CN110793059A - Intelligent combustion comprehensive optimization control method for boiler - Google Patents
Intelligent combustion comprehensive optimization control method for boiler Download PDFInfo
- Publication number
- CN110793059A CN110793059A CN201911105929.XA CN201911105929A CN110793059A CN 110793059 A CN110793059 A CN 110793059A CN 201911105929 A CN201911105929 A CN 201911105929A CN 110793059 A CN110793059 A CN 110793059A
- Authority
- CN
- China
- Prior art keywords
- combustion
- neural network
- optimization
- boiler
- 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.)
- Pending
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23N—REGULATING OR CONTROLLING COMBUSTION
- F23N1/00—Regulating fuel supply
- F23N1/02—Regulating fuel supply conjointly with air supply
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
- F23N—REGULATING OR CONTROLLING COMBUSTION
- F23N1/00—Regulating fuel supply
- F23N1/02—Regulating fuel supply conjointly with air supply
- F23N1/022—Regulating fuel supply conjointly with air supply using electronic means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Abstract
The invention discloses a comprehensive optimization control method for intelligent combustion of a boiler, which comprises the steps of firstly establishing a bidirectional data communication link between an optimization control system and a unit DCS; then, acquiring unit operation historical data, constructing a combustion optimization neural network training sample, and establishing a combustion optimization neural network model; and finally, optimizing the combustion parameters by adopting a multi-objective non-dominated rapid sequencing genetic algorithm (NSGA-II) according to a combustion optimization neural network model, and realizing closed-loop optimization control of boiler combustion by adopting different optimization control decision strategies according to different operating conditions. The neural network training sample acquisition method provided by the invention can effectively improve the quality and modeling efficiency of the combustion optimization model; the combustion optimization algorithm based on the genetic algorithm can be convenient for operators to effectively balance the boiler efficiency and the emission of NOx in flue gas, realize the energy-saving and environment-friendly operation of the boiler, reduce the jump of optimized combustion parameters and is beneficial to the practical application of engineering.
Description
Technical Field
The invention belongs to intelligent control of a thermal power generating unit, and particularly relates to an intelligent combustion comprehensive optimization control method for a boiler.
Background
Big data and artificial intelligence are national science and technology development's war book, and the construction of intelligent power plant is being actively being carried out domestically, and its aim is to adopt advanced control strategy and technique, realizes the meticulous control of the important parameter of process, guarantees unit safety, economy, environmental protection operation. The boiler combustion optimization is one of the key points of intelligent power plant construction, and has important significance for energy conservation and emission reduction of thermal power generating units.
A boiler combustion optimization method based on big data and artificial intelligence generally adopts a neural network and a genetic algorithm technology, and the key of the application is how to establish a high-quality combustion optimization neural network model, which is closely related to the quality of a neural network training sample. The improvement of the boiler efficiency and the reduction of the emission of the flue gas NOx are a pair of contradictions, and because the combustion parameters are more and the influence relationship is complex, how to provide an effective means is convenient for operators to coordinate the boiler efficiency and the flue gas NOx according to the operation condition of the SCR denitration system, and the problem to be solved is the optimization of the boiler combustion. In addition, the safety requirement on the operation of the thermal power generating unit is high, and the combustion optimization algorithm must prevent the optimized parameters from jumping, so that the thermal shock to the unit is avoided, which is a problem that must be considered in the application of boiler combustion optimization engineering. The existing boiler combustion optimization method based on the neural network and the genetic algorithm does not well solve the problems.
Disclosure of Invention
The purpose of the invention is as follows: in order to overcome the problem of insufficient optimization of boiler intelligent combustion control in the prior art, the invention aims to provide a comprehensive optimization control method for boiler intelligent combustion, which is convenient for operators to effectively balance boiler efficiency and smoke NOx emission, reduces the jump of optimized combustion parameters as far as possible, and finally achieves the purposes of improving boiler efficiency and reducing smoke NOx emission.
