CN116227350A

CN116227350A - Multi-target optimization method and device for boiler

Info

Publication number: CN116227350A
Application number: CN202310207029.6A
Authority: CN
Inventors: 王宏武; 杨荣祖; 张奔; 穆祺伟; 于龙文; 王耀文; 刘欢; 王汀; 谢天; 翟鹏程
Original assignee: Xian Thermal Power Research Institute Co Ltd; Xian Xire Energy Saving Technology Co Ltd
Current assignee: Xian Thermal Power Research Institute Co Ltd; Xian Xire Energy Saving Technology Co Ltd
Priority date: 2023-03-06
Filing date: 2023-03-06
Publication date: 2023-06-06

Abstract

The disclosure provides a multi-target optimization method and device for a boiler, which relate to the technical field of deep learning and comprise the following steps: acquiring a boiler NOx emission characteristic model and a boiler efficiency model which are built based on a support vector regression mechanism; determining an objective function; inputting operation parameters into a boiler NOx emission characteristic model and a boiler efficiency model for parameter configuration, wherein the operation parameters comprise the variables to be optimized; optimizing variables to be optimized in a boiler NOx emission characteristic model and a boiler efficiency model based on a multi-target genetic algorithm and an objective function of pre-configured parameters to obtain a pareto solution set containing a plurality of feasible solutions; the target feasible solution is obtained from the pareto solution set. Therefore, the improved multi-target genetic algorithm of NSGA-II can be used for carrying out double-target optimization calculation on boiler efficiency and NOx emission, and the pareto optimal solution set obtained through the algorithm is used for carrying out optimization guidance on boiler operation parameters, so that the prediction accuracy is high, the calculation speed is high, and the generalization capability and the model universality are strong.

Description

Multi-target optimization method and device for boiler

Technical Field

The disclosure relates to the technical field of deep learning, in particular to a multi-objective optimization method and device for a boiler.

Background

The basis and key of the modeling and optimization of the power station boiler combustion numerical model is to build a boiler numerical model with good prediction performance. Early days, computational fluid mechanics mathematical models were the main, but the computation was complicated and time-consuming, the computation accuracy was not high, and the demand of online computation could not be supported. With the advent of neural network algorithm, support vector machine model and genetic algorithm, the theoretical tool of online numerical modeling calculation is enriched, nonlinear modeling calculation and optimization calculation are expected to be carried out by utilizing power plant database data, and the prediction accuracy is high, the calculation speed is high, and the generalization capability and model universality are strong.

How to realize high boiler efficiency and low NOx emission by optimizing the obtained operation data through boiler numerical modeling calculation is a problem which needs to be solved at present.

Disclosure of Invention

The present disclosure aims to solve, at least to some extent, one of the technical problems in the related art.

An embodiment of a first aspect of the present disclosure provides a multi-objective optimization method for a boiler, including:

acquiring a boiler NOx emission characteristic model and a boiler efficiency model which are built based on a support vector regression mechanism;

determining an objective function corresponding to the boiler NOx emission characteristic model and the boiler efficiency model, wherein the objective function comprises a variable to be optimized and a value range corresponding to the variable to be optimized;

Inputting operation parameters corresponding to the current boiler working conditions into the boiler NOx emission characteristic model and the boiler efficiency model for parameter configuration, wherein the operation parameters comprise the variables to be optimized;

optimizing variables to be optimized in the boiler NOx emission characteristic model and the boiler efficiency model based on a multi-objective genetic algorithm with pre-configured parameters and the objective function to obtain a pareto solution set containing a plurality of feasible solutions;

and obtaining a target feasible solution from the pareto solution set containing a plurality of feasible solutions.

