CN111765445A - Boiler on-line combustion optimization control method and system and computer equipment - Google Patents

Boiler on-line combustion optimization control method and system and computer equipment Download PDF

Info

Publication number
CN111765445A
CN111765445A CN202010618483.7A CN202010618483A CN111765445A CN 111765445 A CN111765445 A CN 111765445A CN 202010618483 A CN202010618483 A CN 202010618483A CN 111765445 A CN111765445 A CN 111765445A
Authority
CN
China
Prior art keywords
boiler
data
optimization
algorithm
combustion
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202010618483.7A
Other languages
Chinese (zh)
Other versions
CN111765445B (en
Inventor
梁涛
靳云杰
刘子豪
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hebei University of Technology
Original Assignee
Hebei University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hebei University of Technology filed Critical Hebei University of Technology
Priority to CN202010618483.7A priority Critical patent/CN111765445B/en
Publication of CN111765445A publication Critical patent/CN111765445A/en
Application granted granted Critical
Publication of CN111765445B publication Critical patent/CN111765445B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F22—STEAM GENERATION
    • F22B—METHODS OF STEAM GENERATION; STEAM BOILERS
    • F22B35/00—Control systems for steam boilers
    • F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F23—COMBUSTION APPARATUS; COMBUSTION PROCESSES
    • F23L—SUPPLYING AIR OR NON-COMBUSTIBLE LIQUIDS OR GASES TO COMBUSTION APPARATUS IN GENERAL ; VALVES OR DAMPERS SPECIALLY ADAPTED FOR CONTROLLING AIR SUPPLY OR DRAUGHT IN COMBUSTION APPARATUS; INDUCING DRAUGHT IN COMBUSTION APPARATUS; TOPS FOR CHIMNEYS OR VENTILATING SHAFTS; TERMINALS FOR FLUES
    • F23L9/00—Passages or apertures for delivering secondary air for completing combustion of fuel 
    • 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
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming

Abstract

The invention belongs to the technical field of optimized operation of coal-fired boilers of thermal power plants, and discloses a boiler on-line combustion optimization control method, a system and computer equipment, which are used for acquiring and preprocessing real-time characteristic parameter data of a boiler; substituting the real-time characteristic parameter data into a boiler combustion characteristic model based on an SFPA (Small form factor Power amplifier) optimization SVM (support vector machine); optimizing the boiler operation adjustable parameters by adopting an INSGA-II algorithm as an optimization algorithm; and selecting an optimal solution from the Pareto solution set, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to a DCS (distributed control system) to realize optimal control. According to the invention, the SFPA algorithm is used for optimizing SVM model parameters, so that the accuracy of the established boiler combustion characteristic model is improved; and optimizing by adopting an INSGA-II algorithm to provide decision support for different requirements in the actual operation process of the power station boiler.

