CN107726358B - Boiler Combustion Optimization System based on CFD numerical simulations and intelligent modeling and method - Google Patents
Boiler Combustion Optimization System based on CFD numerical simulations and intelligent modeling and method Download PDFInfo
- Publication number
- CN107726358B CN107726358B CN201710946941.8A CN201710946941A CN107726358B CN 107726358 B CN107726358 B CN 107726358B CN 201710946941 A CN201710946941 A CN 201710946941A CN 107726358 B CN107726358 B CN 107726358B
- Authority
- CN
- China
- Prior art keywords
- boiler
- air distribution
- cluster
- cfd
- sample
- 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.)
- Active
Links
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23N—REGULATING OR CONTROLLING COMBUSTION
- F23N5/00—Systems for controlling combustion
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23N—REGULATING OR CONTROLLING COMBUSTION
- F23N2223/00—Signal processing; Details thereof
- F23N2223/42—Function generator
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Regulation And Control Of Combustion (AREA)
Abstract
The invention discloses a kind of Boiler Combustion Optimization System and method based on CFD numerical simulations and intelligent modeling, system includes DCS control systems interface, CFD computing clusters, sample database cluster, intelligent modeling cluster, central processing cluster and man-machine interface, by the way that CFD analog samples and history run sample are stored, modeled and optimized, optimal air distribution mode real-time matching of the unit with the variations such as network load instruction, as-fired coal coal characteristic, excess air coefficient is realized.The use of CFD numerical simulation technologies improves the accuracy of modeling in the present invention, and when optimization can directly invoke DCS control systems and realize closed-loop control, make air distribution mode quick response load variations, realize the real-time optimization of unit combustion thermal efficiency and NOx emission.
Description
Technical field
The present invention relates to combustion of industrial boiler optimization systems, more particularly to the pot based on CFD numerical simulations and intelligent modeling
Stove combustion optimizing system and method.
Background technology
As country is increasingly stringent to fired power generating unit environmental requirement, how to the greatest extent may be used under the premise of disposal of pollutants is met the requirements
Boiler combustion efficiency can be improved, the problem that domestic fired power generating unit generally faces is had become.Air distribution is adjusted to be fired as fired power generating unit
Important regulative mode during burning is an important factor for influencing boiler thermal efficiency and pollutant generation.Therefore, how preferably
Adjusting air distribution mode becomes the most important thing for improving boiler thermal efficiency and reducing pollutant.In the case of existing, most of units are adopted
Air distribution is controlled with tandem oxygen content control air distribution, or by operation instruction principle and artificial experience, which results in boiler combustions
It is difficult to stablize in optimum, boiler efficiency is low, NOx discharge is high to causing, or even occurs burning partially, the severe feelings such as slagging
Condition.Therefore, there is an urgent need for a kind of methods that can adjust air distribution mode in real time according to load, coal quality, and boiler is made to meet pollutant row
Highest efficiency of combustion is obtained under the premise of putting requirement.
The intelligent modeling methods such as neural network are mainly used for the prioritization scheme of boiler wind speed adjustment mode at present, it is few with power plant
It is Sample Establishing boiler characteristics model to measure operation data and experimental data, according to model above using the methods of genetic algorithm to matching
Wind mode optimizes, such as method disclosed in patent ZL03231306.3 and patent CN106327021A.But above method
There are probelem in two aspects in carrying out boiler modeling process, on the one hand since power plant's operating condition has focused largely on high load capacity,
In shortage, the data sample of underload, while in order to ensure the safety of unit operation, often keeping single air distribution mode,
The data sample coverage area for being accordingly used in modeling is small and mutual independence is poor;On the other hand, mould in existing modeling process
Type input parameter is more, and over-fitting easily occurs.Two above problem seriously affected boiler characteristics model reliability and
Accuracy further results in optimization failure.How in modeling process abundant data sample reduce parameter input, to improve model
Accuracy and optimization reliability, be the main problem of boiler wind speed adjustment method optimizing.In addition existing manual adjustment air distribution side
Formula have certain time lag, and network load instruction be it is continually changing, this requires one kind can to load variations into
The air distribution regulating system of row quick response.