Embodiments of a second aspect of the present disclosure provide a multi-objective optimization apparatus for a boiler, including:

the first acquisition module is used for acquiring a boiler NOx emission characteristic model and a boiler efficiency model which are built based on the support vector regression mechanism;

the determining module is used for determining an objective function corresponding to the boiler NOx emission characteristic model and the boiler efficiency model, wherein the objective function comprises a variable to be optimized and a value range corresponding to the variable to be optimized;

the parameter configuration module is used for inputting operation parameters corresponding to the current boiler working condition into the boiler NOx emission characteristic model and the boiler efficiency model for parameter configuration, and the operation parameters comprise the variable to be optimized;

The optimization module is used for optimizing variables to be optimized in the boiler NOx emission characteristic model and the boiler efficiency model based on a multi-objective genetic algorithm with pre-configured parameters and the objective function to obtain a pareto solution set containing a plurality of feasible solutions;

and the second acquisition module is used for acquiring a target feasible solution from the pareto solution set containing a plurality of feasible solutions.

An embodiment of a third aspect of the present disclosure provides an electronic device, including: the system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor realizes the multi-target optimization method of the boiler according to the embodiment of the first aspect of the present disclosure when the processor executes the program.

An embodiment of a fourth aspect of the present disclosure proposes a non-transitory computer readable storage medium storing a computer program which, when executed by a processor, implements a multi-objective optimization method for a boiler as proposed by an embodiment of the first aspect of the present disclosure.

The multi-objective optimization method and device for the boiler provided by the disclosure have the following beneficial effects:

in the embodiment of the disclosure, a boiler NOx emission characteristic model and a boiler efficiency model which are built based on a support vector regression mechanism are firstly obtained, then an objective function corresponding to the boiler NOx emission characteristic model and the boiler efficiency model is determined, wherein the objective function comprises a variable to be optimized and a value range corresponding to the variable to be optimized, then an operation parameter corresponding to the current boiler working condition is input into the boiler NOx emission characteristic model and the boiler efficiency model for parameter configuration, the operation parameter comprises the variable to be optimized, then a multi-objective genetic algorithm with parameters being pre-configured and the objective function are based on, so that the variable to be optimized in the boiler NOx emission characteristic model and the boiler efficiency model is optimized, a pareto solution set comprising a plurality of feasible solutions is obtained, and finally the target feasible solution is obtained from the pareto solution set comprising the plurality of feasible solutions. Therefore, the improved multi-target genetic algorithm of NSGA-II can be used for carrying out double-target optimization calculation on boiler efficiency and NOx emission, and the pareto optimal solution set obtained through the algorithm is used for carrying out optimization guidance on boiler operation parameters, so that the prediction accuracy is high, the calculation speed is high, and the generalization capability and the model universality are strong.

Additional aspects and advantages of the disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the disclosure.

Drawings

The foregoing and/or additional aspects and advantages of the present disclosure will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a schematic flow chart of a multi-objective optimization method for a boiler according to an embodiment of the present disclosure;

FIG. 2 shows a schematic diagram of boiler thermal efficiency and NOx production for each feasible solution in the pareto solution set;

FIG. 3 is a schematic flow chart of a multi-objective optimization method for a boiler according to an embodiment of the present disclosure;

FIG. 4 is a block diagram of a multi-objective optimization device for a boiler according to an embodiment of the present disclosure;

fig. 5 illustrates a block diagram of an exemplary computer device suitable for use in implementing embodiments of the present disclosure.

Detailed Description

Embodiments of the present disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and intended for the purpose of explaining the present disclosure and are not to be construed as limiting the present disclosure.

The following describes a multi-objective optimization method, apparatus, computer device, and storage medium for a boiler according to an embodiment of the present disclosure with reference to the accompanying drawings.

It should be noted that, the execution body of the boiler multi-objective optimization method in the embodiment of the present disclosure is a boiler multi-objective optimization device, which may be implemented by software and/or hardware, and the boiler multi-objective optimization method set forth in the embodiment of the present disclosure will be described below with "multi-objective optimization device" as the execution body, which is not limited herein.

Fig. 1 is a schematic flow chart of a multi-objective optimization method for a boiler according to an embodiment of the present disclosure.

As shown in fig. 1, the multi-objective optimization method of the boiler may include the steps of:

and step 101, acquiring a boiler NOx emission characteristic model and a boiler efficiency model which are built based on a support vector regression mechanism.

The support vector regression machine can be SVR, is called Support Vactor Regression in English, is a supervised learning model with related learning algorithm, and is used for analyzing data for classification and regression analysis. In support vector regression, the straight line required to fit the data is called the hyperplane.