In order to achieve the purpose, the technical scheme of the invention is as follows:
an intelligent combustion comprehensive optimization control method for a boiler comprises the following steps:
(1) configuring an optimization control system, wherein the optimization control system comprises a workstation and a PLC (programmable logic controller), a data communication port of the PLC is connected with a unit DCS (distributed control system) data communication port, a communication interface of the workstation is connected with the PLC data communication port, and a bidirectional data communication link of the optimization control system and the unit DCS is established;
(2) obtaining unit operation historical data, constructing a combustion optimization neural network training sample omega, and establishing a combustion optimization neural network model M comprising a vector Vin=[Vc,Vo]Neural net as a combustion optimizationInput of a network model, in vectorsAs model output, constituting combustion samples pi together with the time of each data record;
(3) and (3) optimizing combustion parameters through a multi-objective genetic algorithm based on the combustion optimization neural network model M established in the step (2), so as to realize closed-loop optimization control of boiler combustion.
Further, data exchange is carried out between the PLC and the unit DCS in the step (1) in a Modbus mode, and data exchange is carried out between the workstation and the PLC in an OPC mode.
The step (2) comprises the following steps:
(21) acquiring unit operation data of the last year from a unit DCS historical database, wherein the unit operation data comprises working condition variables, combustion parameter variables and other variables and time corresponding to each data record; wherein the working condition variables comprise unit load, coal heat value as fired, water supply temperature and environment temperature; the combustion parameter variables comprise the load of each coal mill, the primary air flow of each coal mill, the oxygen content of the flue gas at the inlet of the air preheater, the opening of a secondary air door and the opening of a burnout air door; other variables include the NOx content of the flue gas at the denitration inlet and a variable for calculating the positive balance furnace efficiency of the boiler;
(22) 5 min sliding average resampling is carried out on the historical operation data obtained in the step (21), the positive balance efficiency η of the boiler corresponding to each data record is calculated, and working condition variables are used for forming working condition variable vectors VcForming a combustion parameter variable vector V from the combustion parameter variablesoBy vector Vin=[Vc,Vo]As input to a combustion optimization neural network model, in vectorsAs a model output, a combustion sample Π is formed along with the time of each data record, consisting of { data record time, V } [in,VoutData record composition of the structure;
(23) obtaining a neural network training sample omega from the combustion sample pi, and calculating the value V according to the neural network training sample omegainAs model input, VoutAnd (5) as model output, establishing a combustion optimization neural network model M by adopting an RBF neural network learning algorithm.
Further, the specific step of obtaining the neural network training sample Ω in step (23) is as follows:
(231) setting a neural network training sample omega to be null;
(232) randomly selecting a sample i from combustion samples pi, and recording the input of a corresponding combustion optimization neural network model as the input
(233) Searching all samples j meeting the following condition in combustion samples pi to form a sample set pi1:
WhereinFor the input of the combustion optimization neural network model corresponding to the jth sample, | | · | | is the euler distance of the vector, δ is a smaller positive real number;
(234) will be provided withCorresponding sample adding sample set II1At pi1Selecting N samples nearest to the current time to be added into a neural network training sample omega;
(235) deletion of a sample set Π included in a combustion sample Π1The sample of (1); after deletion, if the sample pi is empty, ending, otherwise, placing the sample pi set1If empty, return to step (232).
The step (3) comprises the following specific steps:
(31) acquiring and calculating the current time in real time through the data link established in the step (1)Moving average data V of last 5 minutes before the momentc、VoAnd Vout;
(32) Randomly generating N individuals to form an initial populationIs the p-th individual vector formed by combustion parameter variables, p is 1,2, …, N1(ii) a To be provided withComputing an output vector of the model from the model M as an input to the neural network modelIs obtained byCorresponding model output set
(33) With the maximized boiler efficiency η, the NOx content in the denitration inlet flue gas is minimized, even if the output of the neural network model is the maximum optimization target, the combustion parameters are optimized by adopting a multi-target non-dominated rapid sequencing genetic algorithm to obtain an optimized populationAnd its corresponding model output set
(34) The control decision is made to determine the optimum combustion parameter value as followsIf the actual operation operating point V of the boileroutOn the Pareto front curve obtained by multi-objective optimization, the Pareto front curve is superior to VoutAmong the combustion parameters corresponding to the points (a) and (b), the closest VoAs a combustion parameter ofOptimum combustion parameter valueOtherwise, selecting the combustion parameter corresponding to the corresponding point on the Pareto front curve as the optimal combustion parameter value according to the η and the optimization weight of NOx
(35) Vector transformation through the data link of step (1)The combustion parameter value in the step (1) is sent to a set DCS to realize combustion closed loop optimization control;
(36) and (5) turning to the step (31), and repeating the steps (31) to (35).