Description

Boiler on-line combustion optimization control method and system and computer equipment
Technical Field
The invention belongs to the technical field of optimized operation of coal-fired boilers of thermal power plants, and particularly relates to a boiler online combustion optimization control method, a boiler online combustion optimization control system and computer equipment.
Background
At present, the power industry is the life line of national economy and the foundation of the development of modern society. The power generation forms in China include hydroelectric power generation, coal-fired power generation, nuclear power generation, wind power generation, solar power generation and the like, but at present, thermal power generation mainly based on coal still occupies a leading position in the power industry in China and is a main power source of an energy supply system in China. Thermal power generation using coal as a raw material provides electric power for life and industrial operation of people, and simultaneously brings two problems of large coal consumption and air pollution. Therefore, the thermal power plant needs to improve the operation efficiency to reduce the consumption of coal as much as possible and also needs to control the emission of atmospheric pollutants to protect the environment. Therefore, the problem that the boiler is in a working state with low nitrogen emission and high thermal efficiency needs to be solved urgently now is to develop a clean and efficient combustion optimization technology to control the combustion process of the boiler.
At present, combustion adjustment is mainly performed by operators according to own engineering experience in domestic power station combustion operation, but due to the complexity of boiler combustion, the method wastes time and labor and has a poor optimization effect. With the continuous development of machine learning and intelligent algorithms in recent years, a new direction is provided for boiler combustion optimization. An accurate boiler combustion system model is established, and on the basis of the model, adjustable parameters during boiler operation are optimized through an intelligent algorithm, the optimal values of the adjustable parameters are given, and clean and efficient operation of the boiler can be realized. However, the existing methods have the problems of insufficient model accuracy, poor optimization effect and the like.
In conclusion, an accurate and efficient boiler combustion system model and an optimization model need to be established urgently under the current energy and environment situation so as to overcome the problems of insufficient accuracy and poor actual optimization effect of the boiler combustion system model, and achieve the purposes of better guiding operators to adjust boiler production parameters, improving the boiler combustion efficiency and reducing the emission of nitrogen oxides.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a boiler on-line combustion optimization control method, a boiler on-line combustion optimization control system and computer equipment.
In order to achieve the purpose, the invention provides the following technical scheme:
an on-line combustion optimization control method for a boiler comprises the following steps:
step one, acquiring characteristic parameter historical data and NO of coal-fired boiler from DCS (distributed control System) of power plantxDischarge amount and heat efficiency data, and data pretreatment.
And step two, clustering and dividing the processed boiler operation data according to the distribution of the load working conditions by using a Gaussian Mixture Model (GMM) algorithm, and then resampling the data.
Thirdly, constructing a boiler combustion characteristic model by taking the resampled data as sample data; the boiler combustion characteristic model is obtained by optimizing the SVM based on an improved pollination algorithm SFPA.
And step four, acquiring real-time characteristic parameter data of the boiler from the power plant DCS, and preprocessing the data.
And step five, substituting the real-time characteristic parameter data as input into the trained boiler combustion characteristic model.
And sixthly, optimizing the boiler operation adjustable parameters by adopting an improved non-dominated sorting genetic algorithm INSGA-II as an optimization algorithm and combining an optimization target and a boiler combustion characteristic model.
And seventhly, selecting an optimal solution from the Pareto solution set by engineering personnel according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to a DCS (distributed control system) to realize optimal control.
Further, in the first step, the characteristic parameters collected from the power plant DCS system are boiler load, coal supply quantity, flue gas oxygen content, primary air quantity, primary air temperature, secondary air quantity, secondary air temperature and fly ash carbon content.
The characteristic parameter historical data collected from the power plant DCS system is data obtained after data in a boiler fault operation time period are removed.
The data preprocessing method comprises the following steps: and a normalization processing method is adopted to avoid the influence of the acquired data on the established model due to different physical meanings and units, so that the application of the data is facilitated.
Further, in the second step, the method for clustering and resampling the boiler operation parameter data according to the boiler load distribution by using the Gaussian Mixture Model (GMM) includes:
(1) setting a sample set as X and the number of samples as N; according to the value range (x) of the load parameter corresponding to the sample datamin，χmax) The interval is divided into equal parts, wherein the division standard is that each interval conforms to the range of 10MW, and the interval is divided into equal parts, namely K equal to round ((x) andmax-xmin) 10) (K is more than or equal to 2); and a Gaussian mixture model of the boiler load is provided with K components (K is K);
(2) initializing a Gaussian mixture model, wherein the formula is as follows:
Figure BDA0002564416600000031
in the formula, phikIs a mixing coefficient, which satisfies
Figure BDA0002564416600000032
Representing the kth component in the model;
setting the mixing coefficient of each component as 1/k, setting the mean value of each component as the central value of the corresponding coincidence interval, and setting the initial variance as 20;
(3) clustering sample data according to a Gaussian mixture model algorithm to obtain a clustered sample set C ═ C1,C2,C3,...,Ck-if the absolute value of the load mean difference between the two closest clusters of the convergence is less than 5, k-1 → k, and returning to perform step (2);
(4) and after a final clustering result is obtained, resampling the boiler data according to the clustering result in the boiler data. The sampling rule is as follows: the cluster after clustering corresponds to a mixing coefficient phikGreater than 1/K andthe method meets the condition that the data variance is less than 10, and the resampling is carried out when the two conditions are met, wherein the specific sampling formula is as follows:
Figure BDA0002564416600000033
where round () is a rounding function.
Further, in the third step, the boiler combustion characteristic model is established by adopting an improved flower pollination algorithm SFPA optimization SVM and is applied to boiler combustion optimization; the boiler load, coal feeding quantity, flue gas oxygen content, primary air quantity, primary air temperature, secondary air quantity, secondary air temperature and fly ash carbon content are used as input variables of the model, NOxAnd (4) taking the discharge amount and the boiler thermal efficiency as output variables, and training a boiler combustion characteristic model.
Further, in the third step, the simulated annealing algorithm and the flower pollination algorithm are combined to form a new SFPA optimization algorithm; then, the SFPA optimization algorithm is used for carrying out parameter optimization on the kernel function parameter sigma, the penalty coefficient C and the insensitive loss function of the SVM, and the method comprises the following steps,
(1) initializing an SFPA optimization algorithm, determining the value ranges of C, sigma and three parameters, setting the number of flower populations as N and the maximum iteration number as itermaxThe conversion probability is P, the initial temperature is T, and the annealing constant is theta;
(2) randomly generating N solutions, wherein each solution vector corresponds to a three-dimensional vector (C, sigma,), and finding the optimal solution and the minimum error f under the current conditionmin；