The goal of the support vector machine algorithm is to find a hyperplane in n-dimensional space that explicitly classifies data points. The data points on both sides of the hyperplane closest to the hyperplane are referred to as support vectors. These affect the position and orientation of the hyperplane and thus help construct the SVM.

The support vector machine (Support Vector Machine, SVM) is a generalized linear classifier for binary classification of data in a supervised learning mode, and the decision boundary is the maximum margin hyperplane for solving the learning sample.

The method comprises the following steps of selecting input variables of a boiler NOx emission characteristic model: the total fuel quantity, the total air quantity, the power generation, the coal feeding quantity of 6 coal feeders, the primary air quantity of 6 layers, the opening of 6 secondary air baffles, the opening of 6 SOFA air baffles, the oxygen quantity of tail flue gas, the temperature of outlet flue gas of the air preheater, the received base carbon, hydrogen, oxygen, nitrogen, sulfur, ash, moisture, volatile matters and low-position heating value, wherein the total input variables are 38, and the output variables are NOx emission under the standard state of SCR inlet.

The method comprises the following steps of selecting input variables of a boiler efficiency model: in addition to the 38 variables described above, the fly ash carbon content and slag carbon content were increased. A total of 40 input variables and the output variable is boiler efficiency.

Further, according to the regression algorithm of the support vector machine, a boiler NOx model and an efficiency model can be respectively established according to the data and the model structure.

The function model of the boiler NOx emission characteristic model and the boiler efficiency model is established as follows:

wherein, the liquid crystal display device comprises a liquid crystal display device,

y _[η] ＝f _[η] (X _η )

In the method, in the process of the invention,

for NOx emissions, y _[η] For boiler efficiency>

SVM model built for NOx emissions, f _[η] (X _η ) SVM model built for boiler efficiency, X _N For the input vector of the NOx emission SVM model, X _η Is the input vector of the efficient SVM model.

The combustion optimization of the boiler is based on the established combustion characteristic model capable of accurately predicting the combustion performance index of the boiler, and the boiler adjustable operation parameters can be optimized by taking the minimized NOx emission and the maximized boiler efficiency as two objective functions of the combustion multi-objective optimization of the boiler by adopting the established boiler NOx emission characteristic model and the boiler efficiency model based on the support vector regression machine.

It can be understood that the SVM model of boiler efficiency and NOx multivariable boiler efficiency is established, the modeling problem of complex nonlinear coupling relation is effectively solved, the prediction accuracy of the simulation model by utilizing the SVM model is high, the fitting degree is large, the generalization capability is strong, and the operation economy under different working conditions and coal characteristics can be predicted accurately.

Step 102, determining an objective function corresponding to the boiler NOx emission characteristic model and the boiler efficiency model, wherein the objective function comprises a variable to be optimized and a value range corresponding to the variable to be optimized.

The objective function corresponding to the boiler NOx emission characteristic model and the boiler efficiency model may be:

max y _[η] ＝f _[η] (A,SOFA,SA _i )

in the method, in the process of the invention,

for NOx emissions, y _[η] For boiler efficiency, f _[NOx] (A,SOFA,SA _i ) For boiler NOx emission characteristic model, f _[η] (A,SOFA,SA _i ) Is boiler efficiency model, A, SOFA, SA _i The variables to be optimized are a boiler NOx emission characteristic model and a boiler efficiency model. Where i=1, 2, …,6.

The variables to be optimized comprise 11 parameters, namely total air quantity (A) of the boiler, opening degrees (SOFA) of four SOFA air baffles and opening degrees (SA) of 6 secondary air baffles, as target variables for optimization. In the constraint condition, namely, the value range corresponding to the variable to be optimized is as follows: the value range of the total air quantity (A) of the boiler is 8B-11B, the value range of the opening degree (SOFA) of the SOFA air baffle is 0-100%, and the value range of the opening degree (SA) of each secondary air baffle is 0-70%.

And step 103, inputting the operation parameters corresponding to the current boiler working conditions into the boiler NOx emission characteristic model and the boiler efficiency model for parameter configuration, wherein the operation parameters comprise the variables to be optimized.