Further, step (34) is for an optimal combustion parameter valueThe determination and control decisions of (c) are as follows:
(342) if Ψ2If the set is empty, go to (346); otherwise, turning to the next step;
(344) Calculating phi2The following performance indicators for each vector:
min is a small operation;
(346) calculating Ψ1The following performance indicators for each vector:
whereinThe ith element representing the corresponding vector, i ═ 1,2, λ ∈ [0,1 ∈]An optimized weighting factor for NOx;
max is a get big operation;
(348) and (6) ending.
Has the advantages that: compared with the prior art, the invention has the remarkable effects that: 1. the quality of a neural network training sample can be improved, and the quality and the modeling efficiency of a combustion optimization model are improved; 2. a combustion optimization method adopting different optimization strategies for different operation conditions is provided based on a genetic algorithm, so that the boiler efficiency and the smoke NOx emission can be effectively balanced by operators, the optimization effect can be ensured, the jump of combustion parameters is reduced as much as possible, the thermal shock of a unit is avoided, and the practical application of engineering is facilitated.
Drawings
FIG. 1 is a block diagram of the steps of the intelligent combustion comprehensive optimization control method of the boiler of the present invention;
FIG. 2 is a graph of Pareto front curves from multi-objective optimization.
Detailed Description
In order to explain the technical solutions disclosed in the present invention in detail, the present invention is further explained with reference to the accompanying drawings and specific embodiments.
The engineering application object is a 600MW unit, the boiler is a direct-fired pulverizing system four-corner tangential firing mode, and the specific steps of the method are as follows by combining the figure 1:
step 1: the optimization control system is provided with a workstation and a PLC, a data communication port of the PLC is connected with a unit DCS data communication port, and data exchange is carried out in a Modbus mode; a communication interface of the workstation is connected with a PLC data communication interface, data exchange is carried out by adopting an OPC mode, and a bidirectional data communication link between an optimization control system and a unit DCS is established;
in practical engineering application, an IBM server and a Siemens PLC are adopted in a workstation, the PLC is connected with a set DCS through a TS180 gateway, OPC is adopted between the IBM server and the PLC for data exchange, and a Modbus mode is adopted between the PLC and the DCS for data exchange.
Step 2: acquiring unit operation historical data, constructing a combustion optimization neural network training sample omega, and establishing a combustion optimization neural network model M, wherein the specific method comprises the following steps:
(2.1) acquiring unit operation data of the last year from a unit DCS historical database, wherein the data comprises working condition variables, combustion parameter variables, other variables and time corresponding to each data record, and the working condition variables comprise unit load, a heat value of coal as fired, water supply temperature and environment temperature; the combustion parameter variables comprise the load of each coal mill, the primary air flow of each coal mill, the oxygen content of the flue gas at the inlet of the air preheater, the opening of a secondary air door and the opening of a burnout air door; the other variables comprise the nitrogen oxide content NOx of the flue gas at the denitration inlet and a variable for calculating the positive balance furnace efficiency of the boiler;
(2.2) performing 5-minute sliding average resampling on the historical operating data obtained in the step (2.1), calculating the positive balance efficiency η of the boiler corresponding to each data record, and forming a working condition variable vector V by using the working condition variablescForming a combustion parameter variable vector V from the combustion parameter variablesoBy vector Vin=[Vc,Vo]As input to a combustion optimization neural network model, in vectorsAs model output, constituting combustion samples pi together with the time of each data record;
(2.3) according to the combustion sample pi, constructing a neural network training sample omega by adopting the following method:
A1) setting a neural network training sample omega to be null;