(3) Determining the error value of each Sol (i) at the current temperature according to
Figure BDA0002564416600000041
Wherein Sol (i) is a certain value of the current solution vector, best is the optimal value in the whole situation, and T is the initial temperature of annealing;
(4) finding a certain substitute value of the global optimum value from all Sol (i) according to the roulette strategy, and recording the substitute value as best _ plus;
(5) when P > rand, updating the solution vector by using the following formula, and performing the border crossing processing of the solution vector:
Figure BDA0002564416600000042
Figure BDA0002564416600000043
wherein λ is 1.5, M is a constant,
Figure BDA0002564416600000051
(λ) is the standard gamma function;
(6) when P < rand, updating the solution vector by using the following formula, and performing the border crossing processing of the solution vector:
Figure BDA0002564416600000052
wherein the random number is uniformly distributed between 0 and 1,
Figure BDA0002564416600000053
and
Figure BDA0002564416600000054
pollen of different flowers of the same species;
(7) error of the new solution formed in step (5) or (6)
Figure BDA0002564416600000055
Make a judgment if
Figure BDA0002564416600000056
Is less than fminIf not, the current solution is reserved;
(8) if the minimum error corresponding to the formed new solution is smaller than the global minimum error, updating the global optimum solution and the global minimum error;
(9) carrying out annealing operation;
(10) determine f thereofminWhether the prediction accuracy of the boiler combustion modeling is achieved, and if so, the process terminates and outputs a set of solutions (C, σ,) that are optimal at that time and a global minimum error fminOtherwise, jumping to the step (3) to continue searching.
Further, in the sixth step, the adjustable parameters of the boiler operation are coal feeding quantity, primary air quantity, secondary air quantity and flue gas oxygen content, the adjustable parameters are used as optimization variables to carry out optimization, and the rest non-adjustable parameters are used as fixed values to be kept unchanged in the optimization process; the multi-objective optimization objective is as follows:
Figure BDA0002564416600000057
in the formula (I), the compound is shown in the specification,
Figure BDA0002564416600000058
is NO of boilerxDischarge concentration, fη(x) Is the thermal efficiency of the boiler, xiFor the ith optimizable boiler operating variable, AiAnd BiThe value range of the parameters is optimized.
Further, in step six, the INSGA-II algorithm is improved by the NSGA-II algorithm, and the improvement method comprises:
(1) by enlarging the size of the first generation population, the evolution of the population is accelerated at the initial stage, and the convergence of the algorithm is improved; if the population number of individuals is N, the number of individuals of the initialization population may be set to be between 1.5N and 2N.
(2) Probability operation is introduced to the selection operator, so that the diversity of the population is expanded, and the distributivity of the solution set is improved; the formula for the calculation of the probability operation is:
Figure BDA0002564416600000061
in the formula, ω is a probability selection operator parameter, and g is an evolution algebra.
(3) Introducing a mixed crossover operator, and dynamically adjusting the search space of the algorithm; the mixed crossover operator introduces a Gaussian distribution crossover operator NDX on the basis of an SBX crossover operator, and adaptively adjusts the weights of the two crossover operators. Let u be the random number generated by uniform distribution over the interval (0,1), and r ═ N (0,1) | be the value of the gaussian distributed random variable, when u is less than or equal to 0.5, the update formula of the individual under the hybrid crossover operator is:
Figure BDA0002564416600000062
when u > 0.5, the update formula of the individual under the mixed crossover operator is as follows:
Figure BDA0002564416600000063
in the formula, x1,iAnd x2,iThe value of the ith variable of the individual child generated by the hybrid crossover operator; m ═ p1,i+p2,i,N＝p1,i-p2,i，p1,iAnd p2,iIs the value of the ith variable of the parent individual, G is the current iteration number, G is the total iteration number, ηcIs a hybrid crossover operator parameter.
Further, in the sixth step, the INSGA-II optimization algorithm specifically includes:
(1) initializing INSGA-II algorithm: setting the population size to be N, initializing a population P0Scale N of01.5N to 2N, maximum number of iterations G, probability selection operator parameter ω, hybrid crossover operator parameter ηc；
(2) Randomly generating N0Individuals as parent population P0；
(3) The current iteration number g is 0, and the parent population P is treated0Performing probability selection, mixed crossing and mutation operations to generate a filial generation population Q0；
(4) The parent population PtAnd the offspring population QtMerging to generate new population RtAnd for new population RtPerforming fast non-dominated sorting;
(5) using elite strategy, innovationPopulation RtN excellent individuals are selected to generate a new parent population Pt+1；
(6) For new population Pt+1Performing probability selection, mixed crossing and mutation operations to generate a new filial generation population Qt+1；
(7) And (4) if the current iteration time G is not less than the maximum iteration time G, ending the algorithm, otherwise, G is G +1, and jumping to the step (4) for circulation.
Further, in the seventh step, adding an optimization bias logic at the DCS system side, and superimposing an optimization bias on the control instruction of the optimized variable; and the difference between the optimized coal feeding quantity, the optimized primary air quantity, the optimized secondary air quantity and the optimized flue gas oxygen content and the actual coal feeding quantity, the optimized primary air quantity, the optimized secondary air quantity and the optimized flue gas oxygen content is used as an optimization bias and sent to a DCS (distributed control System) to realize the optimized control of the on-line combustion of the boiler.
It is a further object of the invention to provide a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of:
acquiring characteristic parameter historical data and NO of coal-fired boiler from DCS (distributed control System) of power plantxDischarge amount and thermal efficiency data, and carrying out data pretreatment;
clustering and dividing the processed boiler operation data according to the distribution of load working conditions by using a Gaussian Mixture Model (GMM) algorithm, and then resampling the data;
taking the resampled data as sample data to construct a boiler combustion characteristic model; the boiler combustion characteristic model is obtained by optimizing an SVM (support vector machine) based on an improved pollination algorithm SFPA (small form-factor power amplifier);
acquiring real-time characteristic parameter data of a boiler from a power plant DCS (distributed control System), and preprocessing the data;
taking the real-time characteristic parameter data as input, and substituting the input into the trained boiler combustion characteristic model;
an improved non-dominated sorting genetic algorithm INSGA-II is adopted as an optimization algorithm, and an optimization target and a boiler combustion characteristic model are combined to optimize boiler operation adjustable parameters;
and (4) selecting an optimal solution from the Pareto solution set by engineering personnel according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to a DCS (distributed control system) to realize optimal control.
Another object of the present invention is to provide an online combustion optimization control system for a boiler, which implements the online combustion optimization control method for a boiler, the online combustion optimization control system for a boiler comprising:
the first data preprocessing module is used for acquiring the characteristic parameter historical data and NO of the coal-fired boiler from the DCS of the power plantxDischarge amount and thermal efficiency data, and carrying out data pretreatment;
the data resampling module is used for clustering and dividing the processed boiler operation data according to the distribution of the load working conditions by using a Gaussian Mixture Model (GMM) algorithm and resampling the data;
the boiler combustion characteristic model building module is used for building a boiler combustion characteristic model by taking the resampled data as sample data;
the second data preprocessing module is used for acquiring real-time characteristic parameter data of the boiler from the power plant DCS and preprocessing the data;