The boiler condition may be, among other things, a boiler load, such as 526.2MW. In general, the operating parameters of the operating conditions to be optimized may be used as inputs.

It is understood that the condition to be optimized may be a condition where NOx emissions are relatively high and boiler efficiency is relatively low. The focus of optimization is to increase efficiency as much as possible while maintaining low NOx emission levels.

List one

As shown in the above Table one, the current boiler conditions corresponded to a boiler load of 526.2MW and a corresponding NOx emission of 587.8mg/Nm ³ The boiler efficiency is 90.6%, which is a working condition needing to be optimized.

The operation parameters corresponding to the current boiler working condition can comprise 11 variables to be optimized, namely total boiler air quantity (A), four SOFA air baffle opening degrees (SOFA) and 6 secondary air baffle opening degrees (SA), and can also comprise input and output variables, such as total fuel quantity, power generation power, coal feeding quantity of 6 coal feeders, 6 layers of primary air quantity, tail flue gas oxygen quantity, air preheater outlet gas temperature, received base carbon, hydrogen, oxygen, nitrogen, sulfur, ash, moisture, volatile matters, low-level heating value, NOx emission under the standard state of SCR inlet, fly ash carbon content, slag carbon content, boiler efficiency and the like.

Step 104, optimizing variables to be optimized in the boiler NOx emission characteristic model and the boiler efficiency model based on a multi-objective genetic algorithm with pre-configured parameters and an objective function to obtain a pareto solution set containing a plurality of feasible solutions.

The multi-objective genetic algorithm can be an NSGA-II multi-objective genetic algorithm, wherein the initial population size of the NSGA-II multi-objective genetic algorithm is 200, the initial value of the self-adaptive crossover probability is 0.9, the initial value of the self-adaptive mutation probability is 0.009, the optimal front-end individual coefficient is 0.3, and the maximum evolution algebra is 2000.

It should be noted that the optimization scheme can be implemented by using the established boiler NOx emission characteristic model and boiler efficiency model by adopting the improved NSGA-II multi-objective genetic algorithm. It should be noted that, the optimization result is a Pareto (Pareto) solution set containing a plurality of possible solutions, and the obtained results are adjustable parameter sets with less NOx emission and higher boiler efficiency.

As shown in the following tables two and three, a comparison of parameters before and after optimization is shown:

watch II

Watch III

It can be seen that in the optimization proposal given, by properly adjusting the secondary air, the SOFA air baffle and the total combustion air quantity, the goals of high efficiency and low NOx emission, especially the boiler efficiency, can be achieved, and the total air quantity of the boiler after optimization is higher than the value before optimization because of the upward adjustment. The boiler efficiency is improved by 0.6% in the result of the feasible solution b, and the NOx emission is kept at the original lower level, so that the running parameters in the optimization proposal in the feasible solution have stronger guiding significance.

After optimization, the boiler efficiency of all feasible solutions is improved to a certain extent, the boiler efficiency is increased to more than 90.8%, the boiler efficiency and NOx emission almost show a linear relation, the NOx generation amount is increased along with the improvement of the efficiency, the NOx emission and the efficiency are improved at the same time, and the efficiency is improved to 91.3% at the maximum while the NOx emission is kept lower than before optimization.

Step 105, obtaining a target feasible solution from the pareto solution set comprising a plurality of feasible solutions.

The target feasible solution may be a target operation parameter that may be selected currently.

Alternatively, the boiler efficiency value and the NOx emission value corresponding to each feasible solution in the pareto solution set may be first determined, and then the feasible solution with the boiler efficiency value greater than the first threshold and the NOx emission value less than the second threshold may be used as the target feasible solution.

Fig. 2 shows the boiler efficiency values (boiler thermal efficiency) and NOx emission values (NOx production) corresponding to the respective feasible solutions in the pareto solution set.

The first threshold may be a threshold of the boiler efficiency value, such as 91.3%.

Wherein the second threshold may be a threshold for NOx emissions, such as 523.7mg/Nm ³ 。

Fig. 3 is a schematic flow chart of a multi-objective optimization method for a boiler according to an embodiment of the present disclosure.