A2) randomly selecting a sample i from combustion samples pi, and recording the input of a corresponding combustion optimization neural network model as the input
A3) Searching all samples j meeting the following condition in combustion samples pi to form a sample set pi1:
WhereinFor the input of the combustion optimization neural network model corresponding to the jth sample, | | · | | is the euler distance of the vector, δ is a smaller positive real number;
A4) will be provided withCorresponding sample adding sample set II1At pi1Selecting N samples nearest to the current time to be added into a neural network training sample omega;
A5) deletion of a sample set Π included in a combustion sample Π1The sample of (1); after deletion, if the sample pi is empty, ending, otherwise, placing the sample pi set1Empty, return to step a 2);
(2.4) training samples omega according to the neural network, with VinAs model input, VoutAs model output, establishing a combustion optimization neural network model M by adopting a BP neural network learning algorithm;
in the actual engineering implementation, approximately 432000 records of 10-month unit operation historical data are collected from a unit DCS database, 26800 recorded neural network training samples are obtained after the records are processed according to the method in the step 2, the data samples cover the load range from 30% to 100% of rated load, the range of coal quality covers the received basic low-grade heat value from 14MJ/Kg to 25MJ/Kg of coal type, and the neural network model established by the method covers all possible operation working conditions of the unit.
The selected working condition variables comprise the load of a unit, the heat value of coal as fired, the water supply temperature and the environment temperature, the selected combustion parameter variables comprise the load of 6 coal mills, 6 primary air flow, 1 flue gas oxygen amount, 1 total air volume, the differential pressure between 1 auxiliary air box and a hearth, the opening of 6 layers of secondary air doors, the opening of 1 layer of OFA air doors and the opening of 3 layers of SOFA air doors, 25 variables in total are selected, the working condition variables and the combustion parameter variables are used as the input of a neural network model, the boiler efficiency η and the reciprocal 1/NOx of the nitrogen oxide content NOx of the flue gas at a denitration inlet are used as the output of the model, and an RBF neural network is adopted to establish a combustion optimization neural network model.
And step 3: based on the combustion optimization neural network model M established in the step 2, the combustion parameters are optimized by adopting a multi-objective genetic algorithm, the closed-loop optimization control of boiler combustion is realized, and the purposes of improving the boiler efficiency and reducing the emission of NOx in flue gas are achieved, and the method specifically comprises the following steps:
(3.1) acquiring and calculating the moving average data V of the latest 5 minutes before the current time in real time through the data link in the step 1c、VoAnd Vout;
(3.2) randomly generating N individuals to form an initial population phi0=[Vo p],Vo pIs the p-th individual vector formed by combustion parameter variables, p is 1,2, …, N1(ii) a To be provided withComputing an output vector of the model from the model M as an input to the neural network modelIs obtained byCorresponding model output set
(3.3) optimizing combustion parameters by adopting a multi-objective non-dominated rapid sequencing genetic algorithm (NSGA-II) with the maximum boiler efficiency η and the minimum NOx content in the denitration inlet flue gas as optimization targets to obtain an optimized populationAnd its corresponding model output set
A1) From Ψ1Select all vectors satisfying the following condition
A2) if Ψ2Empty set, go to a 6); otherwise, turning to the next step;
A4) Calculating phi2The following performance indicators for each vector:
min is a small operation;
A6) calculating Ψ1The following performance indicators for each vector:
whereinThe ith element representing the corresponding vector, i ═ 1,2, λ ∈ [0,1 ∈]An optimized weighting factor for NOx;
max is a big operation;
A8) and (6) ending.
(3.5) vector transformation through the data link of step 1The combustion parameter value in the step (1) is sent to a set DCS to realize combustion closed loop optimization control;
(3.6) turning to the step (3.1), and repeating the step (3.1) -the step (3.5).
In step 3, VoutFor the actual operating point, Ψ1Is the Pareto front in fig. 2. When the actual operation condition point VoutWhen not on the Pareto front curve, Ψ2Is a non-empty set, namely a Pareto leading edge ab section in fig. 2, any working point is selected on the ab section, and the boiler efficiency is higher than the actual operating working pointMeanwhile, the NOx emission is lower than the actual operation working condition point, and in order to reduce the jump of the combustion parameter, the optimal control decision strategy of the step A5 is adopted, and the point closest to the current working condition in the ab section is used as the optimal solution. When the actual operation condition point VoutOn Pareto front curve, Ψ2For the empty set, the method adopts an optimization control decision strategy of step A7, realizes flexible control of boiler efficiency and NOx emission by optimizing the weight coefficient, and enables operators to conveniently and effectively realize balance of boiler efficiency and NOx emission by adjusting the optimization weight coefficient.
During engineering acceptance, a third party performs boiler performance comparison tests before and after optimization, combustion optimization under full load improves boiler efficiency by about 0.53%, NOx emission is reduced by about 10%, and the optimization target of technical expectation is achieved, which shows that the method is effective.
The method provided by the invention provides a method for acquiring the neural network training samples, so that the samples can cover various working conditions of unit operation as much as possible, a large number of redundant samples can be eliminated, and the quality and the modeling efficiency of a combustion optimization model can be improved; the invention also provides a new combustion optimization method based on a genetic algorithm, and for the current working condition which operates at a non-Pareto front edge, the invention provides that a region which is superior to the current working condition is found out on the Pareto front edge, and the point which is closest to the current working condition in the region is taken as an optimal solution, and the combustion working condition is controlled to the Pareto front edge, thereby not only ensuring the combustion optimization effect, but also reducing the jump of the optimized combustion parameter; for the current working condition running at the Pareto frontier, the optimization weight coefficient lambda of NOx is adopted to comprehensively balance the boiler efficiency and the smoke NOx emission, and operators can conveniently and effectively adjust the balance boiler efficiency and the smoke NOx emission by optimizing the weight coefficient lambda, so that the method is favorable for engineering application.
Claims (6)
1. The intelligent combustion comprehensive optimization control method for the boiler is characterized by comprising the following steps of: the method comprises the following steps:
(1) configuring an optimization control system, wherein the optimization control system comprises a workstation and a PLC (programmable logic controller), a data communication port of the PLC is connected with a unit DCS (distributed control system) data communication port, a communication interface of the workstation is connected with the PLC data communication port, and a bidirectional data communication link of the optimization control system and the unit DCS is established;
(2) obtaining unit operation historical data, constructing a combustion optimization neural network training sample omega, and establishing a combustion optimization neural network model M comprising a vector Vin=[Vc,Vo]As input to a combustion optimization neural network model, in vectorsAs model output, constituting combustion samples pi together with the time of each data record;
(3) and (3) optimizing combustion parameters through a multi-objective genetic algorithm based on the combustion optimization neural network model M established in the step (2), so as to realize closed-loop optimization control of boiler combustion.
2. The intelligent combustion comprehensive optimization control method for the boiler according to claim 1, characterized in that: and (2) performing data exchange between the PLC and the unit DCS in the step (1) in a Modbus mode, and performing data exchange between the workstation and the PLC in an OPC mode.
3. The intelligent combustion comprehensive optimization control method for the boiler according to claim 1, characterized in that: the step (2) comprises the following steps:
(21) acquiring unit operation data of the last year from a unit DCS historical database, wherein the unit operation data comprises working condition variables, combustion parameter variables and other variables and time corresponding to each data record; wherein the working condition variables comprise unit load, coal heat value as fired, water supply temperature and environment temperature; the combustion parameter variables comprise the load of each coal mill, the primary air flow of each coal mill, the oxygen content of the flue gas at the inlet of the air preheater, the opening of a secondary air door and the opening of a burnout air door; the other variables comprise the nitrogen oxide content NOx of the flue gas at the denitration inlet and a variable for calculating the positive balance furnace efficiency of the boiler;
(22) performing a5 minute slip on the historical operating data obtained in step (21)Average re-sampling, calculating the positive balance efficiency η of boiler corresponding to each data record, and forming working condition variable vector V with working condition variablescForming a combustion parameter variable vector V from the combustion parameter variablesoBy vector Vin=[Vc,Vo]As input to a combustion optimization neural network model, in vectorsAs a model output, a combustion sample Π is formed along with the time of each data record, consisting of { data record time, V } [in,VoutData record composition of the structure;
(23) obtaining a neural network training sample omega from the combustion sample pi, and calculating the value V according to the neural network training sample omegainAs model input, VoutAnd (5) as model output, establishing a combustion optimization neural network model M by adopting an RBF neural network learning algorithm.