the real-time data input module is used for taking the real-time characteristic parameter data as input and substituting the input into the trained boiler combustion characteristic model;
the boiler operation adjustable parameter processing module is used for optimizing boiler operation adjustable parameters by adopting an improved non-dominated sorting genetic algorithm INSGA-II as an optimization algorithm and combining an optimization target and a built boiler combustion characteristic model;
and the optimal offset calculation module is used for selecting an optimal solution from the Pareto solution set according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to the DCS to realize optimal control.
By combining all the technical schemes, the invention has the advantages and positive effects that:
according to the boiler online combustion optimization control method provided by the invention, the kernel function parameter sigma, the penalty coefficient C and the insensitive loss function of the SVM are subjected to parameter optimization by utilizing an SFPA optimization algorithm, so that the built boiler combustion characteristic model has higher precision and better prediction effect.
The invention uses an improved non-dominated sorting genetic algorithm INSGA-II to optimize the adjustable parameters of boiler operation, and compared with the method that only one Pareto solution can be searched when an optimization objective function is established by using a weight coefficient method, the INSGA-II algorithm can search a plurality of groups of Pareto solutions, thereby providing decision support for different requirements in the actual operation process of the power station boiler, improving the optimization performance and the optimization effect, and being capable of adapting to various optimization requirements in the actual engineering.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained from the drawings without creative efforts.
FIG. 1 is a flow chart of an on-line combustion optimization control method for a boiler according to an embodiment of the present invention.
FIG. 2 is a diagram of specific implementation steps of an online combustion optimization control method for a boiler according to an embodiment of the present invention.
Fig. 3 is a flowchart of a method for performing parameter optimization on a kernel function parameter σ, a penalty coefficient C, and an insensitive loss function of an SVM by using an SFPA optimization algorithm according to an embodiment of the present invention.
FIG. 4 is a flow chart of boiler combustion optimization provided by an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In order to solve the problems in the prior art, the invention provides an online combustion optimization control method, a system and a computer device for a boiler, and the invention is described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the method for optimizing and controlling boiler on-line combustion provided by the embodiment of the invention comprises the following steps:
s101, acquiring characteristic parameter historical data and NO of coal-fired boiler from DCS (distributed control System) of power plantxDischarge amount and heat efficiency data, and data pretreatment.
And S102, clustering and dividing the processed boiler operation data according to the distribution of the load working conditions by using a Gaussian Mixture Model (GMM) algorithm, and then resampling the data.
S103, constructing a boiler combustion characteristic model by taking the resampled data as sample data; the boiler combustion characteristic model is obtained by optimizing the SVM based on an improved pollination algorithm SFPA.
And S104, acquiring real-time characteristic parameter data of the boiler from the power plant DCS, and preprocessing the data.
And S105, substituting the real-time characteristic parameter data serving as input into the trained boiler combustion characteristic model.
And S106, optimizing the boiler operation adjustable parameters by adopting an improved non-dominated sorting genetic algorithm INSGA-II as an optimization algorithm and combining an optimization target and the established boiler combustion characteristic model.
And S107, selecting an optimal solution from the Pareto solution set by engineering personnel according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to a DCS (distributed control system) to realize optimal control.
The boiler on-line combustion optimization control method provided by the invention can be implemented by adopting other steps by persons skilled in the art, and the boiler on-line combustion optimization control method provided by the invention in fig. 1 is only one specific embodiment.
The technical solution of the present invention is further described with reference to the following examples.
As shown in fig. 2, the method for optimizing and controlling boiler on-line combustion provided by the embodiment of the present invention includes the following steps:
s1, obtaining the characteristic parameter historical data and NO of the coal-fired boiler from the DCS system of the power plantxDischarge amount and heat efficiency data, and data pretreatment.
Specifically, the characteristic parameters collected from the power plant DCS system are boiler load, coal supply quantity, flue gas oxygen content, primary air quantity, primary air temperature, secondary air quantity, secondary air temperature and fly ash carbon content.
Specifically, the historical data collected from the power plant DCS system should be data after data within a period of time of boiler fault operation is removed.
Specifically, the data preprocessing method in step S1 includes: and a normalization processing method is adopted to avoid the influence of the acquired data on the established model due to different physical meanings and units, so that the application of the data is facilitated.
And S2, clustering and dividing the processed boiler operation data according to the distribution of the load working conditions by using a Gaussian Mixture Model (GMM) algorithm, and then resampling the data.
Specifically, in step S2, a Gaussian Mixture Model (GMM) is used to cluster the boiler operation parameter data according to the boiler load distribution, and then resample the data. The method specifically comprises the following steps of,
s2-1, setting a sample set as X and the number of samples as N; according to the value range (x) of the load parameter corresponding to the sample datamin，χmax) The interval is divided into equal parts, wherein the division standard is that each interval conforms to the range of 10MW, and the interval is divided into equal parts, namely K equal to round ((x) andmax-xmin) 10) (K is more than or equal to 2); and a Gaussian mixture model of the boiler load is provided with K components (K is K);
s2-2, initializing a Gaussian mixture model, wherein the formula is as follows:
Figure BDA0002564416600000111
in the formula, phikIs a mixing coefficient, which satisfies
Figure BDA0002564416600000112
Representing the kth component in the model;
setting the mixing coefficient of each component as 1/k, setting the mean value of each component as the central value of the corresponding coincidence interval, and setting the initial variance as 20;
s2-3, clustering the sample data according to the Gaussian mixture model algorithm to obtain a clustered sample set C ═ C1,C2,C3,...,CkH, if the absolute value of the load mean difference between the two closest clusters to the cluster is less than 5, k-1 → k, and returns to perform step S2-2;
and S2-4, resampling the final clustering result according to the clustering result in the boiler data after the final clustering result is obtained. The sampling rule is as follows: the cluster after clustering corresponds to a mixing coefficient phikMore than 1/K, less than 10 of data variance, and resampling when the two conditions are met, wherein the specific sampling formula is as follows:
Figure BDA0002564416600000121
where round () is a rounding function.
S3, constructing a boiler combustion characteristic model by taking the resampled data as sample data; the boiler combustion characteristic model is obtained by optimizing the SVM based on an improved pollination algorithm SFPA.
Specifically, in the step S3, an improved pollination algorithm SFPA optimization SVM is adopted to establish a boiler combustion characteristic model, and the boiler combustion characteristic model is applied to boiler combustion optimization; the boiler load, coal feeding quantity, flue gas oxygen content, primary air quantity, primary air temperature, secondary air quantity, secondary air temperature and fly ash carbon content are used as input variables of the model, NOxAnd (4) taking the discharge amount and the boiler thermal efficiency as output variables, and training a boiler combustion characteristic model.