As shown in fig. 3, the multi-objective optimization method of the boiler may include the steps of:

step 201, acquiring a plurality of groups of boiler operation data corresponding to different boiler loads from a database as a sample set.

The database may include boiler operation data corresponding to various different sizes of boiler loads, and the data may be used as research data.

For example, boiler operation data of about 500MW, about 650MW, and about 800MW are not limited herein. As an example, 200 sets of boiler operation data may be selected, including research data for a variety of different load conditions, such as total fuel amount, total air volume, power generation, 6 coal feeder coal feed amounts, 6 primary air volumes, 6 secondary air baffle openings, 6 SOFA air baffle openings, afterburner stack gas oxygen amounts, air preheater outlet gas temperatures, received base carbon, hydrogen, oxygen, nitrogen, sulfur, ash, moisture, volatile and low heat generation values, NOx emissions at SCR inlet standard conditions, fly ash carbon content, slag carbon content, boiler efficiency, and the like, without limitation herein.

Step 202, dividing the sample set into a training sample set and a test sample set, and respectively performing normalization processing.

For example, the sample sets may be divided into N groups, where N may be a positive integer greater than 50.

M groups are taken as training set samples in the N groups of data to train, learn and establish a support vector machine model, the number of the training set samples at least accounts for more than 80% of the total data, and the remaining N-M groups are taken as test sample sets to verify the quality of the model. If N is 100, 80 sets of samples can be taken as training set samples from the 100 sets of data to build a model, and the other 20 sets of data are taken as test set samples to verify the performance of the built model.

In order to eliminate the influence of the difference in order of magnitude between the input variables on modeling, unified normalization processing is carried out on the training sample set and the test sample set data, wherein the formulas are as follows, and the normalization is between-1 and 1.

Step 203, optimizing the kernel function parameters by using a cross-validation method based on the Drosophila algorithm to obtain an optimal parameter combination.

Further, the most widely used Radial Basis Function (RBF) kernel function may be selected as the kernel function for the build model, as specified below. The nonlinear mapping by Radial Basis Function (RBF) kernel functions can have good nonlinear operation capability. Radial Basis Function (RBF) kernel functions are of broad representatives and utility.

Then, the initial value and the value range of the kernel parameter can be selected.

The kernel function parameter g, the penalty factor C and the insensitive loss coefficient epsilon are optimized by using a drosophila optimization algorithm. The parameters of the Drosophila optimization algorithm were set as follows: the population scale of the drosophila is 20, the iteration number is 100, the optimizing range of C is [0, 100], the optimizing range of g is [0, 50], and the optimizing range of epsilon is [0,0.5].

Three important parameters in the model can be optimized by using a drosophila optimization algorithm (FOV), and the optimal (C, g, epsilon) combination is obtained. In the optimizing modeling process, a 5-fold cross validation method is carried out on training set sample data to improve the generalization capability of the model, and the taste concentration is the cross validation Mean Square Error (MSE) of the following formula.

Specifically, a drosophila optimization algorithm is used to find the optimal (C, g, epsilon) combination, and the optimization target is to minimize the mean square error MSE of the training set samples of the formula.

By confirming whether the MSE of the training set sample reaches the minimum, if so, the optimal (C, g, epsilon) combination is obtained, and if not, the iterative calculation is continued until the minimum.

As an example, after the FOA optimization calculation, the NOx emission model of the FOA-SVM is obtained: c=78.2371, g= 0.1288, epsilon=0.0242; obtaining a boiler efficiency model of the FOA-SVM: c=12.7708, g=0.0177, epsilon=0.0021.

And 204, building a boiler NOx emission characteristic model and a boiler efficiency model which are constructed based on a support vector regression machine based on the optimal parameter combination and the radial basis function.

Alternatively, the best (C, g, epsilon) combination obtained can be used to build a boiler NOx emission characteristic model and a boiler efficiency model based on the feed-forward optimization of a Drosophila optimization algorithm to a support vector regression machine.

And 205, testing and verifying the boiler NOx emission characteristic model and the boiler efficiency model by using the training sample set and the testing sample set respectively to obtain the trained boiler NOx emission characteristic model and the trained boiler efficiency model.