Invention content
Goal of the invention:The purpose of the present invention is to propose to a kind of boiler combustion based on CFD numerical simulations and intelligent modeling is excellent
Change system and method, the uses of wherein CFD numerical simulation technologies improves the accuracy of modeling, and when optimization can directly invoke DCS
Control system realizes closed-loop control, makes air distribution mode quick response load variations, realizes unit combustion thermal efficiency and NOx emission
Real-time optimization.
Technical solution:Boiler Combustion Optimization System the present invention is based on CFD numerical simulations and intelligent modeling includes DCS controls
System interface, CFD computing clusters, sample database cluster, intelligent modeling cluster, central processing cluster and man-machine interface, DCS controls
History run sample and CFD analog samples are transferred to CFD computing clusters by information transmission network by system interface processed respectively
It is preserved in sample database cluster;Intelligent modeling cluster calls the data stored in sample database cluster, selects intelligent algorithm
It is modeled, and boiler thermal efficiency prediction model and NOx discharge prediction model is transferred to central processing cluster;Central processing
Cluster calls the parameter of boiler real time execution in DCS control system interfaces to carry out seeking for air distribution mode while receiving model
It is excellent, so that unit combustion thermal efficiency and NOx emission is reached combination optimal value.
Using the method for CFD numerical simulations and the Boiler Combustion Optimization System of intelligent modeling, include the following steps:
(1) CFD analog samples are calculated by CFD computing clusters;
(2) it analyzes and improves history run sample;
(3) power plant's actual history operation sample by CFD analog samples and after improving is stored in sample database cluster,
So that intelligent modeling cluster calls;
(4) intelligent modeling cluster receives the sample in sample database cluster, is trained using intelligent algorithm;
(5) optimization of air distribution mode is carried out in central processing cluster;
(6) man-machine interface is established to ensure monitoring and control of the operations staff to entire burning optimization blowing system.
In step (1), feelings are arranged according to the practical structures of optimization aim boiler, each heating surface deployment scenarios, air duct flue
Condition and the parameters such as burner nozzle type and position build boiler physical model, using Computational Fluid Dynamics couplingization
The method for learning reaction simulates stove combustion process;When simulation, for different load, as-fired coal coal characteristic, excessive sky
The factors such as gas coefficient, coal pulverizer combination and air distribution mode are simulated, and corresponding unburned carbon in flue dust, NOx discharge are obtained
Etc. parameters, and boiler thermal efficiency is obtained by thermodynamic computing.
In step (2), by boiler combustion optimization Adjustment Tests, the unburned carbon in flue dust of unit under different operating conditions is obtained
And boiler slag carbon content;It is input, flying dust with load, as-fired coal coal characteristic, oxygen amount, coal pulverizer combination, air distribution mode etc.
Phosphorus content and boiler slag carbon content are output, establish nonlinear model;With power plant's actual operating data be input, unburned carbon in flue dust and
Boiler slag carbon content is the prediction unburned carbon in flue dust and boiler slag carbon content under output acquisition practical operation situation, anti-by boiler heat
Balance method calculates, and obtains the boiler thermal efficiency of history run sample.
In step (6), unit operation personnel read sample by human-computer interface control sample database cluster, manually or certainly
Dynamic update boiler thermal efficiency prediction model and NOx discharge prediction model;By selection, manually or automatically Optimizing Mode, control are matched
The optimal way of wind.
In step (3), the data in sample database include load, as-fired coal coal characteristic, excess air coefficient, air distribution
The information such as mode, boiler thermal efficiency, NOx discharge.
It is defeated to include the parameters such as load, as-fired coal coal characteristic, excess air coefficient, air distribution mode in step (4)
Enter, is output with boiler thermal efficiency, NOx discharge, establishes boiler thermal efficiency prediction model and NOx discharge prediction model, lead to
The case where addition regularization term prevents over-fitting is crossed to occur.
In step (5), central processing cluster reads real-time load, as-fired coal coal characteristic, mistake from DCS control system interfaces
The parameters such as air coefficient are measured, are inputted as fixed;The boiler thermal efficiency prediction model trained using intelligent modeling cluster
With NOx discharge prediction model, optimizing is carried out to air distribution mode using optimization algorithm, obtains corresponding under operating condition this moment most
Excellent air distribution mode.
Description of the drawings
Fig. 1 is the Boiler Combustion Optimization System figure based on CFD numerical simulations and intelligent modeling;
Fig. 2 is the Boiler combustion optimization general flow chart based on CFD numerical simulations and intelligent modeling;
Fig. 3 is the intelligent modeling algorithm flow chart by taking the BP neural network of regularization as an example;
Fig. 4 is the central processing cluster burning optimization algorithm flow chart by taking genetic algorithm as an example.
Specific implementation mode
Specific implementation mode is described further below in conjunction with the accompanying drawings.
As shown in Figure 1, the system includes DCS control systems interface, CFD computing clusters, sample database cluster, intelligently builds
Mould cluster, central processing cluster and man-machine interface.Wherein, DCS control systems interface and CFD computing clusters pass through information transmission network
History run sample and CFD analog samples are transferred in sample database cluster and are preserved respectively by network;Intelligent modeling cluster tune
With the data stored in sample database cluster, intelligent algorithm is selected to be modeled, and by boiler thermal efficiency prediction model and NOx
Forecasting of discharged quantity model is transferred to central processing cluster;Central processing cluster calls DCS control systems while receiving model
The parameter of boiler real time execution carries out the optimizing of air distribution mode in interface, and optimizing result can provide open loop to unit operation personnel and refer to
It leads or directly invokes DCS control systems and realize closed-loop control, unit combustion thermal efficiency and NOx emission is made to reach combination optimal value.
As shown in Fig. 2, the method includes the steps of:
(1) CFD analog samples are calculated by CFD computing clusters, according to the practical structures of optimization aim boiler, each heating surface
Deployment scenarios, air duct flue deployment scenarios and the parameters such as burner nozzle type and position, build the boiler object of complete and accurate
Model is managed, wherein secondary air duct controls each layer secondary air flow since air preheater entrance, using baffle opening.
The structure of mathematical model is as follows:The discrete of the differential equation has used Finite Volume Method for Air, pressure x velocity coupling to use three
Stability maintenance state SIMPLE algorithms calculate, and Equations of Turbulence is using the bis- equation simulation turbulent air flow flowings of standard k- ε, using random orbit mould
Type is to pulverized coal particle movement locus into line trace.The chemistry occurred in combustion using non-premixed combustion modeling coal dust
Combustion reaction and each component transport, and gas phase turbulance burning simulates (PDF) model using Hybrid analysis-probability density function;
Radiation heat transfer uses P1 radiation patterns.
After mathematical model is established, it is directed to different load, as-fired coal coal characteristic, excess air coefficient, coal-grinding respectively
Machine combination, air distribution mode are simulated, and obtain the parameters such as corresponding unburned carbon in flue dust, NOx discharge, and pass through heating power meter
Calculation obtains the boiler combustion thermal efficiency.
(2) it analyzes and improves history run sample.By boiler combustion optimization Adjustment Tests, obtain under different operating conditions
The unburned carbon in flue dust and boiler slag carbon content of unit.As shown in figure 3, using BP neural network algorithm, it is special with load, as-fired coal coal quality
Property, oxygen amount, coal pulverizer combination, air distribution mode be input, unburned carbon in flue dust and boiler slag carbon content are output, establish 3 layers of BP
Neural network model.It is input with power plant's actual operating data, unburned carbon in flue dust and boiler slag carbon content are that output obtains practical fortune
Prediction unburned carbon in flue dust in the case of row and boiler slag carbon content.It is calculated by boiler heat back balance method, obtains history run
The boiler thermal efficiency of data.
(3) power plant's actual history operation sample by CFD analog samples and after improving is stored in sample database cluster,
So that intelligent modeling cluster calls.Data in sample database include load, as-fired coal coal characteristic, excess air coefficient,
The information such as air distribution mode, boiler thermal efficiency, NOx discharge.The increase of random groups operating condition can be controlled by Power Plant DCS
The mode that system interface calls in history run sample updates sample database cluster.In addition, in some abnormal runnings,
When running on the lower load or lower heat of combustion coal, corresponding combustion conditions are simulated by CFD and are provided for sample database
Sample realizes the real-time update of sample database, ensures all standing of the sample to operating condition.