Specifically, firstly, combining a simulated annealing algorithm with a pollination algorithm to improve the global search capability and convergence rate of the pollination algorithm and form a new SFPA optimization algorithm; then, an SFPA optimization algorithm is used to perform parameter optimization on the kernel function parameter σ, the penalty coefficient C and the insensitive loss function of the SVM, as shown in fig. 3, the specific steps of the optimization include:
s3-1, initializing an SFPA optimization algorithm, determining the value ranges of C, sigma and three parameters, setting the number of flower populations as N and the maximum iteration number as itermaxThe conversion probability is P, the initial temperature is T, and the annealing constant is theta;
s3-2, randomly generating N solutions, wherein each solution vector corresponds to a three-dimensional vector (C, sigma,) and finding the optimal solution and the minimum error f under the current conditionmin；
S3-3, determining the error value of each Sol (i) at the current temperature according to the following formula
Figure BDA0002564416600000122
Wherein Sol (i) is a certain value of the current solution vector, best is the optimal value in the whole situation, and T is the initial temperature of annealing;
s3-4, finding a certain substitute value of the global optimum value from all Sol (i) according to the roulette strategy, and recording the substitute value as best _ plus;
s3-5, when P > rand, updating the solution vector by using the following formula, and performing the border crossing processing of the solution vector:
Figure BDA0002564416600000131
Figure BDA0002564416600000132
wherein λ is 1.5, M is a constant,
Figure BDA0002564416600000133
(λ) is the standard gamma function;
s3-6, when P < rand, updating the solution vector by using the following formula, and performing the border crossing processing of the solution vector:
Figure BDA0002564416600000134
wherein the random number is uniformly distributed between 0 and 1,
Figure BDA0002564416600000135
and
Figure BDA0002564416600000136
pollen of different flowers of the same species;
s3-7, error of new solution formed in step S3-5 or S3-6
Figure BDA0002564416600000137
Make a judgment if
Figure BDA0002564416600000138
Is less than fminIf not, the current solution is reserved;
s3-8, if the minimum error corresponding to the formed new solution is smaller than the global minimum error, updating the global optimum solution and the global minimum error;
s3-9, performing cooling operation;
s3-10, determining fminWhether the prediction accuracy of the boiler combustion modeling is achieved, and if so, the process terminates and outputs a set of solutions (C, σ,) that are optimal at that time and a global minimum error fminOtherwise, go to step S3-3 to continue searching.
And S4, acquiring real-time characteristic parameter data of the boiler from the power plant DCS, and preprocessing the data.
And S5, taking the real-time characteristic parameter data as input, and substituting the input into the trained boiler combustion characteristic model.
And S6, optimizing the boiler operation adjustable parameters by adopting an improved non-dominated sorting genetic algorithm INSGA-II as an optimization algorithm and combining an optimization target and a boiler combustion characteristic model.
As shown in fig. 4, in step S6, an improved non-dominated sorting genetic algorithm (INSGA-II) is used to optimize the adjustable parameters of the boiler operation, so as to obtain an optimal adjustable parameter combination of the boiler operation, and adjust the actual combustion process of the boiler according to the obtained optimal adjustable parameter combination, thereby achieving the purpose of optimizing the combustion.
Specifically, the adjustable parameters of the boiler operation are coal feeding quantity, primary air quantity, secondary air quantity and flue gas oxygen content, the adjustable parameters are used as optimization variables to carry out optimization, and the rest non-adjustable parameters are used as fixed values to be kept unchanged in the optimization process; the multi-objective optimization objective is as follows:
Figure BDA0002564416600000141
in the formula (I), the compound is shown in the specification,
Figure BDA0002564416600000142
is NO of boilerxDischarge concentration, fη(x) Is the thermal efficiency of the boiler, xiFor the ith optimizable boiler operating variable, AiAnd BiThe value range of the parameters is optimized.
Specifically, the INSGA-II algorithm is improved by the NSGA-II algorithm, and the improvement is divided into the following three aspects:
(1) by enlarging the size of the first generation population, the evolution of the population is accelerated at the initial stage, and the convergence of the algorithm is improved; if the population number of individuals is N, the number of individuals of the initialization population may be set to be between 1.5N and 2N.
(2) Probability operation is introduced to the selection operator, so that the diversity of the population is expanded, and the distributivity of the solution set is improved; the formula for the calculation of the probability operation is:
Figure BDA0002564416600000143
in the formula, ω is a probability selection operator parameter, and g is an evolution algebra.
(3) Introducing a mixed crossover operator, and dynamically adjusting the search space of the algorithm; the mixed crossover operator introduces a Gaussian distribution crossover operator NDX on the basis of an SBX crossover operator, and adaptively adjusts the weights of the two crossover operators. Let u be the random number generated by uniform distribution over the interval (0,1), and r ═ N (0,1) | be the value of the gaussian distributed random variable, when u is less than or equal to 0.5, the update formula of the individual under the hybrid crossover operator is:
Figure BDA0002564416600000151
when u > 0.5, the update formula of the individual under the mixed crossover operator is as follows:
Figure BDA0002564416600000152
in the formula, x1,iAnd x2,iThe value of the ith variable of the individual child generated by the hybrid crossover operator; m ═ p1,i+p2,i,N＝p1,i-p2,i，p1,iAnd p2,iIs the value of the ith variable of the parent individual, G is the current iteration number, G is the total iteration number, ηcIs a hybrid crossover operator parameter.
The INSGA-II algorithm optimization method specifically comprises the following steps:
s6-1, initializing the INSGA-II algorithm: setting the population size to be N, initializing a population P0Scale N of01.5N to 2N, maximum number of iterations G, probability selection operator parameter ω, hybrid crossover operator parameter ηc；
S6-2, randomly generating N0Individuals as parent population P0；
S6-3, setting the current iteration number g as 0, and counting the parent population P0Performing probability selection, mixed crossing and mutation operations to generate a filial generation population Q0；
S6-4, breeding the parent population PtAnd the offspring population QtMerging to generate new population RtAnd for new population RtPerforming fast non-dominated sorting;
s6-5, utilizing elite strategy to obtain new population RtSelect N good onesGenerating a new parent population Pt+1；
S6-6, for new population Pt+1Performing probability selection, mixed crossing and mutation operations to generate a new filial generation population Qt+1；
And S6-7, if the current iteration time G is not less than the maximum iteration time G, ending the algorithm, otherwise, G +1, and jumping to the step S6-4 to circulate.
And S7, selecting an optimal solution from the Pareto solution set by engineering personnel according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to a DCS (distributed control system) to realize optimal control.
Specifically, in step S7, adding an optimization bias logic on the DCS system side, and superimposing the optimization bias on the control command of the optimized variable; and (5) taking the differences of the coal feeding quantity, the primary air quantity, the secondary air quantity and the flue gas oxygen content obtained by optimization in the step (S6) and the actual coal feeding quantity, the primary air quantity, the secondary air quantity and the flue gas oxygen content as optimization offsets, and sending the optimization offsets to a DCS (distributed control system) to realize the online combustion optimization control of the boiler.
It should be noted that the embodiments of the present invention can be realized by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus and its modules of the present invention may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of hardware circuits and software, e.g., firmware.
The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.

