CN109107744A - A kind of medium-speed pulverizer coal-air ratio and oil pressure dynamic become excellent method of adjustment - Google Patents
A kind of medium-speed pulverizer coal-air ratio and oil pressure dynamic become excellent method of adjustment Download PDFInfo
- Publication number
- CN109107744A CN109107744A CN201810617355.3A CN201810617355A CN109107744A CN 109107744 A CN109107744 A CN 109107744A CN 201810617355 A CN201810617355 A CN 201810617355A CN 109107744 A CN109107744 A CN 109107744A
- Authority
- CN
- China
- Prior art keywords
- coal
- oil pressure
- medium
- speed pulverizer
- adjustment
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C25/00—Control arrangements specially adapted for crushing or disintegrating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B02—CRUSHING, PULVERISING, OR DISINTEGRATING; PREPARATORY TREATMENT OF GRAIN FOR MILLING
- B02C—CRUSHING, PULVERISING, OR DISINTEGRATING IN GENERAL; MILLING GRAIN
- B02C23/00—Auxiliary methods or auxiliary devices or accessories specially adapted for crushing or disintegrating not provided for in preceding groups or not specially adapted to apparatus covered by a single preceding group
Landscapes
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Crushing And Grinding (AREA)
- Disintegrating Or Milling (AREA)
Abstract
The invention discloses a kind of medium-speed pulverizer coal-air ratios and oil pressure dynamic to become excellent method of adjustment, including regularly updates sample data;Fineness of pulverized coal On-line Estimation value is obtained by the soft-sensing model that the method for least squares support vector machines establishes fineness of pulverized coal;Limit subsystem boundary condition range;Using K-means clustering algorithm, the synchronous cluster of multivariable is carried out to primary air flow and load oil pressure in each typical power output neighborhood of coal pulverizer, excavate under the conditions of different inlet conditions corresponding to mill load oil pressure and the corresponding the smallest class of conversion unit consumption of primary air flow;Using central point as the primary air flow of the medium-speed pulverizer in each typical power output neighborhood section and the operation a reference value of load oil pressure;Finally it is fitted medium-speed pulverizer operating parameter curve.The method of the present invention has obtained primary air flow and load oil pressure a reference value curve by online real-time cluster, reduces coal-grinding unit consumption as far as possible under the premise of meeting three norms, realizes the excellent adjustment that dynamically becomes.
Description
Technical field
The present invention relates to a kind of coal pulverizer coal-air ratio and adjustment of oil pressure method, more particularly to a kind of medium-speed pulverizer coal-air ratio and
Oil pressure dynamic becomes excellent method of adjustment.
Background technique
Medium-speed pulverizer is widely applied because of the advantages such as its start-stop time is short, power consumption is low in coal-burning power plant of China.In common
Speed coal mill mills component under spring force, hydraulic coupling or other external forces, and raw coal therebetween is squeezed and milled, final broken
It is broken into coal dust;By component rotation of milling, broken coal dust is thrown to vane room, flows through the stream of hot air of vane room for these coals
Powder takes the coal dust classifier on medium-speed pulverizer top to, and excessively thick coal dust is separated to be regrinded again.In this process, it removes
Outside coal-supplying amount, a variety of input quantities include air quantity, the adjusting for rolling external force or even separator etc., affect the height of coal pulverizer power consumption
It is low.However, the medium-speed pulverizer operation of majority power plant is mainly provided with coal pulverizer producer at present operating parameter or field experience carry out
Operation.And the operating parameter reference value that producer provides is mostly foreign factory settings, lacks and combines with scene operation concrete condition
Operating parameter, cause to export coal dust phenomenon excessively up to standard or not up to standard and happen occasionally, the abrasion of medium-speed pulverizer grinding roller increases.Using
Artificial field experience, and be theoretically unsound, cause the event of maloperation to happen occasionally.It after all is extremely to rely on other factory of state
The operating parameter that family provides is not bound with live practical.
A reference value refers to the corresponding each parameter value of unit optimum condition under current running boundary, general multi-pass cross design value,
Variable condition calculation, test etc. determine a reference value, may be based on the theory of similarity and live thermal test result determines that unit gives work
Optimization target values under condition, or optimal operating parameter is determined using lagrangian optimization algorithm.However actual moving process is past
There are deviations, such as inferior equipment function, coal quality to deviate all possible Influencing Mechanism model of the factors such as design conditions for past and mechanism
Reliability and accuracy.
Summary of the invention
In view of the above-mentioned defects in the prior art, the present invention provides a kind of medium-speed pulverizer coal-air ratios and oil pressure dynamic to become excellent adjustment
Method solves the problem of that there are operating parameters when deviation to adjust in due course for actual moving process and setting operating condition.
Technical solution of the present invention is as follows: a kind of medium-speed pulverizer coal-air ratio and oil pressure dynamic become excellent method of adjustment, including following step
It is rapid:
Step 1 regularly updates collecting sample data;
Step 2, to obtain coal dust by the soft-sensing model that the method for least squares support vector machines establishes fineness of pulverized coal thin
Spend On-line Estimation value;
Step 3 limits several boundary condition ranges of subsystem;
Step 4 divides several typical power output neighborhood sections in medium-speed pulverizer power output range;
Step 5 to primary air flow and load oil pressure and carries out class calculation and arrive for 2 respectively in each typical power output neighborhood section
The k-means of K is clustered, and K is more than or equal to 2;
Step 6 calculates Silhouette cluster effective evaluation function, determines class calculation;
Step 7 finds optimal Silhouette profile diagram to the k-means cluster of determining class calculation, finds class central point
The corresponding conversion the smallest class of unit power consumption of coal pulverizing in place is benchmark class;
Step 8 as the primary air flow of the medium-speed pulverizer in each typical power output neighborhood section and adds the central point of benchmark class
Carry the operation a reference value of oil pressure;
The operation a reference value of step 9, the primary air flow for the medium-speed pulverizer in neighborhood sections of being contributed with different typical cases and load oil pressure
It is fitted medium-speed pulverizer operating parameter curve.
Further, in the step 2 least squares support vector machines Radial basis kernel function are as follows:
Wherein xiFor kernel function center, if x and xiVery close so kernel function value is 1, if differing greatly so core letter
Numerical value is approximately equal to 0.Since this function is similar to Gaussian Profile, therefore also referred to as gaussian kernel function.σ is the width parameter of function, σ control
The radial effect range of function is made.
Further, boundary condition range includes that environment temperature, coal quality, disintegrating outlet temperature, coal dust are thin in the step 3
Degree, air preheater outlet temperature, mill heat primary air main-piping pressure.
Preferably, the environment temperature is -10~0 DEG C, and the coal quality is 17.19~17.89MJ/kg, the disintegrating outlet
Temperature is 70~80 DEG C, and the fineness of pulverized coal is R90< 27%, the air preheater outlet temperature is 290~320 DEG C, the mill heat
First air main-piping pressure is 8~9kPa.
Preferably, K is more than or equal to (sample data quantity) in the step 51/2And closest to (sample data quantity)1/2
Integer.
Preferably, acquisition time interval is not more than 1min when regularly updating collecting sample data in the step 1, periodically more
New time interval is not more than 10d.
Preferably, typical power output neighborhood section is not less than 6 in the step 4.
Preferably, in the step 4 typical power output neighborhood section be 40 ± 0.2t/h, 45 ± 0.2t/h, 50 ± 0.2t/h,
55 ± 0.2t/h, 60 ± 0.2t/h and 65 ± 0.2t/h.
Technical solution of the present invention is built by regularly updating collected sample by the method for least squares support vector machines
The soft-sensing model of vertical fineness of pulverized coal, obtains fineness of pulverized coal On-line Estimation value, further utilizes K-means clustering algorithm, is grinding
To primary air flow cluster synchronous with load oil pressure progress multivariable under each typical power output of coal machine, different inlet conditions conditions are excavated
Lower corresponding mill load oil pressure and primary air flow a reference value, reduce coal-grinding list as far as possible under the premise of meeting three norms
Consumption.A reference value can be found by historical data and real time data.Algorithm is easy easily to be realized, can be applied to scene.In unit reality
In operation, according to new collected achievement data, dynamic adjusts a reference value operating condition library, to approach under mill current operating conditions
Optimum operating condition, to provide Operating Guideline opinion for field operator.
Detailed description of the invention
Fig. 1 is that medium-speed pulverizer coal-air ratio and oil pressure dynamic become excellent method of adjustment flow diagram.
Fig. 2 is Silhouette profile diagram under inhomogeneity calculation K in embodiment.
Fig. 3 is cluster result figure under 55t/h typical condition in embodiment.
Fig. 4 is powder power consumption difference schematic diagram processed under identical power output in embodiment.
Fig. 5 is Clustering Effect figure under 55t/h typical condition in embodiment.
Fig. 6 is primary air flow a reference value distribution map in embodiment.
Fig. 7 is load oil pressure a reference value distribution map in embodiment.
Specific embodiment
Below with reference to embodiment, the invention will be further described, but not as a limitation of the invention.
Incorporated by reference to shown in Fig. 1, by taking certain power plant 600MW Subcritical Units as an example, pulverized coal preparation system is cold First air positive pressure direct
Formula, six coal pulverizer model ZGM123 are blown, fire coal is train bituminous coal, and composition variation is little.Middling speed involved in the present embodiment
Mill coal-air ratio and oil pressure dynamic become excellent method of adjustment, comprising:
Step 1, in SIS (plant information monitoring system), choose the history data that B in January, 2015 is ground one month,
It is female that data label is coal-supplying amount (power output), disintegrating outlet temperature, environment temperature, air preheater export an average wind-warm syndrome, heat primary air
Pipe pressure, mill primary air flow, load oil pressure, conversion powder power consumption processed.Sampling interval is 1min, through data cleansing (remove break point,
Obvious abnormal point) after, amount in 37088 groups of sample deposit floor data libraries, is with 10d to the update of database in actual motion
Interval carries out.Since the situation of change of coal quality cannot be directly read from power plant's SIS system, and current measurement means are difficult to reality
Therefore the on-line analysis of existing coal quality selects the low heat valve of the as-fired coal in day registration form to characterize coal characteristic.
Step 2, to obtain coal dust by the soft-sensing model that the method for least squares support vector machines establishes fineness of pulverized coal thin
Spend On-line Estimation value, the Radial basis kernel function of supporting vector machine model are as follows:
Wherein xiFor kernel function center, if x and xiVery close so kernel function value is 1, if differing greatly so core letter
Numerical value is approximately equal to 0.Since this function is similar to Gaussian Profile, therefore also referred to as gaussian kernel function.σ is the width parameter of function, σ control
The radial effect range of function is made.
Use all parameters for characterizing coal pulverizer operating status and can get measured value as fineness of pulverized coal hard measurement mould
The input of type, fineness of pulverized coal is as output.R90=f (B, Q, Tin,poil, in Δ p) formula, B is coal-supplying amount, th-1;Q is primary
Air quantity, th-1;TinFor import wind-warm syndrome, DEG C;poilOil pressure, MPa are loaded for mill;Δ p is inlet and outlet differential pressure, kPa;R90For coal dust
Fineness, %.
Step 3 limits subsystem boundary condition range environment temperature as -10~0 DEG C, and coal quality is 17.19~17.89MJ/
Kg, disintegrating outlet temperature are 70~80 DEG C, fineness of pulverized coal R90< 27%, air preheater outlet temperature is 290~320 DEG C, and it is primary to grind heat
Wind main-piping pressure is 8~9kPa.
Under step 4, coal pulverizer actual operating mode, power output is maintained between 35~70t/h.Respectively in six typical conditions
40 ± 0.2t/h, 45 ± 0.2t/h, 50 ± 0.2t/h, 55 ± 0.2t/h, 60 ± 0.2t/h and 65 ± 0.2t/h carry out neighborhood area
Between divide.
Step 5, (setting i is in 1 to 6 traversal, the neighborhood of corresponding six typical conditions in each typical power output neighborhood section
Section) the k-means cluster that class calculation is 2 to K is carried out with oil pressure is loaded to primary air flow respectively, K is more than or equal to 2, cf=(sample
Notebook data quantity)1/2;
Step 6, Fig. 2 give the sample under 55 ± 0.2t/h typical condition and carry out k-means cluster, and K value is 11
When Silhouette profile diagram.It can be seen that under the operating condition, each data point Silhouette value is polymerized to when 4 class closer to 1, and
The opposite unreasonable point of 5~11 classes is less.Therefore, typical condition Types Below selection 4 is more reasonable.
Step 7, the k-means for being 4 to class calculation are clustered, and cluster result is as shown in figure 3, find optimal Silhouette
Profile diagram, finding the corresponding conversion the smallest class of unit power consumption of coal pulverizing of class center is benchmark class, in partial enlargement Fig. 4 near 55t/h
Conversion powder power consumption-coal-supplying amount figure.It can be seen that the part that doping is not separated by may be not consider coal accurately together
Caused by the moisture of matter, ash content, grindability and other restrictive conditions, but general trend is bright and clear, it may be said that this cluster reaches substantially
Ideal effect is arrived.Since each cluster centre point is to extract under the current efficiency operation level of unit from mass data,
The optimization operation potentiality of coal pulverizer can be truly reacted, therefore a reference value that clustering algorithm determines has the advantages that reality is reachable.
Step 8, Fig. 5 point out that the corresponding conversion powder power consumption processed of the 4th class cluster centre is smaller, therefore select the cluster of the 4th class
Primary air flow 94.18t/h, the mill load oil pressure 12.79MPa at center are as this operating condition target component a reference value.
The operation base of step 9, the primary air flow for the medium-speed pulverizer in neighborhood sections that all different typical cases are contributed and load oil pressure
Quasi- value fitting medium-speed pulverizer operating parameter curve, result are as shown in Figure 6,7.
Claims (8)
- The excellent method of adjustment 1. a kind of medium-speed pulverizer coal-air ratio and oil pressure dynamic become, which comprises the following steps:Step 1 regularly updates collecting sample data;Step 2 obtains fineness of pulverized coal by the soft-sensing model that the method for least squares support vector machines establishes fineness of pulverized coal and exists Line estimated value;Step 3 limits several boundary condition ranges of subsystem;Step 4 divides several typical power output neighborhood sections in medium-speed pulverizer power output range;Step 5 to primary air flow and load oil pressure and carries out class calculation and arrive K's for 2 respectively in each typical power output neighborhood section K-means cluster, K are more than or equal to 2;Step 6 calculates Silhouette cluster effective evaluation function, determines class calculation;Step 7 finds optimal Silhouette profile diagram to the k-means cluster of determining class calculation, finds class center pair It is benchmark class that the smallest class of unit power consumption of coal pulverizing, which should be converted,;The central point of benchmark class as the primary air flow of the medium-speed pulverizer in each typical power output neighborhood section and is loaded oil by step 8 The operation a reference value of pressure;Step 9 is fitted with the primary air flow of medium-speed pulverizer and the operation a reference value of load oil pressure in different typical power output neighborhood sections Medium-speed pulverizer operating parameter curve.
- The excellent method of adjustment 2. medium-speed pulverizer coal-air ratio according to claim 1 and oil pressure dynamic become, which is characterized in that the step The Radial basis kernel function of least squares support vector machines in rapid 2 are as follows:Wherein xiFor kernel function center, σ is the width parameter of function.
- The excellent method of adjustment 3. medium-speed pulverizer coal-air ratio according to claim 1 and oil pressure dynamic become, which is characterized in that the step Boundary condition range includes environment temperature, coal quality, disintegrating outlet temperature, fineness of pulverized coal, air preheater outlet temperature, mill heat one in rapid 3 Secondary wind main-piping pressure.
- The excellent method of adjustment 4. medium-speed pulverizer coal-air ratio according to claim 3 and oil pressure dynamic become, which is characterized in that the ring Border temperature is -10~0 DEG C, and the coal quality is 17.19~17.89MJ/kg, and the disintegrating outlet temperature is 70~80 DEG C, the coal Powder fineness is R90< 27%, the air preheater outlet temperature is 290~320 DEG C, the mill heat primary air main-piping pressure for 8~ 9kPa。
- The excellent method of adjustment 5. medium-speed pulverizer coal-air ratio according to claim 1 and oil pressure dynamic become, which is characterized in that the step K is more than or equal to (sample data quantity) in rapid 51/2And closest to (sample data quantity)1/2Integer.
- The excellent method of adjustment 6. medium-speed pulverizer coal-air ratio according to claim 1 and oil pressure dynamic become, which is characterized in that the step Acquisition time interval is not more than 1min when regularly updating collecting sample data in rapid 1, regularly updates time interval no more than 10d.
- The excellent method of adjustment 7. medium-speed pulverizer coal-air ratio according to claim 1 and oil pressure dynamic become, which is characterized in that the step Typical power output neighborhood section is not less than 6 in rapid 4.
- The excellent method of adjustment 8. medium-speed pulverizer coal-air ratio according to claim 1 and oil pressure dynamic become, which is characterized in that the step Typical power output neighborhood section is 40 ± 0.2t/h, 45 ± 0.2t/h, 50 ± 0.2t/h, 55 ± 0.2t/h, 60 ± 0.2t/h in rapid 4 With 65 ± 0.2t/h.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810617355.3A CN109107744B (en) | 2018-06-15 | 2018-06-15 | Medium-speed mill air-coal ratio and oil pressure dynamic optimization-approaching adjusting method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810617355.3A CN109107744B (en) | 2018-06-15 | 2018-06-15 | Medium-speed mill air-coal ratio and oil pressure dynamic optimization-approaching adjusting method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109107744A true CN109107744A (en) | 2019-01-01 |
CN109107744B CN109107744B (en) | 2020-09-01 |
Family
ID=64821880
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810617355.3A Active CN109107744B (en) | 2018-06-15 | 2018-06-15 | Medium-speed mill air-coal ratio and oil pressure dynamic optimization-approaching adjusting method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109107744B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113385294A (en) * | 2021-02-02 | 2021-09-14 | 浙江浙能技术研究院有限公司 | Energy-saving control method for reducing coal grinding unit consumption of medium-speed coal mill by adjusting speed change of rare earth motor |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101038277A (en) * | 2007-04-19 | 2007-09-19 | 东北大学 | Soft measurement method for coal power fineness in powdering producer |
CN104374675A (en) * | 2014-10-15 | 2015-02-25 | 国家电网公司 | Coal mill pulverized coal fineness online monitoring method |
CN106622620A (en) * | 2016-09-27 | 2017-05-10 | 华北电力大学(保定) | Medium-speed coal mill model building method based on system dynamics |
-
2018
- 2018-06-15 CN CN201810617355.3A patent/CN109107744B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101038277A (en) * | 2007-04-19 | 2007-09-19 | 东北大学 | Soft measurement method for coal power fineness in powdering producer |
CN104374675A (en) * | 2014-10-15 | 2015-02-25 | 国家电网公司 | Coal mill pulverized coal fineness online monitoring method |
CN106622620A (en) * | 2016-09-27 | 2017-05-10 | 华北电力大学(保定) | Medium-speed coal mill model building method based on system dynamics |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113385294A (en) * | 2021-02-02 | 2021-09-14 | 浙江浙能技术研究院有限公司 | Energy-saving control method for reducing coal grinding unit consumption of medium-speed coal mill by adjusting speed change of rare earth motor |
Also Published As
Publication number | Publication date |
---|---|
CN109107744B (en) | 2020-09-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US20130030573A1 (en) | Computer-based method and device for automatically providing control parameters for a plurality of coal mills supplying coal powder to a plant | |
CN109541168B (en) | Coal powder economic fineness on-line monitoring and adjusting method | |
CN107297269B (en) | The control method of Material Level In Ball Mills | |
CN111047104B (en) | Energy consumption optimization method of grinding system | |
CN108393146A (en) | A kind of ball mill pulverizing system adaptive optimal decoupling control method | |
CN109238760B (en) | Online monitoring method of coal mill of intelligent power plant coal-fired generator set based on typical correlation analysis and slow characteristic analysis | |
CN108672025B (en) | A kind of multi-state optimal decoupling control method of steel ball coal-grinding pulverized coal preparation system | |
CN106391286A (en) | System and method for controlling fineness of pulverized coal | |
CN109107744A (en) | A kind of medium-speed pulverizer coal-air ratio and oil pressure dynamic become excellent method of adjustment | |
CN109342279B (en) | Mixed soft measurement method based on grinding mechanism and neural network | |
Gao et al. | Optimization control of a pulverizing system on the basis of the estimation of the outlet coal powder flow of a coal mill | |
CN106594793A (en) | Outlet temperature control optimizing method of medium-speed coal pulverizer of thermal power generating unit | |
US20130006428A1 (en) | System and associated method for monitoring and controlling a power plant | |
CN111443598A (en) | Cement vertical mill control method and device | |
CN111612212A (en) | On-line optimization model updating method for coal powder fineness of coal mill | |
CN109821646A (en) | A kind of coal pulverizer outlet pulverized coal flow flexible measurement method | |
CN110296431A (en) | A kind of fineness of pulverized coal automatic regulating system and method based on static separation device | |
CN113743672B (en) | Online analysis and operation optimization guiding method for economic coal fines fineness of coal mill | |
CN111651910B (en) | Coal mill-oriented sectional probabilistic state monitoring method | |
Lian et al. | Research on slag grinding process control | |
Zhang et al. | Application of neural network predictive control in cement combined grinding | |
JP2015530922A (en) | Grinding method | |
Simmons et al. | Operational experience from the United States' first vertical roller mill for cement grinding | |
CN217962902U (en) | Online adjustable static separator of coal mill | |
CN113792255B (en) | Method for calculating coal mill outlet coal powder fineness on line based on power plant big data |
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 |