For example, if the training sample set has M groups, the learning may be trained by substituting 38 input variables and NOx emissions in the M groups of training sample set samples into the NOx emission SVM model. Input variables and efficiencies in the M sets of training set samples are substituted into the efficiency SVM model to train learning.

And verifying the model by using the N-M group test sample sets, substituting the input variables in the N-M group test sample sets into the NOx emission SVM model to obtain N-M group test NOx emission values.

Substituting the input variables in the N-M group test sample set into the efficiency SVM model to obtain N-M group test efficiency values, and obtaining N-M group test NOx emission values and N-M group test efficiency values. And comparing the tested NOx emission value with the NOx emission value in the tested sample set to obtain a relative error value. And comparing the test efficiency emission value with the efficiency emission value in the test sample set to obtain a relative error value. The prediction accuracy of the model is evaluated.

And 206, determining an objective function corresponding to the boiler NOx emission characteristic model and the boiler efficiency model, wherein the objective function comprises a variable to be optimized and a value range corresponding to the variable to be optimized.

Step 207, inputting the operation parameters corresponding to the current boiler working condition into a boiler NOx emission characteristic model and a boiler efficiency model for parameter configuration, wherein the operation parameters comprise the variable to be optimized.

Step 208, optimizing variables to be optimized in the boiler NOx emission characteristic model and the boiler efficiency model based on a multi-objective genetic algorithm with pre-configured parameters and an objective function to obtain a pareto solution set containing a plurality of feasible solutions.

Step 209, obtaining a target feasible solution from the pareto solution set comprising a plurality of feasible solutions.

It should be noted that, the specific implementation manner of the steps 206 to 209 may refer to the above embodiments, and are not described herein.

In the embodiment of the disclosure, firstly, a plurality of groups of boiler operation data corresponding to different boiler loads are obtained from a database as sample sets, the sample sets are divided into a training sample set and a test sample set, normalization processing is performed respectively, kernel function parameters are optimized by using a cross validation method based on a drosophila algorithm to obtain an optimal parameter combination, then a boiler NOx emission characteristic model and a boiler efficiency model constructed based on a support vector regression machine are established based on the optimal parameter combination and a radial basis kernel function, finally, the training sample set and the test sample set are used for testing and verifying the boiler NOx emission characteristic model and the boiler efficiency model respectively to obtain trained target functions corresponding to the boiler NOx emission characteristic model and the boiler efficiency model, the target functions comprise variables to be optimized, the operating parameters corresponding to the current boiler working conditions are input into the boiler NOx emission characteristic model and the boiler efficiency model to be parameter configuration, the operating parameters comprise the variables to be optimized, then the target parameters can be optimized according to the target genetic algorithm and the target parameters to the target parameters, and the target functions can be obtained from the target models, and the target functions can be optimized, and the target functions can be obtained from the target models. Therefore, a single-target boiler NOx emission characteristic model and a boiler efficiency model can be established, the modeling problem of complex nonlinear coupling relation is effectively solved, the simulation model utilizes the SVM model to predict high precision, high fitting degree and strong generalization capability, can predict operation economy under different working conditions and coal characteristics accurately, establishes a support vector regression feed-forward optimized NOx emission SVM model and a boiler efficiency SVM model based on a drosophila optimization algorithm, performs double-target optimization calculation on boiler efficiency and NOx emission by utilizing an improved NSGA-II multi-target genetic algorithm, and performs optimization guidance on boiler operation parameters through a pareto optimal solution set obtained by the algorithm so as to obtain optimal operation data.

In order to achieve the above embodiments, the present disclosure further proposes a multi-objective optimization device for a boiler.

Fig. 4 is a block diagram of a boiler multi-objective optimization device according to a fourth embodiment of the present disclosure.