Claims (9)

1. An online combustion optimization control method for a boiler is characterized by comprising the following steps:
step one, acquiring characteristic parameter historical data and NO of coal-fired boiler from DCS (distributed control System) of power plantxDischarge amount and thermal efficiency data, and carrying out data pretreatment;
secondly, clustering and dividing the processed boiler operation data according to the distribution of the load working conditions by using a Gaussian Mixture Model (GMM) algorithm, and then resampling the data;
thirdly, constructing a boiler combustion characteristic model by taking the resampled data as sample data; the boiler combustion characteristic model is obtained by optimizing an SVM (support vector machine) based on an improved pollination algorithm SFPA (small form-factor power amplifier);
acquiring real-time characteristic parameter data of the boiler from a power plant DCS, and preprocessing the data;
step five, substituting the real-time characteristic parameter data as input into the trained boiler combustion characteristic model;
step six, optimizing boiler operation adjustable parameters by adopting an improved non-dominated sorting genetic algorithm INSGA-II as an optimization algorithm and combining an optimization target and a boiler combustion characteristic model;
and seventhly, selecting an optimal solution from the Pareto solution set by engineering personnel according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to a DCS (distributed control system) to realize optimal control.
2. The method according to claim 1, wherein in the first step, the characteristic parameters collected from the power plant DCS system are boiler load, coal supply amount, flue gas oxygen content, primary air amount, primary air temperature, secondary air amount, secondary air temperature, fly ash carbon content; the characteristic parameter historical data collected from the power plant DCS system is data obtained after data in a boiler fault operation time period are removed; and preprocessing the data by adopting a normalization processing method.
3. The method for the optimized control of the on-line combustion of the boiler as claimed in claim 1, wherein in the second step, the method for clustering and resampling the boiler operation parameter data according to the boiler load distribution by using the Gaussian mixture model GMM comprises the following steps:
(1) setting a sample set as X and the number of samples as N; according to the value range (x) of the load parameter corresponding to the sample datamin，χmax) The interval is divided into equal parts, wherein the division standard is that each interval conforms to the range of 10MW, and the interval is divided into equal parts, namely K equal to round ((x) andmax-xmin) 10) (K is more than or equal to 2); and a Gaussian mixture model of the boiler load is provided with K components (K is K);
(2) initializing a Gaussian mixture model, wherein the formula is as follows:
Figure FDA0002564416590000021
in the formula, phikIs a mixing coefficient, which satisfies
Figure FDA0002564416590000022
Figure FDA0002564416590000023
Representing the kth component in the model;
setting the mixing coefficient of each component as 1/k, setting the mean value of each component as the central value of the corresponding coincidence interval, and setting the initial variance as 20;
(3) clustering the sample data according to a Gaussian mixture model algorithm to obtainTo the clustered sample set C ═ C1,C2,C3,...,Ck-if the absolute value of the load mean difference between the two closest clusters of the convergence is less than 5, k-1 → k, and returning to perform step (2);
(4) after the final clustering result is obtained, resampling the boiler data according to the clustering result in the boiler data; the sampling rule is as follows: the cluster after clustering corresponds to a mixing coefficient phikMore than 1/K, less than 10 of data variance, and resampling when the two conditions are met, wherein the specific sampling formula is as follows:
Figure FDA0002564416590000024
where round () is a rounding function.
4. The boiler on-line combustion optimization control method of claim 1, characterized in that in the third step, the improved pollination algorithm SFPA optimization SVM is adopted to establish a boiler combustion characteristic model, and the boiler combustion characteristic model is applied to boiler combustion optimization; the boiler load, coal feeding quantity, flue gas oxygen content, primary air quantity, primary air temperature, secondary air quantity, secondary air temperature and fly ash carbon content are used as input variables of the model, NOxAnd (4) taking the discharge amount and the boiler thermal efficiency as output variables, and training a boiler combustion characteristic model.
5. The boiler on-line combustion optimization control method of claim 1, characterized in that in the third step, the simulated annealing algorithm and the pollination algorithm are combined to form a new SFPA optimization algorithm; then, the SFPA optimization algorithm is used for carrying out parameter optimization on the kernel function parameter sigma, the penalty coefficient C and the insensitive loss function of the SVM, and the method comprises the following steps,
(1) initializing an SFPA optimization algorithm, determining the value ranges of C, sigma and three parameters, setting the number of flower populations as N and the maximum iteration number as itermaxThe conversion probability is P, the initial temperature is T, and the annealing constant is theta;
(2) randomly generating N solutions, wherein each solution vector corresponds to a three-dimensional vector (C, sigma,), and finding the optimal solution and the minimum error f under the current conditionmin；
(3) Determining the error value of each Sol (i) at the current temperature according to
Figure FDA0002564416590000031
Wherein Sol (i) is a certain value of the current solution vector, best is the optimal value in the whole situation, and T is the initial temperature of annealing;
(4) finding a certain substitute value of the global optimum value from all Sol (i) according to the roulette strategy, and recording the substitute value as best _ plus;
(5) when P > rand, updating the solution vector by using the following formula, and performing the border crossing processing of the solution vector:
Figure FDA0002564416590000032
Figure FDA0002564416590000033
wherein λ is 1.5, M is a constant,
Figure FDA0002564416590000034
(λ) is the standard gamma function;
(6) when P < rand, updating the solution vector by using the following formula, and performing the border crossing processing of the solution vector:
Figure FDA0002564416590000035
wherein the random number is uniformly distributed between 0 and 1,
Figure FDA0002564416590000041
and
Figure FDA0002564416590000042
pollen of different flowers of the same species;
(7) error of the new solution formed in step (5) or (6)
Figure FDA0002564416590000043
Make a judgment if
Figure FDA0002564416590000044
Is less than fminIf not, the current solution is reserved;
(8) if the minimum error corresponding to the formed new solution is smaller than the global minimum error, updating the global optimum solution and the global minimum error;
(9) carrying out annealing operation;
(10) determine f thereofminWhether the prediction accuracy of the boiler combustion modeling is achieved, and if so, the process terminates and outputs a set of solutions (C, σ,) that are optimal at that time and a global minimum error fminOtherwise, jumping to the step (3) to continue searching.
6. The boiler on-line combustion optimization control method according to claim 1, wherein in the sixth step, the adjustable boiler operation parameters are coal feeding amount, primary air amount, secondary air amount and flue gas oxygen content, the adjustable boiler operation parameters are used as input variables of an optimization process to carry out optimization, and the rest non-adjustable parameters are kept unchanged as fixed values in the optimization process; the multi-objective optimization objective is as follows:
Figure FDA0002564416590000045
in the formula (I), the compound is shown in the specification,
Figure FDA0002564416590000046
is NO of boilerxDischarge concentration, fη(x) Is the thermal efficiency of the boiler, xiFor the ith optimizable boiler operating variable, AiAnd BiThe value range of the optimized parameter is obtained;
in the sixth step, the INSGA-II algorithm is improved by an NSGA-II algorithm, and the improvement method comprises the following steps:
(1) by enlarging the size of the first generation population, the evolution of the population is accelerated at the initial stage, and the convergence of the algorithm is improved; if the population individuals are N, the number of the individuals of the initialized population can be set to be between 1.5N and 2N;
(2) probability operation is introduced to the selection operator, so that the diversity of the population is expanded, and the distributivity of the solution set is improved; the formula for the calculation of the probability operation is:
Figure FDA0002564416590000047
in the formula, omega is a probability selection operator parameter, and g is an evolution algebra;
(3) introducing a mixed crossover operator, and dynamically adjusting the search space of the algorithm; the mixed crossover operator introduces a Gaussian distribution crossover operator NDX on the basis of an SBX crossover operator, and adaptively adjusts the weights of the two crossover operators; let u be the random number generated by uniform distribution over the interval (0,1), and r ═ N (0,1) | be the value of the gaussian distributed random variable, when u is less than or equal to 0.5, the update formula of the individual under the hybrid crossover operator is:
Figure FDA0002564416590000051
when u > 0.5, the update formula of the individual under the mixed crossover operator is as follows:
Figure FDA0002564416590000052
in the formula, x1,iAnd x2,iThe value of the ith variable of the individual child generated by the hybrid crossover operator; m ═ p1,i+p2,i,N＝p1,i-p2,i，p1,iAnd p2,iIs the value of the ith variable of the parent individual, G is the current iteration number, G is the total iteration number, ηcIs a hybrid crossover operator parameter;
the optimization method of the INSGA-II algorithm comprises the following steps:
(1) initializing INSGA-II algorithm: setting the population size to be N, initializing a population P0Scale N of01.5N to 2N, maximum number of iterations G, probability selection operator parameter ω, hybrid crossover operator parameter ηc；
(2) Randomly generating N0Individuals as parent population P0；
(3) The current iteration number g is 0, and the parent population P is treated0Performing probability selection, mixed crossing and mutation operations to generate a filial generation population Q0；
(4) The parent population PtAnd the offspring population QtMerging to generate new population RtAnd for new population RtPerforming fast non-dominated sorting;
(5) from the new population R using elite strategytN excellent individuals are selected to generate a new parent population Pt+1；
(6) For new population Pt+1Performing probability selection, mixed crossing and mutation operations to generate a new filial generation population Qt+1；
(7) And (4) if the current iteration time G is not less than the maximum iteration time G, ending the algorithm, otherwise, G is G +1, and jumping to the step (4) for circulation.
7. The boiler on-line combustion optimization control method according to claim 1, wherein in step seven, the optimization bias logic is added on the DCS system side, and the optimization bias is superposed on the control command of the optimized variable; and the difference between the optimized coal feeding quantity, the optimized primary air quantity, the optimized secondary air quantity and the optimized flue gas oxygen content and the actual coal feeding quantity, the optimized primary air quantity, the optimized secondary air quantity and the optimized flue gas oxygen content is used as an optimization bias and sent to a DCS (distributed control System) to realize the optimized control of the on-line combustion of the boiler.
8. A computer device, characterized in that the computer device comprises a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to carry out the steps of:
acquiring characteristic parameter historical data and NO of coal-fired boiler from DCS (distributed control System) of power plantxDischarge amount and thermal efficiency data, and carrying out data pretreatment;
clustering and dividing the processed boiler operation data according to the distribution of load working conditions by using a Gaussian Mixture Model (GMM) algorithm, and then resampling the data;
taking the resampled data as sample data to construct a boiler combustion characteristic model; the boiler combustion characteristic model is obtained by optimizing an SVM (support vector machine) based on an improved pollination algorithm SFPA (small form-factor power amplifier);
acquiring real-time characteristic parameter data of a boiler from a power plant DCS (distributed control System), and preprocessing the data;
taking the real-time characteristic parameter data as input, and substituting the input into the trained boiler combustion characteristic model;
an improved non-dominated sorting genetic algorithm INSGA-II is adopted as an optimization algorithm, and an optimization target and a boiler combustion characteristic model are combined to optimize boiler operation adjustable parameters;
and (4) selecting an optimal solution from the Pareto solution set by engineering personnel according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to a DCS (distributed control system) to realize optimal control.
9. An online combustion optimization control system of a boiler for implementing the online combustion optimization control method of the boiler according to any one of claims 1 to 7, the online combustion optimization control system of the boiler comprising:
the first data preprocessing module is used for acquiring the characteristic parameter historical data and NO of the coal-fired boiler from the DCS of the power plantxDischarge amount and thermal efficiency data, and carrying out data pretreatment;
the data resampling module is used for clustering and dividing the processed boiler operation data according to the distribution of the load working conditions by using a Gaussian Mixture Model (GMM) algorithm and resampling the data;
the boiler combustion characteristic model building module is used for building a boiler combustion characteristic model by taking the resampled data as sample data;
the second data preprocessing module is used for acquiring real-time characteristic parameter data of the boiler from the power plant DCS and preprocessing the data;
the real-time data input module is used for taking the real-time characteristic parameter data as input and substituting the input into the trained boiler combustion characteristic model;
the boiler operation adjustable parameter processing module is used for optimizing boiler operation adjustable parameters by adopting an improved non-dominated sorting genetic algorithm INSGA-II as an optimization algorithm and combining an optimization target and a built boiler combustion characteristic model;
and the optimal offset calculation module is used for selecting an optimal solution from the Pareto solution set according to the actual operation requirement of the power plant boiler, comparing the optimal solution with the real-time parameters of the boiler, calculating the optimal offset, and sending the optimal offset to the DCS to realize optimal control.
CN202010618483.7A 2020-07-01 2020-07-01 Boiler on-line combustion optimization control method and system and computer equipment Active CN111765445B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010618483.7A CN111765445B (en) 2020-07-01 2020-07-01 Boiler on-line combustion optimization control method and system and computer equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010618483.7A CN111765445B (en) 2020-07-01 2020-07-01 Boiler on-line combustion optimization control method and system and computer equipment

Publications (2)

Publication Number Publication Date
CN111765445A true CN111765445A (en) 2020-10-13
CN111765445B CN111765445B (en) 2021-10-15

Family

ID=72724473

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010618483.7A Active CN111765445B (en) 2020-07-01 2020-07-01 Boiler on-line combustion optimization control method and system and computer equipment

Country Status (1)

Country Link
CN (1) CN111765445B (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112197262A (en) * 2020-10-24 2021-01-08 四川泸天化创新研究院有限公司 Intelligent control method for coal-fired boiler of circulating fluidized bed
CN112349357A (en) * 2020-11-05 2021-02-09 大连英达士智能科技有限公司 Long-term batching method and system for production of non-ferrous copper
CN112651568A (en) * 2020-12-31 2021-04-13 新奥数能科技有限公司 Boiler load dynamic adjustment method and device, control terminal and storage medium
CN112668117A (en) * 2020-12-28 2021-04-16 上海康恒环境股份有限公司 Design method of primary air main pipe of garbage incinerator
CN112966436A (en) * 2021-03-02 2021-06-15 南京工程学院 Boiler efficiency and NO are taken into considerationxMulti-objective operational optimization method for emissions
CN113222244A (en) * 2021-05-10 2021-08-06 刘旸 Online boiler combustion optimization method based on flame combustion image judgment
CN113705890A (en) * 2021-08-27 2021-11-26 太原理工大学 Diesel engine pollutant emission control method based on approximate model
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
CN114060825A (en) * 2021-11-15 2022-02-18 南方电网电力科技股份有限公司 Multi-pollutant co-processing method in waste incinerator and related device
CN114880927A (en) * 2022-04-29 2022-08-09 广东大唐国际雷州发电有限责任公司 Intelligent power plant monitoring method, system, equipment and storage medium
CN117034787A (en) * 2023-10-10 2023-11-10 南京远思智能科技有限公司 Modelica modeling and genetic algorithm-based coal-fired boiler operation method
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 (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101498457A (en) * 2009-03-02 2009-08-05 杭州电子科技大学 Boiler combustion optimizing method
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
CN104763999A (en) * 2015-03-04 2015-07-08 内蒙古瑞特优化科技股份有限公司 Power plant pulverized coal boiler combustion performance online optimizing method and system
CN109325313A (en) * 2018-11-01 2019-02-12 大唐环境产业集团股份有限公司 Based on improvement quantum telepotation boiler of power plant NOXPrediction model device
CN110793059A (en) * 2019-11-13 2020-02-14 南京昆岳智能电力科技有限公司 Intelligent combustion comprehensive optimization control method for boiler
US20200175435A1 (en) * 2018-11-30 2020-06-04 Doosan Heavy Industries & Construction Co., Ltd. System and method for optimizing combustion of boiler

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101498457A (en) * 2009-03-02 2009-08-05 杭州电子科技大学 Boiler combustion optimizing method
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
CN104763999A (en) * 2015-03-04 2015-07-08 内蒙古瑞特优化科技股份有限公司 Power plant pulverized coal boiler combustion performance online optimizing method and system
CN109325313A (en) * 2018-11-01 2019-02-12 大唐环境产业集团股份有限公司 Based on improvement quantum telepotation boiler of power plant NOXPrediction model device
US20200175435A1 (en) * 2018-11-30 2020-06-04 Doosan Heavy Industries & Construction Co., Ltd. System and method for optimizing combustion of boiler
CN110793059A (en) * 2019-11-13 2020-02-14 南京昆岳智能电力科技有限公司 Intelligent combustion comprehensive optimization control method for boiler

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112197262B (en) * 2020-10-24 2023-06-27 四川泸天化创新研究院有限公司 Intelligent control method for circulating fluidized bed coal-fired boiler
CN112197262A (en) * 2020-10-24 2021-01-08 四川泸天化创新研究院有限公司 Intelligent control method for coal-fired boiler of circulating fluidized bed
CN112349357A (en) * 2020-11-05 2021-02-09 大连英达士智能科技有限公司 Long-term batching method and system for production of non-ferrous copper
CN112668117A (en) * 2020-12-28 2021-04-16 上海康恒环境股份有限公司 Design method of primary air main pipe of garbage incinerator
CN112651568A (en) * 2020-12-31 2021-04-13 新奥数能科技有限公司 Boiler load dynamic adjustment method and device, control terminal and storage medium
CN112966436A (en) * 2021-03-02 2021-06-15 南京工程学院 Boiler efficiency and NO are taken into considerationxMulti-objective operational optimization method for emissions
CN112966436B (en) * 2021-03-02 2022-02-18 南京工程学院 Boiler efficiency and NO are taken into considerationxMulti-objective operational optimization method for emissions
CN113222244A (en) * 2021-05-10 2021-08-06 刘旸 Online boiler combustion optimization method based on flame combustion image judgment
CN113705890A (en) * 2021-08-27 2021-11-26 太原理工大学 Diesel engine pollutant emission control method based on approximate model
CN113705890B (en) * 2021-08-27 2023-06-20 太原理工大学 Diesel engine emission pollutant control method based on approximate model
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
CN114060825A (en) * 2021-11-15 2022-02-18 南方电网电力科技股份有限公司 Multi-pollutant co-processing method in waste incinerator and related device
CN114060825B (en) * 2021-11-15 2024-01-26 南方电网电力科技股份有限公司 Multi-pollutant cooperative treatment method in garbage incinerator and related device
CN114880927A (en) * 2022-04-29 2022-08-09 广东大唐国际雷州发电有限责任公司 Intelligent power plant monitoring method, system, equipment and storage medium
CN117034787A (en) * 2023-10-10 2023-11-10 南京远思智能科技有限公司 Modelica modeling and genetic algorithm-based coal-fired boiler operation method
CN117034787B (en) * 2023-10-10 2023-12-29 南京远思智能科技有限公司 Modelica modeling and genetic algorithm-based coal-fired boiler operation method

Also Published As

Publication number Publication date
CN111765445B (en) 2021-10-15

Similar Documents

Publication Publication Date Title
CN111765445B (en) Boiler on-line combustion optimization control method and system and computer equipment
Kusiak et al. Combustion efficiency optimization and virtual testing: A data-mining approach
CN104534507A (en) Optimal control method for combustion of boiler
Tang et al. Adaptive nonlinear model predictive control of NOx emissions under load constraints in power plant boilers
CN108490790A (en) A kind of overheating steam temperature active disturbance rejection cascade control method based on multiple-objection optimization
Zhang et al. Optimization of wind power and its variability with a computational intelligence approach
CN110935567A (en) Thermal power generating unit dry-type electric precipitator optimization control method and system
Xie et al. Dynamic multi-objective intelligent optimal control toward wastewater treatment processes
Ding et al. Application of fuzzy immune algorithm and soft computing in the design of 2-DOF PID controller
CN114117925A (en) Comprehensive energy system planning method considering high-proportion distributed photovoltaic consumption
Kusiak et al. Clustering-based performance optimization of the boiler–turbine system
Zheng et al. Data-driven based multi-objective combustion optimization covering static and dynamic states
CN111027760A (en) Power load prediction method based on least square vector machine
CN107038489B (en) Multi-target unit combination optimization method based on improved NBI method
CN109506248B (en) Case reasoning boiler combustion optimization method based on online optimization
Ling et al. A thermal load forecasting algorithm based on trajectory tracking
Hu et al. Adaptive internal model control of SCR denitration system based on multi-objective optimization
CN114637212B (en) Oxygen content prediction control method and system
Liang et al. Multi-objective stochastic optimal day-ahead scheduling for micro-grid based on scenario and PSO
Gao et al. Artificial bee colony optimization of NOx emission and reheat steam temperature in a 1000 MW boiler
Shang et al. Parameter optimization of PID controllers by reinforcement learning
Vakil-Baghmisheh et al. The design of PID controllers for a Gryphon robot using four evolutionary algorithms: a comparative study
Kang et al. Brain storming optimization algorithm for heating dispatch scheduling of thermal power plant
CN113824116B (en) Hybrid time scale DRL-based comprehensive energy online scheduling method
Logan et al. Improved Operation Forecasting of Air Blowers in a Wastewater Treatment Plant via Machine Learning-based Models

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