>
<mo>)</mo>
</mrow>
<mn>2</mn>
</msup>
</mrow>
<mrow>
<mi>n</mi>
<mo>-</mo>
<mn>1</mn>
</mrow>
</mfrac>
<mo>.</mo>
</mrow>
1
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611125089.XA CN107341281B (en) | 2016-12-08 | 2016-12-08 | Coal-fired unit smoke concentration analysis method based on big data technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611125089.XA CN107341281B (en) | 2016-12-08 | 2016-12-08 | Coal-fired unit smoke concentration analysis method based on big data technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107341281A true CN107341281A (en) | 2017-11-10 |
| CN107341281B CN107341281B (en) | 2021-03-30 |
Family
ID=60222797
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611125089.XA Active CN107341281B (en) | 2016-12-08 | 2016-12-08 | Coal-fired unit smoke concentration analysis method based on big data technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107341281B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187804A (en) * | 2006-11-15 | 2008-05-28 | 北京华电天仁电力控制技术有限公司 | Thermal power unit operation optimization rule extraction method based on data excavation |
| CN104077656A (en) * | 2014-06-27 | 2014-10-01 | 国家电网公司 | Dust removal electricity price monitoring method for coal-fired power generator set |
| CN104504498A (en) * | 2014-12-04 | 2015-04-08 | 国家电网公司 | Coal-fired power generating set ultralow emission environmental protection electricity price monitoring method |
| CN104536388A (en) * | 2014-11-21 | 2015-04-22 | 国家电网公司 | Operation staff behavior analyzing and extracting method of coal-fired generating set |
| JP2015189918A (en) * | 2014-03-28 | 2015-11-02 | 新日鐵住金株式会社 | A method for measuring and repairing cracks in furnace body |
| CN105335798A (en) * | 2015-11-03 | 2016-02-17 | 国家电网公司 | Pollutant discharge capacity prediction method based on operation team characteristic analysis |
-
2016
- 2016-12-08 CN CN201611125089.XA patent/CN107341281B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101187804A (en) * | 2006-11-15 | 2008-05-28 | 北京华电天仁电力控制技术有限公司 | Thermal power unit operation optimization rule extraction method based on data excavation |
| JP2015189918A (en) * | 2014-03-28 | 2015-11-02 | 新日鐵住金株式会社 | A method for measuring and repairing cracks in furnace body |
| CN104077656A (en) * | 2014-06-27 | 2014-10-01 | 国家电网公司 | Dust removal electricity price monitoring method for coal-fired power generator set |
| CN104536388A (en) * | 2014-11-21 | 2015-04-22 | 国家电网公司 | Operation staff behavior analyzing and extracting method of coal-fired generating set |
| CN104504498A (en) * | 2014-12-04 | 2015-04-08 | 国家电网公司 | Coal-fired power generating set ultralow emission environmental protection electricity price monitoring method |
| CN105335798A (en) * | 2015-11-03 | 2016-02-17 | 国家电网公司 | Pollutant discharge capacity prediction method based on operation team characteristic analysis |
Non-Patent Citations (2)
| Title |
|---|
| DUO SUN等: "Measurement of soot temperature, emissivity and concentration of a heavy-oil flame through pyrometric imaging", 《2012 IEEE INTERNATIONAL INSTRUMENTATION AND MEASUREMENT TECHNOLOGY CONFERENCE PROCEEDINGS》 * |
| 卢青: "基于改进的K-means聚类算法的火电厂锅炉燃烧优化研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107341281B (en) | 2021-03-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106599271A (en) | Emission monitoring time series data abnormal value detection method for coal-fired unit | |
| CN106054104A (en) | Intelligent ammeter fault real time prediction method based on decision-making tree | |
| CN111103565B (en) | Data transformation method and system based on intelligent electric energy meter metering error analysis | |
| CN109190967B (en) | Carbon emission accounting method and system for thermal generator set | |
| CN110676488B (en) | Online proton exchange membrane fuel cell fault diagnosis method based on low-frequency impedance and electrochemical impedance spectrum | |
| CN102221654A (en) | Monitoring and evaluation system of electrostatic deduster operation efficiency | |
| CN109034546A (en) | A kind of intelligent Forecasting of city gas Buried Pipeline risk | |
| CN104111379B (en) | A kind of flexible parser flow of platform area line loss per unit | |
| CN110994599B (en) | Extra-high voltage AC/DC transmission output type power grid stability judgment method based on big data | |
| CN110930057A (en) | Quantitative evaluation method for reliability of distribution transformer test result based on LOF algorithm | |
| CN104331736A (en) | RBF (Radial Basis Function) neural network based supercritical boiler nitric oxide discharging dynamic predication method | |
| CN110135466A (en) | A kind of exceeded vehicle judgment method of pollutant emission and system | |
| CN113063897A (en) | Air pollutant tracing method and device | |
| CN106048131A (en) | Converter electric precipitator system-based explosion-proof control method and device | |
| CN102735799A (en) | Engine exhaust monitoring system | |
| CN116432123A (en) | Electric energy meter fault early warning method based on CART decision tree algorithm | |
| CN109856321A (en) | The determination method of abnormal high level point | |
| CN107341281A (en) | A kind of coal unit dust concentration analysis method based on big data technology | |
| CN107491882A (en) | A kind of construction method of thermal power plant's electric power green color index model | |
| Zeng et al. | Promoting low-carbon development of electric power industry in China: A circular economy efficiency perspective | |
| CN104793167B (en) | The automatic trace analysis method and system of metering automation terminal | |
| CN107506832A (en) | The hidden danger method for digging aided in is maked an inspection tour to monitoring | |
| CN108872487B (en) | Multi-pollutant online automatic monitoring device for atmosphere pollution source | |
| CN114184911B (en) | Method and device for detecting defect type of equipment and electronic equipment | |
| CN115587331A (en) | Power grid equipment operation state diagnosis and prediction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |