CN107066659A - A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature - Google Patents
A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature Download PDFInfo
- Publication number
- CN107066659A CN107066659A CN201611223114.8A CN201611223114A CN107066659A CN 107066659 A CN107066659 A CN 107066659A CN 201611223114 A CN201611223114 A CN 201611223114A CN 107066659 A CN107066659 A CN 107066659A
- Authority
- CN
- China
- Prior art keywords
- learning machine
- temperature
- msub
- cement
- mrow
- 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.)
- Pending
Links
- 239000004568 cement Substances 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000005457 optimization Methods 0.000 claims abstract description 18
- 239000003245 coal Substances 0.000 claims description 26
- 239000011159 matrix material Substances 0.000 claims description 9
- 238000002485 combustion reaction Methods 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 7
- 230000004913 activation Effects 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 8
- 238000005259 measurement Methods 0.000 abstract description 5
- 239000000843 powder Substances 0.000 abstract description 3
- 239000002817 coal dust Substances 0.000 description 4
- 238000000354 decomposition reaction Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 239000002994 raw material Substances 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000010304 firing Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000007789 gas Substances 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000000197 pyrolysis Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000009423 ventilation Methods 0.000 description 1
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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J5/00—Radiation pyrometry, e.g. infrared or optical thermometry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Curing Cements, Concrete, And Artificial Stone (AREA)
Abstract
The invention belongs to the calciner temperature electric powder prediction in cement production process, the method that specifically a kind of limit of utilization learning machine predicts cement decomposing furnace temperature.This method comprises the following steps：Step 1: determining the principal element of influence calciner temperature；Step 2: gathered data；Step 3: user's group hunting algorithm carries out parameter optimization to extreme learning machine；Step 4: setting up forecast model；Step 5: predicting the outcome；Step 6: real-time optimization forecast model；The present invention is a kind of method of the prediction cement decomposing furnace temperature of utilization crowd searching algorithm optimization extreme learning machine weights, this method can accurately predict calciner temperature, there is provided support for the calciner temperature control in later stage manufacture of cement, it is to avoid the problem of existing measurement cement decomposing furnace temperature utilizes the expensive survey tools such as infrared thermometer, complex operation.
Description
Technical field
The invention belongs to the calciner temperature electric powder prediction in cement production process, specifically one kind utilizes pole
Limit the method that learning machine predicts cement decomposing furnace temperature.
Background technology
In the production process of cement, the efficiency of combustion that precalcining system is produced for clinker plays vital
Effect.Dore furnace is the core of precalcining system, therefore, it is necessary to detect calciner temperature.Because dore furnace is non-linear,
The characteristics of large dead time, it is difficult to set up accurately mathematical modeling, great difficulty is brought to predicted temperature.
Although calciner temperature, because its is expensive, complex operation can be measured with infrared thermometer, it is difficult to
Extensive use in the cement producing line of long-play.Traditional modelling by mechanism is main from kinetics angle, or
By some combustion reaction mechanisms, dore furnace model is set up.These models are primary concern is that some inside dore furnace are reacted
Mechanism, not in view of the functional relation between major influence factors and calciner temperature.Therefore, modelling by mechanism does not have too
Big reference value.
The content of the invention
The invention provides a kind of cement decomposing furnace temperature of utilization crowd searching algorithm optimization extreme learning machine weights is pre-
Survey method, this method can accurately predict calciner temperature, and branch is provided for the calciner temperature control in later stage manufacture of cement
Hold, solve the above-mentioned deficiency of existing measurement calciner temperature.
Technical solution of the present invention is described with reference to the drawings as follows：
A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature, this method comprises the following steps：
Step 1: determining the principal element of influence calciner temperature；
By dore furnace internal-combustion mechanism, with reference to the summary of experience of field personnel, it can be deduced that influence dore furnace
The principal element of temperature, which has, feeds coal amount, feeding capacity and tertiary air pressure；
Step 2: gathered data；
The data for feeding coal amount and feeding capacity are obtained from the DCS system of cement plant, is surveyed using infrared thermometer and decomposes furnace temperature
Degree；The data for feeding coal amount, feeding capacity and calciner temperature take 100 groups；
Step 3: user's group hunting algorithm carries out parameter optimization to extreme learning machine；
User's group hunting algorithm carries out optimizing to the weights of extreme learning machine, and selects best weight value；Extreme learning machine
Regression equation be：
Wherein：Y is output, xjFor input, K is the number of hidden nodes, ωijFor input weight matrix；biBiased for hidden layer
Matrix；βiFor output weight matrix；gi(ωij·xj+bi) it is activation primitive；
Step 4: setting up forecast model；
Parameter to extreme learning machine is configured；According to step one, input number of nodes：3, output node number：1, according to
Experience sets the number of hidden nodes：9, the best weight value that optimizing in step 3 is obtained is assigned to extreme learning machine；
Step 5: predicting the outcome；
Coal amount, feeding capacity and tertiary air pressure data, the model prediction obtained using step 4 are fed according to what is currently collected
Calciner temperature；
Step 6: real-time optimization forecast model；
According to predicting the outcome obtained by step 5, it is possible to use step 3 and step 4 are again to parameter optimization, again
Model is set up, precision is improved
Described step three is comprised the following steps that：
Step1：Space dimensionality in calculating crowd's searching algorithm is weights number, produces initial search person；
Step2：The weights of initialization are assigned to extreme learning machine, are trained, and it is poor with prediction data and True Data
The mean square deviation of value is used as fitness value；
Step3：Iteration optimizing, by comparing, obtains the minimum weights of fitness value；
Step4：When reaching stop condition, stop iteration, export best initial weights；Otherwise, return to Step3 and continue optimizing.
Beneficial effects of the present invention are：
1st, the model that limit of utilization learning machine of the present invention is set up only needs in cement production process DCS system data and red
The calciner temperature data that outside line temperature measurer is measured, you can model is set up by training, model prediction calciner temperature is utilized.
2nd, the problem of avoiding the expensive survey tools such as infrared thermometer, complex operation.
3rd, without correlation theories knowledge, the relevant data in DCS system are the temperature of predictable dore furnace.
4th, the forecast model set up using extreme learning machine, can be according to live operating condition, real-time correction model.
Brief description of the drawings
Fig. 1 is to feed coal amount datagram from what DCS system was obtained；
Fig. 2 is the feeding capacity datagram obtained from DCS system；
Fig. 3 is that the tertiary air obtained from DCS system presses datagram；
Fig. 4 is a kind of decomposition furnace structure schematic diagram；
Fig. 5 is another decomposition furnace structure schematic diagram；
Fig. 6 is the calciner temperature datagram that infrared thermometer is measured；
In Fig. 7 behaviour group hunting algorithm optimization parametric procedures, per the change curve of generation fitness value；
Fig. 8 is to be used for the figure that predicts the outcome of training set using the extreme learning machine of best weight value；
Fig. 9 is to be used for the figure that predicts the outcome of forecast set using the extreme learning machine of best weight value.
Figure 10 is the flow chart of this method specific implementation step
Embodiment
A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature, this method comprises the following steps：
Step 1: determining the principal element of influence calciner temperature；
By dore furnace internal-combustion mechanism, with reference to the summary of experience of field personnel, it can be deduced that influence dore furnace
The principal element of temperature, which has, feeds coal amount, feeding capacity and tertiary air pressure；
(1) influence of coal amount is fed
Actual cement production process shows, in the case that other conditions are constant in dore furnace, the temperature of dore furnace with
Feed the increase of coal amount and increase, reduced with the reduction for feeding coal amount.But when feeding coal amount more than certain value, have remaining coal
Powder, burns away into next stage preheater.Which results in the rise of the CO concentration in gas, cause preheater Coating clogging, lead
Cause calciner temperature not high.As can be seen here, when other conditions are constant in dore furnace, in calciner temperature and the function for feeding coal amount
In, feed coal amount and there is an extreme point, calciner temperature is reduced afterwards as the increase for feeding coal amount is first raised.In addition, the matter of coal dust
Amount also has certain influence on calciner temperature.Therefore, it is to influence the principal element of calciner temperature to feed coal amount.
(2) influence of feeding capacity
It can be seen from the technological process of nsp kiln, raw material are mixed in preheaters at different levels with thermal current, preheating, most laggard
Enter dore furnace.Feeding capacity is given at pre-heater inlet, is not adjustable at dore furnace.Decomposed when feeding capacity abruptly increase
Furnace temperature can be reduced；When feeding capacity rapid drawdown, calciner temperature can be raised.When feeding capacity is smaller, less, the system of raw material heat absorption
Temperature is too high, is easily caused preheater skinning；When feeding capacity is larger, local stoppages are easily caused again.Therefore, feeding capacity is to dividing
The disturbance for solving stove is very big, in order to maintain the stable operation of dore furnace, should avoid the abruptly increase rapid drawdown of feeding capacity, it is suitable to be maintained at as far as possible
In the range of.
(3) influence of tertiary air quantity
Tertiary air is the wind that grate-cooler returns to dore furnace, and oxygen is provided for coal dust firing, influences combustion efficiency.When three times
When wind is smaller, raw material heat exchange is insufficient, influences the burning of coal dust and the decomposition of material；When tertiary air is larger, then it can cause kiln
Interior improper ventilation, or even the material that collapsed at kiln tail necking can be caused, and because of the increase of decomposition furnace outlet exhausted air quantity, can cause heat consumption with
The increase of power consumption.Influence of the tertiary air quantity to coal dust firing effect is more direct, it can be seen that, for the change of calciner temperature
Change, tertiary air quantity be also an important influence because.
The factor of influence calciner temperature also has a lot, but most of be not principal element and can not adjust in real time.Therefore,
This method chooses hello coal amount, feeding capacity and tertiary air and presses these three variables as the input of calciner temperature forecast model.
Step 2: gathered data；
Refering to Fig. 1-Fig. 3, first, the data for feeding coal amount, feeding capacity and tertiary air pressure are obtained in the DCS system of cement plant.
In the present embodiment, took one group to feed coal amount and feeding capacity data every 1 hour, 100 groups are taken altogether.Taken every 6 minutes one time three times
Wind pressure data, takes 10 groups of data in 1 hour, average as one group of data, totally 100 groups.Refering to Fig. 4~Fig. 7, then,
Calciner temperature is measured using infrared thermometer：Choose to measure section at fuel-pyrolysis and combustion zone, both sides tertiary air enters
It is measuring point at mouthful.Infrared sensor is contactless temperature-measuring, and measurement range is -50 DEG C~3000 DEG C, and measurement accuracy is up to 0.02
~0.1 DEG C, therefore sensor selects infrared sensor.The temperature that two side sensers are measured is averaged.Choose coal amount of feeding, feed
At the time of doses and tertiary air pressure data, the moment is chosen as calciner temperature, one group of temperature data was recorded every 6 minutes, 1 is small
When it is interior take 10 groups of data, average as one group of data, totally 100 groups.
Step 3: user's group hunting algorithm carries out parameter optimization to extreme learning machine
First, the input number of nodes for determining extreme learning machine is 3, and output node number is 1, and the number of hidden nodes is 9.And use
Crowd's searching algorithm carries out optimizing to the weights of extreme learning machine, and selects best weight value；The regression equation of extreme learning machine is：
Wherein：Y is output, xjFor input, K is the number of hidden nodes, ωijFor input weight matrix；biBiased for hidden layer
Matrix；βiFor output weight matrix；gi(ωij·xj+bi) it is activation primitive；
Extreme learning machine Optimization Steps based on crowd's algorithm are as follows：
Step1：The number of calculating limit learning machine weights sets the space dimension of crowd's searching algorithm as space dimensionality
Number, and initial search person is generated at random；
Step2：Initial search person is trained for extreme learning machine, and it is square with prediction data and True Data deviation
Difference is used as fitness value；
Step3：Start iteration, calculate the fitness value of each searcher, be compared, select fitness in comparing every time
It is worth less searcher as optimal searcher zbest；
Step4：When reaching maximum iteration, stop iteration, and regard final zbest as best initial weights；Otherwise,
Return to Step3.
In the present embodiment, the parameter of crowd's searching algorithm is set to：Population scale：30, maximum iteration：50, the limit
Learning machine inputs number：3, the number of hidden nodes：9, export number：1, space dimensionality：36, the excursions of Optimal Parameters for [-
10,10].Obtaining best weight value is：
ω11=[- 8.5998 1.0498-10.0000]
ω12=[2.5288 1.1687-1.4402]
ω13=[4.0307-5.4728-1.2693]
ω21=[4.2394-5.4643 10.0000]
ω22=[0.1258 2.1132-4.3216]
ω23=[- 2.5763 0.5363-0.6820]
ω31=[2.0390 6.8172-0.8140]
ω32=[- 0.7877 4.5656-0.5437]
ω33=[5.0381-3.6668 0.2628]
β1=[8.1040-9.0870-2.3459]
β2=[3.1803 1.2765-0.6899]
β3=[- 5.1972-0.6229 3.6674]
Step 4: setting up forecast model；
Parameter to extreme learning machine is configured；According to step one, input number of nodes：3, output node number：1, according to
Experience sets the number of hidden nodes：9, the best weight value that optimizing in step 3 is obtained is assigned to extreme learning machine；
Refering to Fig. 8, coal amount, feeding capacity and tertiary air pressure will be fed as input, calciner temperature is used as output, limit of utilization
Learning machine sets up model, will optimize obtained best weight value zbest in step 2 and is assigned to extreme learning machine and is trained.Use 80 groups
Data are trained, prediction effect such as Fig. 7 of training.Wherein, the predicted temperature of model is：
Dore furnace true temperature is：
0.2624 DEG C of the mean absolute error of calciner temperature prediction, mean square deviation is 0.1436, and precision is higher.
Step 5: predicting the outcome；
Coal amount, feeding capacity and tertiary air pressure data, the model prediction obtained using step 4 are fed according to what is currently collected
Calciner temperature；
Refering to Fig. 9, tested using remaining 20 groups of data, prediction effect such as Fig. 8 of test.Model prediction temperature is：
Dore furnace true temperature is：
0.2625 DEG C of the mean absolute error of calciner temperature prediction, mean square deviation is 0.1443, and precision is higher.
Step 6: real-time optimization forecast model；
According to predicting the outcome obtained by step 5, it is possible to use step 3 and step 4 are again to parameter optimization, again
Model is set up, precision is improved.When predicted temperature and detection temperature error are not reaching to desired value (desired value can sets itself),
Re-optimization parameter can be carried out with return to step two and step 3, set up model.Refering to Figure 10, i.e. this method implementation steps stream
Cheng Tu.
In summary, the present invention feeds coal amount, feeding capacity and tertiary air pressure using in cement plant DCS (Distributed Control System)
Real time data, and the calciner temperature measured by infrared thermometer, the extreme learning machine of user's group hunting optimization builds
A kind of forecast model of vertical cement decomposing furnace temperature, it is proposed that method of measurement calciner temperature.Simple to operate, economy of the invention
Practicality, without measuring other data, you can complete the on-line checking to cement decomposing furnace temperature, real-time is high, and there is provided one kind
Effective measuring method, guarantee is provided for control calciner temperature, optimization manufacture of cement.
Claims (2)
1. a kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature, it is characterised in that this method includes following step
Suddenly：
Step 1: determining the principal element of influence calciner temperature；
By dore furnace internal-combustion mechanism, with reference to the summary of experience of field personnel, it can be deduced that influence calciner temperature
Principal element have feed coal amount, feeding capacity and tertiary air pressure；
Step 2: gathered data；
The data for feeding coal amount and feeding capacity are obtained from the DCS system of cement plant, calciner temperature is surveyed using infrared thermometer；Hello
The data of coal amount, feeding capacity and calciner temperature take 100 groups；
Step 3: user's group hunting algorithm carries out parameter optimization to extreme learning machine；
User's group hunting algorithm carries out optimizing to the weights of extreme learning machine, and selects best weight value；Time of extreme learning machine
The equation is returned to be：
<mrow>
<mi>y</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>K</mi>
</munderover>
<msub>
<mi>&beta;</mi>
<mi>i</mi>
</msub>
<msub>
<mi>g</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>&omega;</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>&CenterDot;</mo>
<msub>
<mi>x</mi>
<mi>j</mi>
</msub>
<mo>+</mo>
<msub>
<mi>b</mi>
<mi>i</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mn>2</mn>
<mo>,</mo>
<mn>3</mn>
<mo>,</mo>
</mrow>
Wherein：Y is output, xjFor input, K is the number of hidden nodes, ωijFor input weight matrix；biFor hidden layer bias matrix；
βiFor output weight matrix；gi(ωij·xj+bi) it is activation primitive；
Step 4: setting up forecast model；
Parameter to extreme learning machine is configured；According to step one, input number of nodes：3, output node number：1, rule of thumb
The number of hidden nodes is set：9, the best weight value that optimizing in step 3 is obtained is assigned to extreme learning machine；
Step 5: predicting the outcome；
Coal amount, feeding capacity and tertiary air pressure data are fed according to what is currently collected, the model prediction obtained using step 4 is decomposed
Furnace temperature；
Step 6: real-time optimization forecast model；
According to predicting the outcome obtained by step 5, it is possible to use step 3 and step 4 are re-established again to parameter optimization
Model, improves precision.
2. the method that a kind of limit of utilization learning machine according to claim 1 predicts cement decomposing furnace temperature, its feature exists
In comprising the following steps that for, described step three：
Step1：Space dimensionality in calculating crowd's searching algorithm is weights number, produces initial search person；
Step2：The weights of initialization are assigned to extreme learning machine, are trained, and with prediction data and True Data difference
Mean square deviation is used as fitness value；
Step3：Iteration optimizing, by comparing, obtains the minimum weights of fitness value；
Step4：When reaching stop condition, stop iteration, export best initial weights；Otherwise, return to Step3 and continue optimizing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611223114.8A CN107066659A (en) | 2016-12-27 | 2016-12-27 | A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611223114.8A CN107066659A (en) | 2016-12-27 | 2016-12-27 | A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107066659A true CN107066659A (en) | 2017-08-18 |
Family
ID=59624044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611223114.8A Pending CN107066659A (en) | 2016-12-27 | 2016-12-27 | A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107066659A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113420461A (en) * | 2021-07-20 | 2021-09-21 | 山东恒拓科技发展有限公司 | Cement decomposing furnace online simulation system and establishing method thereof |
| CN113589693A (en) * | 2021-07-22 | 2021-11-02 | 燕山大学 | Cement industry decomposing furnace temperature model prediction control method based on neighborhood optimization |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104537033A (en) * | 2014-12-23 | 2015-04-22 | 清华大学 | Interval type index forecasting method based on Bayesian network and extreme learning machine |
| CN104656436A (en) * | 2014-10-27 | 2015-05-27 | 济南大学 | Decomposing furnace outlet temperature modeling method |
-
2016
- 2016-12-27 CN CN201611223114.8A patent/CN107066659A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104656436A (en) * | 2014-10-27 | 2015-05-27 | 济南大学 | Decomposing furnace outlet temperature modeling method |
| CN104537033A (en) * | 2014-12-23 | 2015-04-22 | 清华大学 | Interval type index forecasting method based on Bayesian network and extreme learning machine |
Non-Patent Citations (2)
| Title |
|---|
| 余胜威 等: "基于人群搜索算法的PID 控制器参数优化", 《计算机仿真》 * |
| 孙伟 等: "基于改进极根学习机的回转窑煅烧带温度预测方法"", 《计算机测量与控制》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113420461A (en) * | 2021-07-20 | 2021-09-21 | 山东恒拓科技发展有限公司 | Cement decomposing furnace online simulation system and establishing method thereof |
| CN113589693A (en) * | 2021-07-22 | 2021-11-02 | 燕山大学 | Cement industry decomposing furnace temperature model prediction control method based on neighborhood optimization |
| CN113589693B (en) * | 2021-07-22 | 2023-05-09 | 燕山大学 | Cement industrial decomposing furnace temperature model predictive control method based on neighborhood optimization |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109583585B (en) | Construction method of power station boiler wall temperature prediction neural network model | |
| CN109147878B (en) | Soft measurement method for free calcium of cement clinker | |
| CN107016176A (en) | A kind of hybrid intelligent overall boiler burning optimization method | |
| CN106906351B (en) | A kind of board briquette forecasting model and optimum furnace method | |
| CN105886680B (en) | A kind of blast furnace ironmaking process molten iron silicon content dynamic soft measuring system and method | |
| CN106327004A (en) | Cement firing process optimizing method based on clinker quality index | |
| CN111665809B (en) | Segmentation mechanism modeling method and system for rotary cement kiln | |
| CN106709197A (en) | Molten iron silicon content predicting method based on slide window T-S fuzzy neural network model | |
| BRPI1015989B1 (en) | online optimization of wet iron ore pellet hardening in a moving grid | |
| CN104498706B (en) | A kind of drying grate-rotary kiln-circular cooler Trinitarian pelletizing production optimization method | |
| CN105674326B (en) | A kind of industrial combustion gas boiler multiple target multiple constraint burning optimization method | |
| CN109342703B (en) | Method and system for measuring content of free calcium in cement clinker | |
| CN108760592B (en) | Fly ash carbon content online measurement method based on BP neural network | |
| CN103839110A (en) | Modeling method of prediction of production of nitrogen oxide in boiler | |
| CN106594794A (en) | Hybrid and intelligent updating method for boiler efficiency combustion optimization model | |
| CN103332878A (en) | Optimization method for production full process of novel dry-process cement clinker | |
| CN107066659A (en) | A kind of method that limit of utilization learning machine predicts cement decomposing furnace temperature | |
| CN105808945B (en) | A kind of hybrid intelligent boiler efficiency burning optimization method | |
| CN111174569A (en) | Method and system for online prediction of flue gas temperature of calcining section in rotary kiln | |
| CN105468799B (en) | Forecast the emulation mode of high-temp waste gas cycle sintering process heat state parameter | |
| CN103939941A (en) | Method for boiler combustion optimization with combination of irreversible thermodynamics | |
| Durdán et al. | Modeling of temperatures by using the algorithm of queue burning movement in the UCG Process. | |
| Alves et al. | Improving the sustainability of porcelain tile manufacture by flowsheet simulation | |
| CN105400951B (en) | A kind of humidification hybrid control system and its control method for annealing process of silicon steel | |
| CN103440390A (en) | Coupling simulation method for radiation section of industrial steam cracking furnace |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170818 |
|
| WD01 | Invention patent application deemed withdrawn after publication |