(4) intelligent modeling cluster receives the sample in sample database cluster, is trained using intelligent algorithm.Together
Sample uses BP neural network, by the sample including CFD analog samples and history run sample in sample database cluster
It imports in intelligent modeling cluster, is input with load, as-fired coal coal characteristic, excess air coefficient, air distribution mode, with boiler hot
Efficiency, NOx discharge are output, establish the BP neural network model of 3 layers of regularization.Hidden layer and output layer use sigmoid
Function is as activation primitive, after initializing each layer weight, the output net of i-th of node of hidden layer of networkiFor:
The output o of i-th of node of hidden layeriFor:
Wherein, φ is the excitation function of hidden layer, and φ is logsig functions in this formula;M is input layer number, i.e.,
Load, excess air coefficient, as-fired coal coal characteristic, air distribution mode;wijFor j-th of node of input layer to i-th of section of hidden layer
Weights between point;θiIndicate the threshold value of hidden layer.
The output net of k-th of node of output layerkFor:
The output o of k-th of node of output layerkFor:
Wherein, ψ is the excitation function of output layer, and ψ is logsig functions in this formula, and q is output layer neuron number, at this
It is q=2, i.e. the boiler combustion thermal efficiency, NOx discharge in formula;wkiFor implicit i-th of node layer by layer to k-th of node of output layer
Between weights;akIndicate the threshold value of output layer.
The error cost function of neural network is:
Wherein,For j-th of sample, the actual value of k-th of neuron of output layer;λ is regularization parameter, and effect is anti-
Only over-fitting;L is the neural network number of plies, and sl is certain layer of neuron number, and from i=1, i.e. input layer starts to calculate.
The target of neural metwork training is exactly to make network export to reach minimum, use with the difference of practical y values minimum and J (w)
Gradient descent method solves, and output layer error is:
δ(3)=o(3)-y
Wherein 3 indicate the outermost layer of neural network, i.e. output layer.
Hidden layer error is:
δ(2)=(w(2))Tδ(3).*g'(o(2))
The variable gradient of each layer of weights is:
Δ(l)=Δ(l)+δ(l+1)(net(l))T
I.e.
Input layer is constantly updated to hidden layer, the connection weight of hidden layer to output layer by above formula, and Algorithm Error is made to receive
Hold back designated value ε.Boiler thermal efficiency prediction model and NOx discharge prediction model can be obtained at this time.
Two above model can manually or automatically update:Manual mode is by operations staff in man-machine interface
Manually select update;Automated manner can be arranged to be automatically updated after new samples quantity reaches setting value.
(5) optimization of air distribution mode is carried out in central processing cluster.Central processing cluster is read from DCS control system interfaces
The parameters such as real-time load, as-fired coal coal characteristic, excess air coefficient are taken, are inputted as fixed;Utilize intelligent modeling collection
The boiler thermal efficiency prediction model and NOx discharge prediction model that group trains carry out optimizing to air distribution mode, are corresponded to
Air distribution mode optimal under operating condition this moment.As shown in figure 4, using genetic algorithm, specific implementation process is:
1. given constrained parameters, i.e., the value range of each input parameter, including load, excess air coefficient, as-fired coal
Coal characteristic, air distribution mode, the boiler combustion thermal efficiency, NOx discharge.It is initialized, initial evolutionary generation t=0 is set, most
Macroevolution algebraically t=T, it is random to generate n individual as parent and filial generation.
2. evaluating the fitness of individual, the foundation of fitness function is the boiler thermal effect that BP neural network is calculated
Rate and NOx discharge, if NOx discharge meets group setup standard, which is:
Eva=exp (- 0.005*Eff)
Wherein Eff is the boiler thermal efficiency of BP neural network output.
If NOx discharge exceeds group setup standard, which is set as 0.01, during evolution by
It eliminates.
3. n individual composition parent of selection and female generation from all individuals, to parent and mother for each variable in gene
The random coefficient between one [- 0.25,1.25] is taken, then carries out arithmetic friendship between male parent and female parent according to the random coefficient
Fork obtains 2n new individuals.Random variation operation is carried out to individual.
4. carrying out Fitness analysis to 2n newly-generated individual, highest 2 individuals of fitness in 2n individual are selected
New population is added, others are eliminated, and so far complete a wheel and evolve.The condition terminated of evolving is the continuous n of evolutionary process for most
The object function of excellent individual is basically unchanged, or when maximum algebraically is more than certain value T, loop termination.
5. after loop termination, obtained last generation individual, as optimal adaptation degree individual.The individual is represented in given pot
Under the premise of stove load, excess air coefficient, as-fired coal coal characteristic, how air distribution mode is adjusted, boiler thermal effect can be reached
The combination optimal value of rate and NOx discharge.
Wherein, optimal air distribution mode can select optimal multiple-objection optimization value, can also select control NOx discharge not
The optimal value of optimal boiler thermal efficiency is obtained in the case of exceeded.
(6) man-machine interface is established to ensure monitoring and control of the operations staff to entire burning optimization blowing system.Pass through
Man-machine interface, unit operation personnel can control sample database cluster and read sample, update or automatically update boiler hot manually
EFFICIENCY PREDICTION model and NOx discharge prediction model.In addition, operations staff can also be by selecting optimization or Automatic Optimal manually
Pattern controls the optimal way of air distribution, the form that manual mode can be instructed by open loop, by the recommendation air distribution mode after optimization
It is shown in man-machine interface, is adjusted manually by operations staff;Automated manner can directly adjust air distribution side by DCS control systems
Formula realizes that the closed loop of optimization system is adjusted.
Claims (1)
1. a kind of Boiler combustion optimization based on CFD numerical simulations and intelligent modeling, it is characterised in that:Including following step
Suddenly:
(1) CFD analog samples are calculated by CFD computing clusters;
(2) it analyzes and improves history run sample;
(3) power plant's actual history operation sample by CFD analog samples and after improving is stored in sample database cluster, for
Intelligent modeling cluster calls;
(4) intelligent modeling cluster receives the sample in sample database cluster, is trained using BP neural network;
(5) optimization of air distribution mode is carried out in central processing cluster;
(6) man-machine interface is established to ensure monitoring and control of the operations staff to entire burning optimization blowing system;
In the step (1), feelings are arranged according to the practical structures of optimization aim boiler, each heating surface deployment scenarios, air duct flue
Condition and burner nozzle type and location parameter build boiler physical model, using Computational Fluid Dynamics coupling chemistry
Reaction method simulates stove combustion process;
When being simulated to combustion process using CFD, for different load, as-fired coal coal characteristic, excess air coefficient, coal-grinding
Machine combination and air distribution mode factor are simulated, and obtain corresponding unburned carbon in flue dust, NOx discharge parameter, and pass through heat
Boiler thermal efficiency is calculated in power;
In the step (2), by boiler combustion optimization Adjustment Tests, the unburned carbon in flue dust of unit under different operating conditions is obtained
And boiler slag carbon content;It is input with load, as-fired coal coal characteristic, oxygen amount, coal pulverizer combination, air distribution mode, flying dust contains
Carbon amounts and boiler slag carbon content are output, establish nonlinear model;It is input, unburned carbon in flue dust and stove with power plant's actual operating data
Slag phosphorus content is the prediction unburned carbon in flue dust and boiler slag carbon content under output acquisition practical operation situation, is put down by the way that boiler heat is counter
Weighing apparatus method calculates, and obtains the boiler thermal efficiency of history run sample;
In the step (6), unit operation personnel read sample by human-computer interface control sample database cluster, manually or certainly
Dynamic update boiler thermal efficiency prediction model and NOx discharge prediction model;By selection, manually or automatically Optimizing Mode, control are matched
The optimal way of wind;
In the step (3), the data in sample database include load, as-fired coal coal characteristic, excess air coefficient, air distribution
Mode, boiler thermal efficiency, NOx discharge information;
In the step (4), with include load, as-fired coal coal characteristic, excess air coefficient, air distribution mode parameter be input,
It is output with boiler thermal efficiency, NOx discharge, boiler thermal efficiency prediction model and NOx discharge prediction model is established, by adding
The case where adding regularization term to prevent over-fitting, occurs;
In the step (5), central processing cluster reads real-time load, as-fired coal coal characteristic, mistake from DCS control system interfaces
Air coefficient parameter is measured, is inputted as fixed;The boiler thermal efficiency prediction model that is trained using intelligent modeling cluster and
NOx discharge prediction model carries out optimizing to air distribution mode using optimization algorithm, obtains corresponding to optimal under operating condition this moment
Air distribution mode.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710946941.8A CN107726358B (en) | 2017-10-12 | 2017-10-12 | Boiler Combustion Optimization System based on CFD numerical simulations and intelligent modeling and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710946941.8A CN107726358B (en) | 2017-10-12 | 2017-10-12 | Boiler Combustion Optimization System based on CFD numerical simulations and intelligent modeling and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107726358A CN107726358A (en) | 2018-02-23 |
CN107726358B true CN107726358B (en) | 2018-11-09 |
Family
ID=61210433
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710946941.8A Active CN107726358B (en) | 2017-10-12 | 2017-10-12 | Boiler Combustion Optimization System based on CFD numerical simulations and intelligent modeling and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107726358B (en) |
Families Citing this family (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110389570A (en) * | 2018-04-19 | 2019-10-29 | 株洲中车时代电气股份有限公司 | A kind of locomotive traction system trouble-shooter and method |
CN108984842A (en) * | 2018-06-20 | 2018-12-11 | 内蒙古工业大学 | A kind of design method and design platform of aggregate drying coal burner |
CN108954375B (en) * | 2018-07-18 | 2020-06-19 | 厦门邑通软件科技有限公司 | Coal-saving control method for boiler |
CN109063359A (en) * | 2018-08-16 | 2018-12-21 | 燕山大学 | A kind of dynamic modelling method of Process of Circulating Fluidized Bed Boiler |
CN109506248B (en) * | 2018-12-04 | 2020-06-26 | 华北电力大学 | Case reasoning boiler combustion optimization method based on online optimization |
CN109858125B (en) * | 2019-01-22 | 2023-10-27 | 中国大唐集团科学技术研究院有限公司火力发电技术研究院 | Thermal power unit power supply coal consumption calculation method based on radial basis function neural network |
CN110673485B (en) * | 2019-10-21 | 2020-11-24 | 京东城市(南京)科技有限公司 | Model training method, device, electronic apparatus, and medium for combustion control |
CN111222685B (en) * | 2019-11-21 | 2022-07-08 | 江苏方天电力技术有限公司 | Boiler wide-load NOx emission concentration prediction method based on model migration |
CN110986084B (en) * | 2019-12-25 | 2022-01-25 | 华润电力技术研究院有限公司 | Air distribution control method and system of pulverized coal fired boiler and related equipment |
CN111695249B (en) * | 2020-05-29 | 2023-08-01 | 广东省特种设备检测研究院顺德检测院 | Prediction method for heat efficiency of gas boiler |
CN111596548B (en) * | 2020-05-29 | 2023-04-07 | 华润电力技术研究院有限公司 | Boiler secondary air regulating and controlling method, device, equipment and medium |
CN111881120B (en) * | 2020-06-16 | 2024-04-26 | 国能信控互联技术有限公司 | Intelligent operation optimization method for boiler |
CN111931436A (en) * | 2020-08-10 | 2020-11-13 | 哈尔滨锅炉厂有限责任公司 | Burner nozzle air quantity prediction method based on numerical simulation and neural network |
CN112036091A (en) * | 2020-09-14 | 2020-12-04 | 哈尔滨锅炉厂有限责任公司 | Method for predicting air quantity of nozzle of four-corner tangential burner by using regression algorithm |
CN112216354A (en) * | 2020-09-17 | 2021-01-12 | 江苏集萃工业过程模拟与优化研究所有限公司 | Intelligent dosing system and method based on CFD numerical simulation and machine learning |
CN112163380B (en) * | 2020-09-24 | 2024-06-18 | 哈尔滨锅炉厂有限责任公司 | Neural network hearth oxygen concentration prediction system and method based on numerical simulation |
CN113074459A (en) * | 2021-03-11 | 2021-07-06 | 国家能源集团新能源技术研究院有限公司 | Boiler control optimization method and system |
CN113177352A (en) * | 2021-04-07 | 2021-07-27 | 华南理工大学 | Boiler combustion optimization system and method based on numerical simulation and artificial intelligence |
CN113158586A (en) * | 2021-05-26 | 2021-07-23 | 华能新能源股份有限公司 | Wind power plant numerical simulation method and device combined with neural network algorithm and storage medium |
CN113609663B (en) * | 2021-07-29 | 2024-04-09 | 大唐南京发电厂 | Combustion optimization method based on physical model and historical data analysis |
CN113864814B (en) * | 2021-09-15 | 2024-04-26 | 华能国际电力股份有限公司上海石洞口第一电厂 | Variable screening-based boiler combustion optimization method, device and medium |
CN117910170B (en) * | 2024-03-19 | 2024-06-11 | 西安慧金科技有限公司 | Method and system for designing upper air duct of industrial furnace |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106327021A (en) * | 2016-08-31 | 2017-01-11 | 西安艾贝尔科技发展有限公司 | Boiler combustion optimization air distribution method based on online model prediction |
CN106503380B (en) * | 2016-10-28 | 2019-11-15 | 中国科学院自动化研究所 | Coking nitrogen oxides in effluent concentration prediction method and forecasting system |
-
2017
- 2017-10-12 CN CN201710946941.8A patent/CN107726358B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN107726358A (en) | 2018-02-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107726358B (en) | Boiler Combustion Optimization System based on CFD numerical simulations and intelligent modeling and method | |
CN111829003B (en) | Power plant combustion control system and control method | |
CN105864797B (en) | Real-time prediction system and method for boiler entering heat value of circulating fluidized bed household garbage incineration boiler | |
CN104181900B (en) | Layered dynamic regulation method for multiple energy media | |
CN102129259B (en) | Neural network proportion integration (PI)-based intelligent temperature control system and method for sand dust environment test wind tunnel | |
CN105020705A (en) | Method and system for optimizing and controlling combustion performance of circulating fluidized bed boiler in real time | |
CN104763999A (en) | Power plant pulverized coal boiler combustion performance online optimizing method and system | |
CN103064289A (en) | Multiple-target operation optimizing and coordinating control method and device of garbage power generator | |
CN102889598A (en) | Control method for assisting stable combustion of garbage by predicting garbage calorific value | |
CN109886471A (en) | Fired power generating unit load distribution method based on neural network and intelligent optimization algorithm | |
CN106527176A (en) | MFOA (modified fruit fly optimization algorithm)-SVM (support vector machine)-based boiler thermal efficiency and NOX modeling method | |
CN104778361B (en) | The method of modified EMD Elman neural network prediction molten iron silicon contents | |
WO2015035134A1 (en) | Combined cycle power generation optimization system | |
CN105240846A (en) | Method for controlling combustion process of circulating fluidized bed boiler on basis of multivariable generalized predictive control optimization | |
CN103870878B (en) | A kind of Combustion Characteristics in Utility Boiler neural network model | |
CN112016754A (en) | Power station boiler exhaust gas temperature advanced prediction system and method based on neural network | |
CN108760592A (en) | A kind of unburned carbon in flue dust On-line Measuring Method based on BP neural network | |
CN110097929A (en) | A kind of blast furnace molten iron silicon content on-line prediction method | |
CN104729888B (en) | Flue fly ash constant-speed sampling system based on BP neural network and control method | |
CN111931436A (en) | Burner nozzle air quantity prediction method based on numerical simulation and neural network | |
CN106610589A (en) | Online hardware closed-loop network source coordination linear active-disturbance-rejection control method | |
CN105700383B (en) | A kind of positive pressed baker optimal control method | |
CN114909706A (en) | Secondary network balance regulation and control method based on reinforcement learning algorithm and pressure difference control | |
CN103870877A (en) | System and method for intelligently controlling boiler combustion based on neural network | |
CN109858125A (en) | A kind of fired power generating unit net coal consumption rate calculation method based on radial base neural net |
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 |