4. The intelligent combustion comprehensive optimization control method for the boiler according to claim 3, characterized in that: the specific steps of obtaining the neural network training sample omega in the step (23) are as follows:
(231) setting a neural network training sample omega to be null;
(232) randomly selecting a sample i from combustion samples pi, and recording the input of a corresponding combustion optimization neural network model as the input
(233) Searching all samples j meeting the following condition in combustion samples pi to form a sample set pi1:
WhereinIs the jthInputting a combustion optimization neural network model corresponding to a sample, | | | · | | is the Euler distance of the vector, and delta is a smaller positive real number;
(234) will be provided withCorresponding sample adding sample set II1At pi1Selecting N samples nearest to the current time to be added into a neural network training sample omega;
(235) deletion of a sample set Π included in a combustion sample Π1The sample of (1); after deletion, if the sample pi is empty, ending, otherwise, placing the sample pi set1If empty, return to step (232).
5. The intelligent combustion comprehensive optimization control method for the boiler according to claim 1, characterized in that: the step (3) comprises the following specific steps:
(31) acquiring and calculating the moving average data V of the latest 5 minutes before the current time in real time through the data link established in the step (1)c、VoAnd Vout;
(32) Randomly generating N individuals to form an initial population Is the p-th individual vector formed by combustion parameter variables, p is 1,2, …, N1(ii) a To be provided withComputing an output vector of the model from the model M as an input to the neural network modelIs obtained byCorrespond toModel output set of
(33) With the maximized boiler efficiency η, the NOx content in the denitration inlet flue gas is minimized, even if the output of the neural network model is the maximum optimization target, the combustion parameters are optimized by adopting a multi-target non-dominated rapid sequencing genetic algorithm to obtain an optimized populationAnd its corresponding model output set
If the actual operation operating point V of the boileroutOn the Pareto front curve obtained by multi-objective optimization, the Pareto front curve is superior to VoutAmong the combustion parameters corresponding to the points (a) and (b), the closest VoAs an optimum combustion parameter valueOtherwise, selecting the combustion parameter corresponding to the corresponding point on the Pareto front curve as the optimal combustion parameter value according to the η and the optimization weight of NOx
(35) Vector transformation through the data link of step (1)The combustion parameter value in the step (1) is sent to a set DCS to realize combustion closed loop optimization control;
(36) and (5) turning to the step (31), and repeating the steps (31) to (35).
6. The intelligent combustion comprehensive optimization control method for the boiler according to claim 5, characterized in that: step (34) for optimal combustion parameter valuesThe determination and control decisions of (c) are as follows:
(342) if Ψ2If the set is empty, go to (346); otherwise, turning to the next step;
(344) Calculating phi2The following performance indicators for each vector:
min is a small operation;
(346) calculating Ψ1The following performance indicators for each vector:
whereinThe ith element representing the corresponding vector, i ═ 1,2, λ ∈ [0,1 ∈]An optimized weighting factor for NOx;
max is a get big operation;
(348) and (6) ending.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911105929.XA CN110793059A (en) | 2019-11-13 | 2019-11-13 | Intelligent combustion comprehensive optimization control method for boiler |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911105929.XA CN110793059A (en) | 2019-11-13 | 2019-11-13 | Intelligent combustion comprehensive optimization control method for boiler |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110793059A true CN110793059A (en) | 2020-02-14 |
Family
ID=69444310
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911105929.XA Pending CN110793059A (en) | 2019-11-13 | 2019-11-13 | Intelligent combustion comprehensive optimization control method for boiler |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110793059A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111506031A (en) * | 2020-04-21 | 2020-08-07 | 大唐环境产业集团股份有限公司 | Denitration ammonia-spraying optimization auxiliary debugging system based on WinCC and P L C |
CN111765445A (en) * | 2020-07-01 | 2020-10-13 | 河北工业大学 | Boiler on-line combustion optimization control method and system and computer equipment |
CN113834091A (en) * | 2021-10-12 | 2021-12-24 | 中国矿业大学 | Control method for combustion optimization air supply system of gas-fired boiler |
CN113834092A (en) * | 2021-10-12 | 2021-12-24 | 中国矿业大学 | Combustion optimization control method for gas-fired boiler |
CN113864814A (en) * | 2021-09-15 | 2021-12-31 | 华能国际电力股份有限公司上海石洞口第一电厂 | Boiler combustion optimization method, device and medium based on variable screening |
CN114779722A (en) * | 2022-03-14 | 2022-07-22 | 上海发电设备成套设计研究院有限责任公司 | Intelligent combustion optimization control system and method for coal-fired power plant boiler |
CN115903473A (en) * | 2022-10-20 | 2023-04-04 | 中国大唐集团科学技术研究总院有限公司华东电力试验研究院 | Anti-blocking control method and system for rotary air pre-heater of thermal power plant |
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 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120053720A1 (en) * | 2009-02-13 | 2012-03-01 | Abb Research Ltd | Method and a system for optimization of parameters for a recovery boiler |
CN102880795A (en) * | 2012-09-18 | 2013-01-16 | 江西省电力科学研究院 | Method for acquiring target values of boiler optimized operation economic parameters |
CN103324862A (en) * | 2013-07-11 | 2013-09-25 | 中国石油大学(华东) | Coal-fired boiler optimization method based on improved neural network and genetic algorithm |
CN105717882A (en) * | 2014-12-05 | 2016-06-29 | 西安申科电子研究所 | Large-scale thermal power generating unit EDPF-NT distributed control system |
CN108644805A (en) * | 2018-05-08 | 2018-10-12 | 南京归图科技发展有限公司 | Boiler intelligent combustion optimal control method based on big data |
-
2019
- 2019-11-13 CN CN201911105929.XA patent/CN110793059A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120053720A1 (en) * | 2009-02-13 | 2012-03-01 | Abb Research Ltd | Method and a system for optimization of parameters for a recovery boiler |
CN102880795A (en) * | 2012-09-18 | 2013-01-16 | 江西省电力科学研究院 | Method for acquiring target values of boiler optimized operation economic parameters |
CN103324862A (en) * | 2013-07-11 | 2013-09-25 | 中国石油大学(华东) | Coal-fired boiler optimization method based on improved neural network and genetic algorithm |
CN105717882A (en) * | 2014-12-05 | 2016-06-29 | 西安申科电子研究所 | Large-scale thermal power generating unit EDPF-NT distributed control system |
CN108644805A (en) * | 2018-05-08 | 2018-10-12 | 南京归图科技发展有限公司 | Boiler intelligent combustion optimal control method based on big data |
Non-Patent Citations (5)
Title |
---|
刘开磊等: "基于K-最近邻筛选的BMA集合预报模型研究", 《水利学报》 * |
孙伟: "基于改进模糊神经网络的电站锅炉燃烧系统建模及优化控制研究", 《工程科技Ⅱ辑》 * |
岳艳娜: "模板匹配的最邻近改进算法研究", 《信息科技辑》 * |
张本贤等: "《热工控制与运行》", 31 May 2006 * |
李海山等: "基于智能计算的锅炉燃烧优化指导系统及应用", 《中国电力》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111506031A (en) * | 2020-04-21 | 2020-08-07 | 大唐环境产业集团股份有限公司 | Denitration ammonia-spraying optimization auxiliary debugging system based on WinCC and P L C |
CN111765445A (en) * | 2020-07-01 | 2020-10-13 | 河北工业大学 | Boiler on-line combustion optimization control method and system and computer equipment |
CN111765445B (en) * | 2020-07-01 | 2021-10-15 | 河北工业大学 | Boiler on-line combustion optimization control method and system and computer equipment |
CN113864814A (en) * | 2021-09-15 | 2021-12-31 | 华能国际电力股份有限公司上海石洞口第一电厂 | Boiler combustion optimization method, device and medium based on variable screening |
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 |
CN113834091A (en) * | 2021-10-12 | 2021-12-24 | 中国矿业大学 | Control method for combustion optimization air supply system of gas-fired boiler |
CN113834092A (en) * | 2021-10-12 | 2021-12-24 | 中国矿业大学 | Combustion optimization control method for gas-fired boiler |
CN113834092B (en) * | 2021-10-12 | 2023-02-14 | 中国矿业大学 | Combustion optimization control method for gas-fired boiler |
CN114779722A (en) * | 2022-03-14 | 2022-07-22 | 上海发电设备成套设计研究院有限责任公司 | Intelligent combustion optimization control system and method for coal-fired power plant boiler |
CN115903473A (en) * | 2022-10-20 | 2023-04-04 | 中国大唐集团科学技术研究总院有限公司华东电力试验研究院 | Anti-blocking control method and system for rotary air pre-heater of thermal power plant |
CN115903473B (en) * | 2022-10-20 | 2023-11-07 | 中国大唐集团科学技术研究总院有限公司华东电力试验研究院 | Anti-blocking control method and system for rotary air preheater of thermal power plant |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110793059A (en) | Intelligent combustion comprehensive optimization control method for boiler | |
CN103576655B (en) | A kind of power boiler burning subspace modeling and Multipurpose Optimal Method and system | |
WO2022048127A1 (en) | Optimization and regulation method and system for thermoelectric heat pump-thermoelectricity combined system | |
CN108644805A (en) | Boiler intelligent combustion optimal control method based on big data | |
CN107168062A (en) | A kind of load forecasting method in supercritical coal-fired units coordinated control system | |
CN106372756A (en) | Thermal power plant load optimization distribution method based on breeding particle swarm optimization | |
CN111562744B (en) | Boiler combustion implicit generalized predictive control method based on PSO algorithm | |
CN111416356B (en) | Transmission and distribution network linkage optimization method based on alternate direction multiplier method and optimal power flow | |
CN112200348B (en) | Regional comprehensive energy system multi-target operation decision method considering comprehensive demand response | |
CN111881554B (en) | Optimization control method for boiler changing along with air temperature | |
CN112016207A (en) | Economic load distribution optimization method for generator set | |
CN116167483A (en) | Park comprehensive energy system robust scheduling method considering stepped demand response | |
CN112000012A (en) | Unit sliding pressure control optimization method and system based on thermoelectric load condition | |
CN111027258A (en) | Intelligent prediction method for generating load and heating load of supercritical unit | |
CN113673863A (en) | Power system economic dispatching method based on quantum longicorn whisker algorithm | |
CN114330869A (en) | Multi-energy day-ahead scheduling optimization method for iron and steel enterprise considering multi-level indexes | |
CN108984979B (en) | Design method of ultra-supercritical unit depth peak regulation controller based on combination of multivariate frequency domain method and heuristic search algorithm | |
CN113836819A (en) | Bed temperature prediction method based on time sequence attention | |
CN113486509A (en) | Multi-objective optimization control method for comprehensive energy system | |
Zheng et al. | Double fuzzy pitch controller of wind turbine designed by genetic algorithm | |
CN111520740B (en) | Method for coordinately optimizing operation of multiple porous medium combustors | |
CN110794679A (en) | Prediction control method and system for load regulation of industrial steam supply system | |
CN114779722B (en) | Intelligent combustion optimization control system and method for coal-fired power station boiler | |
CN113298351B (en) | Cold-hot electric energy system planning method based on time domain analysis dynamic controllability index | |
CN111881120B (en) | Intelligent operation optimization method for boiler |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200214 |
|
RJ01 | Rejection of invention patent application after publication |