As shown in fig. 4, the boiler multi-objective optimization apparatus 400 may include:

a first obtaining module 410, configured to obtain a boiler NOx emission characteristic model and a boiler efficiency model that are built based on a support vector regression mechanism;

a determining module 420, configured to determine an objective function corresponding to the boiler NOx emission characteristic model and the boiler efficiency model, where the objective function includes a variable to be optimized and a value range corresponding to the variable to be optimized;

the parameter configuration module 430 is configured to input an operation parameter corresponding to a current boiler working condition into the boiler NOx emission characteristic model and the boiler efficiency model for parameter configuration, where the operation parameter includes the variable to be optimized;

the optimizing module 440 is configured to optimize variables to be optimized in the boiler NOx emission characteristic model and the boiler efficiency model based on a multi-objective genetic algorithm with pre-configured parameters and the objective function, so as to obtain a pareto solution set including a plurality of feasible solutions;

A second obtaining module 450, configured to obtain a target feasible solution from the pareto solution set including a plurality of feasible solutions.

Optionally, the variable to be optimized at least includes:

the total air quantity of the boiler, the opening degree of at least 4 SOFA air baffles and the opening degree of at least 6 secondary air baffles;

the value range corresponding to the total air quantity of the boiler is 8B-11B, the value range corresponding to the opening of the SOFA air baffle is 0-100%, and the value range of the opening of the secondary air baffle is 0-70%.

Optionally, the second obtaining module is specifically configured to:

determining a boiler efficiency value and a NOx emission value corresponding to each feasible solution in the pareto solution set;

and taking a feasible solution, in which the boiler efficiency value is larger than a first threshold value and the NOx emission value is smaller than a second threshold value, as a target feasible solution.

Optionally, the multi-objective genetic algorithm is an NSGA-II multi-objective genetic algorithm, wherein the initial population size of the NSGA-II multi-objective genetic algorithm is 200, the initial value of the self-adaptive crossover probability is 0.9, the initial value of the self-adaptive mutation probability is 0.009, the optimal front end individual coefficient is 0.3, and the maximum evolution algebra is 2000.

Optionally, the first obtaining module is specifically configured to:

Acquiring a plurality of groups of boiler operation data corresponding to different boiler loads from a database as a sample set;

dividing the sample set into a training sample set and a test sample set, and respectively carrying out normalization treatment;

optimizing kernel function parameters by using a cross verification method based on a drosophila algorithm to obtain an optimal parameter combination;

based on the optimal parameter combination and the radial basis function, establishing a boiler NOx emission characteristic model and a boiler efficiency model which are constructed based on a support vector regression machine;

and respectively testing and verifying the boiler NOx emission characteristic model and the boiler efficiency model by using the training sample set and the testing sample set so as to obtain the trained boiler NOx emission characteristic model and the trained boiler efficiency model.

To achieve the above embodiments, the present disclosure further proposes a computer device including: the system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to realize the multi-target optimization method of the boiler according to the previous embodiment of the disclosure.

To achieve the above embodiments, the present disclosure further proposes a non-transitory computer readable storage medium storing a computer program which, when executed by a processor, implements a multi-objective optimization method for a boiler as proposed in the foregoing embodiments of the present disclosure.

Fig. 5 illustrates a block diagram of an exemplary computer device suitable for use in implementing embodiments of the present disclosure. The computer device 12 shown in fig. 5 is merely an example and should not be construed as limiting the functionality and scope of use of the disclosed embodiments.

As shown in FIG. 5, the computer device 12 is in the form of a general purpose computing device. Components of computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, a bus 18 that connects the various system components, including the system memory 28 and the processing units 16.

Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, and a local bus using any of a variety of bus architectures. By way of example, and not limitation, such architectures include industry Standard architecture (Industry Standard Architecture; hereinafter ISA) bus, micro channel architecture (Micro Channel Architecture; hereinafter MAC) bus, enhanced ISA bus, video electronics standards Association (Video Electronics Standards Association; hereinafter VESA) local bus, and peripheral component interconnect (Peripheral Component Interconnection; hereinafter PCI) bus.

Computer device 12 typically includes a variety of computer system readable media. Such media can be any available media that is accessible by computer device 12 and includes both volatile and nonvolatile media, removable and non-removable media.

Memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (Random Access Memory; hereinafter: RAM) 30 and/or cache memory 32. The computer device 12 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be used to read from or write to non-removable, nonvolatile magnetic media (not shown in FIG. 5, commonly referred to as a "hard disk drive"). Although not shown in fig. 5, a magnetic disk drive for reading from and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading from or writing to a removable non-volatile optical disk (e.g., a compact disk read only memory (Compact Disc Read Only Memory; hereinafter CD-ROM), digital versatile read only optical disk (Digital Video Disc Read Only Memory; hereinafter DVD-ROM), or other optical media) may be provided. In such cases, each drive may be coupled to bus 18 through one or more data medium interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to carry out the functions of the various embodiments of the disclosure.

A program/utility 40 having a set (at least one) of program modules 42 may be stored in, for example, memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of which may include an implementation of a network environment. Program modules 42 generally perform the functions and/or methods in the embodiments described in this disclosure.

The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the computer device 12, and/or any devices (e.g., network card, modem, etc.) that enable the computer device 12 to communicate with one or more other computing devices. Such communication may occur through an input/output (I/O) interface 22. Moreover, the computer device 12 may also communicate with one or more networks such as a local area network (Local Area Network; hereinafter LAN), a wide area network (Wide Area Network; hereinafter WAN) and/or a public network such as the Internet via the network adapter 20. As shown, network adapter 20 communicates with other modules of computer device 12 via bus 18. It should be appreciated that although not shown, other hardware and/or software modules may be used in connection with computer device 12, including, but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, data backup storage systems, and the like.

The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, for example, implementing the methods mentioned in the foregoing embodiments.

In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.

Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present disclosure, the meaning of "a plurality" is at least two, such as two, three, etc., unless explicitly specified otherwise.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and additional implementations are included within the scope of the preferred embodiment of the present disclosure in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present disclosure.

Logic and/or steps represented in the flowcharts or otherwise described herein, e.g., a ordered listing of executable instructions for implementing logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer readable medium may even be paper or other suitable medium on which the program is printed, as the program may be electronically captured, via, for instance, optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

It should be understood that portions of the present disclosure may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. As with the other embodiments, if implemented in hardware, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable Gate Arrays (PGAs), field Programmable Gate Arrays (FPGAs), and the like.

Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.

Furthermore, each functional unit in the embodiments of the present disclosure may be integrated in one processing module, or each unit may exist alone physically, or two or more units may be integrated in one module. The integrated modules may be implemented in hardware or in software functional modules. The integrated modules may also be stored in a computer readable storage medium if implemented in the form of software functional modules and sold or used as a stand-alone product.

The above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, or the like. Although embodiments of the present disclosure have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the present disclosure, and that variations, modifications, alternatives, and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the present disclosure.

Claims

1. A method for multi-objective optimization of a boiler, comprising:

2. The method of claim 1, wherein the step of determining the position of the probe comprises,

the variables to be optimized at least comprise:

3. The method of claim 1, wherein the obtaining the target viable solution from the pareto solution set comprising a plurality of viable solutions comprises:

4. The method of claim 1, wherein the multi-objective genetic algorithm is an NSGA-II multi-objective genetic algorithm, wherein the initial population size of the NSGA-II multi-objective genetic algorithm is 200, the initial value of the adaptive crossover probability is 0.9, the initial value of the adaptive mutation probability is 0.009, the optimal front end individual coefficient is 0.3, and the maximum evolution algebra is 2000.

5. The method of claim 1, wherein the obtaining a model of boiler NOx emission characteristics and a model of boiler efficiency constructed based on a support vector regression mechanism comprises:

6. A multi-objective optimization device for a boiler, comprising:

7. The apparatus of claim 6, wherein,

the variables to be optimized at least comprise:

8. The apparatus of claim 6, wherein the second acquisition module is specifically configured to:

9. The apparatus of claim 6, wherein the multi-objective genetic algorithm is an NSGA-II multi-objective genetic algorithm, wherein the initial population size of the NSGA-II multi-objective genetic algorithm is 200, the initial value of the adaptive crossover probability is 0.9, the initial value of the adaptive mutation probability is 0.009, the optimal front end individual coefficient is 0.3, and the maximum evolution algebra is 2000.

10. The apparatus of claim 6, wherein the first acquisition module is specifically configured to: