Intelligent optimization method for biomass blending combustion condition of coal-fired boiler

Abstract

The invention relates to the technical field of boiler engineering and artificial intelligence modeling, in particular to a coal-fired boiler biomass blending combustion working condition intelligent optimization method, which comprises the following steps: constructing a three-dimensional mathematical model of coal-fired boiler biomass blending combustion based on experimental test data; performing multi-working-condition numerical simulation according to boiler combustion working condition fluctuation in the three-dimensional mathematical model; according to the numerical simulation data, constructing a time-space coupling model for co-combustion of biomass in the coal-fired boiler; the modeling accuracy is improved through a numerical simulation technology, the boiler operation condition is dynamically optimized in combination with space-time modeling, and theoretical support is provided for energy conservation and emission reduction and blending combustion proportion optimization; the method is suitable for environment-friendly transformation and operation optimization of the coal-fired boiler, and carbon emission and pollutant emission can be effectively reduced.

Description

Technical Field

[0001] The present invention relates to boiler engineering technology and artificial intelligence modeling technology, and specifically to an intelligent optimization method for biomass blending conditions in a coal-fired boiler. Background Art

[0002] Coal-fired boilers are important energy production equipment and are widely used around the world. However, their traditional operation mode relies on coal combustion, which not only consumes a huge amount of fossil fuels, but also produces a large amount of greenhouse gases (such as carbon dioxide) and pollutants (such as sulfur oxides and nitrogen oxides), which seriously affect the ecological environment. Driven by the goal of carbon neutrality, biomass, as a renewable energy source, has become one of the important fuels to replace part of coal due to its carbon neutrality characteristics and wide sources.

[0003] Although biomass fuel blending technology has significant environmental and economic benefits, it still faces the following technical difficulties in practical application:

[0004] (1) The physical and chemical properties of biomass fuels vary significantly

[0005] The calorific value of biomass fuel is usually lower than that of coal (about 10-20MJ / kg), and its volatile matter is higher and fixed carbon is lower. This characteristic may lead to uneven combustion rate, thus affecting combustion efficiency; biomass contains high alkali metals (such as potassium, sodium) and chlorine elements, which may form low melting point compounds under high temperature conditions, which may easily lead to slagging and corrosion problems on the boiler heating surface.

[0006] (2) Effect of blending ratio on combustion stability

[0007] When the biomass blending ratio is high, the temperature field and flow field in the coal-fired boiler will change significantly, which may cause the local temperature in the combustion chamber to be too high or too low, thereby affecting the combustion stability; too high a biomass blending ratio may cause the center position of the furnace flame to shift, reduce the thermal efficiency of the boiler, and even cause the risk of flameout.

[0008] (3) Complexity of boiler operating parameter optimization

[0009] Biomass blending requires dynamic adjustment of key boiler operating parameters (such as secondary air volume distribution, fuel delivery rate, furnace negative pressure, etc.). Traditional experience-based operation methods are difficult to adapt to complex operating conditions in real time; during the blending process, the mixing ratio and combustion characteristics of different biomass fuels need to be flexibly adjusted. The existing control system lacks sufficient predictive capabilities and is difficult to achieve operation optimization within the full range of operating conditions.

[0010] (4) Conflict between pollutant emission characteristics and boiler design

[0011] The combustion of biomass fuels may produce high concentrations of particulate matter (PM2.5), chlorides and alkali metal salts. These substances are easy to adhere to the boiler tube wall to form ash, which increases the difficulty of cleaning. The NO generated when biomass is mixed with combustion x Emissions are closely related to combustion temperature and oxygen content. Optimizing the blending conditions to control pollutant emissions requires complex multi-physics coupling analysis.

[0012] (5) Lack of technology for model prediction and dynamic optimization

[0013] In existing research, most analyses are based on experimental studies or static model simulations, which cannot reflect the dynamic changes caused by the combustion of biomass during boiler operation; there is a lack of predictive tools suitable for real-time adjustment of boiler operating conditions, especially in the processing and analysis of multivariate, high-dimensional data. The current technical means are insufficient and cannot provide sufficient accuracy and dynamics.

[0014] In addition, traditional boiler design and operating parameters are often optimized for pure coal combustion, without fully considering the complex characteristics of biomass fuel. After the biomass is blended, the heat transfer, combustion reaction rate and emission characteristics inside the boiler will change. This change not only increases the technical difficulty of operation, but also puts higher requirements on the stability and life of the boiler system. In the face of the above problems, the practical application of the technology of blending biomass in coal-fired boilers urgently needs an optimization method that integrates numerical simulation and dynamic prediction, which can achieve dynamic optimization of the blending ratio and operating parameters while ensuring combustion stability and emission compliance. However, existing research mainly focuses on the impact analysis of a single variable, lacking an overall solution under multi-variable and multi-operating conditions. This has become the main bottleneck restricting the development of biomass blending technology in coal-fired boilers. Summary of the invention

[0015] The purpose of the present invention is to provide a small target detection method based on an unmanned aerial vehicle, an electronic device and a storage medium to solve the problems raised in the above background technology.

[0016] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0017] A method for intelligently optimizing the biomass blending combustion conditions of a coal-fired boiler, the method comprising:

[0018] S100, construct a three-dimensional mathematical model of coal-fired boilers burning biomass based on experimental test data;

[0019] S200, perform multi-condition numerical simulation based on the boiler combustion condition fluctuations in the three-dimensional mathematical model;

[0020] S300. Construct a spatiotemporal coupling model of biomass combustion in coal-fired boilers based on numerical simulation data.

[0021] Preferably, S100 comprises:

[0022] S101. According to the engineering drawings and on-site survey data, collect combustion-related parameters, including furnace assembly structure, combustion device layout and fuel feed characteristic parameters, accurately restore the coal-fired boiler in three dimensions, and build a preliminary test model;

[0023] S102. Calibrate the preliminary model using experimental measurement results to ensure that its accuracy in combustion temperature, flow field and heat transfer characteristics is less than 10%.

[0024] Preferably, S200 includes:

[0025] S201, dividing the calorific value of biomass into two groups, low calorific value and high calorific value, combining the coal feed amount of the coal-fired boiler to change the blending ratio from 10% to 60% to form multiple groups of working conditions, and conducting multi-condition numerical simulation;

[0026] S202. Obtain the combustion characteristics of coal-fired boilers mixed with different types and amounts of biomass under different operating conditions, analyze the heating surface and furnace temperature, and flue gas composition, and extract and save information data based on the cross section and each node on the cross section as characteristics.

[0027] Preferably, S300 includes:

[0028] S301, building based on CFD simulation data, standardizing CFD simulation results, and generating a training set for deep learning, including: input variables and output targets;

[0029] S302, using the graph convolutional network GCN to extract the characteristics of different areas of the boiler combustion process, capturing the spatial distribution characteristics of the combustion process, and using the bidirectional long short-term memory network to extract the temperature distribution, pressure, and distribution changes of the flue gas outlet;

[0030] S303, using an adaptive learning rate optimization algorithm to improve model accuracy by minimizing prediction errors.

[0031] The optimization method, S301, comprises:

[0032] S301-1. Extract key data sets of different cross sections of coal-fired boilers mixed with biomass from CFD simulations, including:

[0033] Temperature field T(x,y,z,t);

[0034] Flow field velocity v(x,y,z,t) pollutant;

[0035] Biomass blending ratio r and calorific value Q;

[0036] S301-2. Discretize CFD data into a space-time grid and construct a feature matrix X∈R of different cross-section nodesN×F , where N is the number of grid nodes and F is the feature dimension; construct the spatial topological relationship A∈R N×N , reflecting the spatial adjacency between grid nodes.

[0037] Preferably, S302 includes:

[0038] S302-1. Spatial feature extraction based on GCN, H (l+1) =σ(AH (l) W (l) );

[0039] Among them, H (l) is the node feature matrix of the lth layer, H (0) =X represents the characteristic matrix X of the initial test section state,

[0040] A=D -1 / 2 A'D -1 / 2 represents the symmetric normalized adjacency matrix, D is the degree matrix, which reflects the adjacency pointing relationship between grid nodes in different sections; W (l) represents the weight matrix of the lth layer grid, σ represents the activation function, which is used to stimulate the training of grid data;

[0041] S302-2, based on the time dynamic characteristics extraction of Bi-LSTM, its dynamic behavior is described by the following formula, the gate structure and unit under the LSTM unit state update:

[0042] A1. Forget gate: f t =σ(W f ·[h t-1 ,x t ]+b f );

[0043] A2. Input gate:

[0044] i t =σ(W i ·[h t-1 ,x t ]+b i );

[0045] C' t =tanh(W C ·[h t-1 ,x t ]+b C );

[0046] A3. Memory unit update: C t =f t ⊙C t-1 +i t ⊙C' t ;

[0047] A4, output gate:

[0048] o t =σ(W o ·[h t-1 ,x t ]+b o );

[0049] h t =o t ⊙tanh(C t );

[0050] Among them, σ(*) represents the sigmoid function, W f , W i , W C and W o Both represent the weight matrix of the gate structure, [h t-1 ,x t ] represents the hidden state h at the previous moment t-1 and the current input x t connection, b f , b i , b C , b o Both represent the bias vector of the gate structure, C' t represents candidate memory cells, h t represents the hidden state of the output gate;

[0051] S302-3, bidirectional structure, the bidirectional memory network forms two hidden states, the front and the back, for information selection. In the optimization of coal-fired boiler operating conditions, the future change trend of boiler parameters may have important guiding significance for current predictions. In CFD simulation, the furnace temperature at certain moments is not only related to the previous combustion process, but may also be affected by future changes in fuel flow rate. The bidirectional structure can comprehensively consider the information in the upper and lower time periods and more accurately construct complex time dynamic characteristics.

[0052] Preferably, S303 includes:

[0053] The spatial feature H extracted by GCN space The temporal feature H extracted by Bi-LSTM time Fusion, through the fully connected layer to generate the prediction output: H output =FC(concat(H space ,H time ));

[0054] Among them, the output values ​​include thermal efficiency η, combustion stability index and pollutant emission index. Finally, the optimization objective function is obtained in the form of product and addition of the sum of the three outputs, and the boiler operation parameters are predicted and optimized based on the prediction results.

[0055] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the above-mentioned method for intelligently optimizing the biomass blending combustion conditions of a coal-fired boiler.

[0056] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the above-mentioned method for intelligently optimizing the biomass blending combustion conditions of a coal-fired boiler are implemented.

[0057] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0058] The present invention constructs a three-dimensional mathematical model of biomass blending in a coal-fired boiler; uses the model to perform multi-operating condition CFD simulations to analyze the effects of different biomass calorific values ​​and blending ratios on boiler combustion stability, thermal efficiency and emission characteristics; constructs a spatiotemporal characteristic model of biomass blending based on numerical simulation results, and optimizes the operating conditions using the spatiotemporal coupling characteristics of the model; the method improves modeling accuracy through numerical simulation technology, dynamically optimizes boiler operating conditions in combination with spatiotemporal modeling, and provides theoretical support for achieving energy conservation and emission reduction and blending ratio optimization; the present invention is suitable for environmental protection transformation and operation optimization of coal-fired boilers, and can effectively reduce carbon emissions and pollutant emissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0060] Figure 1 It is a flow chart of an intelligent optimization method for biomass blending conditions in a coal-fired boiler according to the present invention;

[0061] Figure 2 It is the construction of three-dimensional mathematical model and CFD simulation flow chart of biomass combustion in coal-fired boiler;

[0062] Figure 3 It is a flow chart of the spatiotemporal coupling model built on the basis of graph convolutional neural network and bidirectional long short-term memory network. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] See also Figure 1-Figure 3 , the present invention provides a technical solution:

[0065] Embodiment 1:

[0066] A method for intelligently optimizing the biomass blending combustion conditions of a coal-fired boiler, the method comprising:

[0067] S100, construct a three-dimensional mathematical model of coal-fired boilers burning biomass based on experimental test data;

[0068] The three-dimensional mathematical model is constructed based on engineering drawings and on-site survey data, and the calibration between historical data and simulation data is achieved through CFD simulation method;

[0069] Preferably, S100 comprises:

[0070] S101. According to the engineering drawings and on-site survey data, collect combustion-related parameters, including furnace assembly structure, combustion device layout and fuel feed characteristic parameters, accurately restore the coal-fired boiler in three dimensions, and build a preliminary test model;

[0071] S102. Calibrate the preliminary model using experimental measurement results to ensure that its accuracy in combustion temperature, flow field and heat transfer characteristics is less than 10%.

[0072] S200, perform multi-condition numerical simulation based on the boiler combustion condition fluctuations in the three-dimensional mathematical model;

[0073] The three-dimensional mathematical model multi-condition numerical simulation is carried out based on the historical operating characteristics of coal-fired boilers and changes in biomass characteristics, and CFD simulation data is converted into a data matrix;

[0074] Preferably, S200 includes:

[0075] S201. The calorific value of biomass is divided into two groups: low calorific value (straw, rice husk, etc.) and high calorific value (wood chips, fruit shells, etc.). Combined with the coal feed of the coal-fired boiler, the blending ratio is changed from 10% to 60% to form multiple groups of working conditions, and multi-condition numerical simulation is performed;

[0076] S202. Obtain the combustion characteristics of coal-fired boilers mixed with different types and amounts of biomass under different operating conditions, analyze the heating surface and furnace temperature, and flue gas composition, and extract and save information data based on the cross section and each node on the cross section as characteristics.

[0077] Preferably, Figure 2 As shown in the figure, the construction of the three-dimensional mathematical model is based on the combination of coal-fired boiler drawings and field survey data. Detailed optimization is carried out for the problems encountered in software modeling. After the model is built, it is calibrated with historical data. By changing the type of biomass (high and low calorific values), adjusting the combustion conditions of the coal-fired boiler, and changing the proportion of coal and biomass, CFD simulation is carried out under multiple conditions. The mathematical model is constructed as follows: the discretization of the differential equation uses the finite volume method, the turbulence model uses the standard k-ε two-equation to simulate turbulent flow, the chemical reaction model uses the finite rate model to simulate the reaction characteristics of the mixed combustion of coal and biomass, and the thermal radiation model uses the P1 radiation model to evaluate the heat distribution in the furnace. The simulation conditions cover the combination of different blending ratios (such as 10%-60%, divided into 10% steps) and biomass calorific values, analyze the temperature field, flow field and pollutant generation characteristics, and save the numerical simulation data;

[0078] Preferably, the algorithm process of S100 and S200 is:

[0079] Obtain the coal-fired boiler drawings and on-site survey data, optimize the details of the problems encountered in the software modeling, calibrate the model with historical data after the model is built, and perform CFD simulation under multiple conditions by changing the type of biomass, adjusting the combustion conditions of the coal-fired boiler, and changing the ratio of coal and biomass. The mathematical model is constructed as follows:

[0080] Analyze experimental test data using differential equations, turbulence models, chemical reaction models, and thermal radiation models;

[0081] The differential equation is discretized using the finite volume method;

[0082] The turbulence model uses the standard k-ε dual equation to simulate turbulent flow;

[0083] The chemical reaction model uses a finite rate model to simulate the reaction characteristics of the mixed combustion of coal and biomass;

[0084] The thermal radiation model uses the P1 radiation model to evaluate the heat distribution in the furnace. The simulation conditions cover the combination of different blending ratios and biomass calorific values, analyze the temperature field, flow field and pollutant generation characteristics, and save the numerical simulation data.

[0085] S300, constructing a spatiotemporal coupling model of biomass combustion in coal-fired boilers based on numerical simulation data;

[0086] The time-space coupling model is constructed based on CFD simulation data, and extracts important information such as temperature, pressure, and components of each important section and node of the three-dimensional mathematical model in space to form a directed adjacency matrix; in time, it extracts the transfer relationship between nodes in each time transfer direction. Finally, the time-space coupling model can replace the CFD simulation process to predict and optimize the working conditions of coal-fired boilers burning biomass.

[0087] Preferably, S300 includes:

[0088] S301, based on CFD simulation data, standardize the CFD simulation results and generate a training set for deep learning, including: input variables (such as biomass calorific value, blending ratio, etc.) and output targets (such as boiler efficiency, furnace temperature distribution uniformity, etc.);

[0089] S302, using the graph convolutional network GCN to extract the characteristics of different areas of the boiler combustion process, capturing the spatial distribution characteristics of the combustion process, and using the bidirectional long short-term memory network to extract the temperature distribution, pressure, and distribution changes of the flue gas outlet;

[0090] S303, using an adaptive learning rate optimization algorithm to improve model accuracy by minimizing prediction errors.

[0091] Preferably, Figure 3 As shown, the algorithm of S300 includes:

[0092] S301-1. Extract key data sets of different cross sections of coal-fired boilers mixed with biomass from CFD simulations, including:

[0093] Temperature field T(x,y,z,t);

[0094] Flow field velocity v(x,y,z,t) pollutant;

[0095] Biomass blending ratio r and calorific value Q;

[0096] S301-2. Discretize CFD data into a space-time grid and construct a feature matrix X∈R of different cross-section nodes N×F , where N is the number of grid nodes and F is the feature dimension; construct the spatial topological relationship A∈R N×N , reflects the spatial adjacency between grid nodes;

[0097] S302-1. Spatial feature extraction based on GCN, H (l+1) =σ(AH (l) W (l) );

[0098] Among them, H (l) is the node feature matrix of the lth layer, H (0)=X represents the characteristic matrix X of the initial test section state,

[0099] A=D -1 / 2 A'D -1 / 2 represents the symmetric normalized adjacency matrix, D is the degree matrix, which reflects the adjacency pointing relationship between grid nodes in different sections; W (l) represents the weight matrix of the lth layer grid, σ represents the activation function, which is used to stimulate the training of grid data;

[0100] S302-2, based on the time dynamic characteristics extraction of Bi-LSTM, its dynamic behavior is described by the following formula, the gate structure and unit under the LSTM unit state update:

[0101] A1. Forget gate: f t =σ(W f ·[h t-1 ,x t ]+b f );

[0102] A2. Input gate:

[0103] i t =σ(W i ·[h t-1 ,x t ]+b i );

[0104] C' t =tan h(W C ·[h t-1 ,x t ]+b C );

[0105] A3. Memory unit update: C t =f t ⊙C t-1 +i t ⊙C' t ;

[0106] A4, output gate:

[0107] o t =σ(W o ·[h t-1 ,x t ]+b o );

[0108] h t =o t ⊙tanh(C t );

[0109] Among them, σ(*) represents the sigmoid function, W f , Wi , W C and W o Both represent the weight matrix of the gate structure, [h t-1 ,x t ] represents the hidden state h at the previous moment t-1 and the current input x t connection, b f , b i , b C , b o Both represent the bias vector of the gate structure, C' t represents candidate memory cells, h t represents the hidden state of the output gate;

[0110] S302-3, bidirectional structure, the bidirectional memory network forms two hidden states, front and back, for information selection. In the optimization of coal-fired boiler operating conditions, the future change trend of boiler parameters may have important guiding significance for current predictions. In CFD simulation, the furnace temperature at certain moments is not only related to the previous combustion process, but may also be affected by future changes in fuel flow rate. The bidirectional structure can comprehensively consider the information in the upper and lower time periods and more accurately construct complex time dynamic characteristics;

[0111] The spatial feature H extracted by GCN space The temporal feature H extracted by Bi-LSTM time Fusion, through the fully connected layer to generate the prediction output: H output =FC(concat(H space ,H time ));

[0112] Among them, the output values ​​include thermal efficiency η, combustion stability index (temperature distribution uniformity, etc.), pollutant emission index, etc., and finally the optimization objective function is obtained in the form of the product and addition of the sum of the three outputs, and the boiler operation parameters (secondary air distribution, biomass blending ratio, fuel delivery rate, etc.) are predicted and optimized based on the prediction results.

[0113] Embodiment 2:

[0114] The computer-readable storage medium of this embodiment stores a computer program thereon, which, when executed by a processor, implements the steps of the method for intelligently optimizing the biomass blending combustion conditions of a coal-fired boiler in Example 1.

[0115] The computer-readable storage medium of this embodiment may be an internal storage unit of the terminal, such as a hard disk or memory of the terminal; the computer-readable storage medium of this embodiment may also be an external storage device of the terminal, such as a plug-in hard disk, a smart memory card, a secure digital card, a flash memory card, etc. equipped on the terminal; further, the computer-readable storage medium may also include both an internal storage unit of the terminal and an external storage device.

[0116] The computer-readable storage medium of this embodiment is used to store computer programs and other programs and data required by the terminal. The computer-readable storage medium can also be used to temporarily store data that has been output or is to be output.

[0117] Embodiment 3:

[0118] The computer device of this embodiment includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps of the method for intelligently optimizing the biomass blending conditions of a coal-fired boiler in Embodiment 1 are implemented.

[0119] In this embodiment, the processor may be a central processing unit, or other general-purpose processors, digital signal processors, application-specific integrated circuits, readily available programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory. For example, the memory may also store information about the device type.

[0120] Those skilled in the art will appreciate that the disclosed content of the embodiments may be provided as methods, systems, or computer program products. Therefore, the present solution may adopt the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Moreover, the present solution may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program codes.

[0121] The present solution is described with reference to the method according to the embodiment of the present solution and the flowchart and / or block diagram of the computer program product. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions; these computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 one or more processes and / or methods Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0122] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 one or more processes and / or methods Figure 1 A function specified in one or more boxes.

[0123] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 one or more processes and / or methods Figure 1 The steps for the functions specified in one or more boxes.

[0124] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).

[0125] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent optimization method for biomass blending in coal-fired boilers, characterized in that: The method comprises: S100, construct a three-dimensional mathematical model of coal-fired boilers burning biomass based on experimental test data; S200, perform multi-condition numerical simulation based on the boiler combustion condition fluctuations in the three-dimensional mathematical model; S300. Construct a spatiotemporal coupling model of biomass combustion in coal-fired boilers based on numerical simulation data.

2. The method for intelligent optimization of biomass blending conditions in a coal-fired boiler according to claim 1, characterized in that: The S100 includes: S101. According to the engineering drawings and on-site survey data, collect combustion-related parameters, including furnace assembly structure, combustion device layout and fuel feed characteristic parameters, accurately restore the coal-fired boiler in three dimensions, and build a preliminary test model; S102. Calibrate the preliminary model using experimental measurement results to ensure that its accuracy in combustion temperature, flow field and heat transfer characteristics is less than 10%.

3. The method for intelligent optimization of biomass blending conditions in a coal-fired boiler according to claim 1, characterized in that: The S200 includes: S201, dividing the calorific value of biomass into two groups, low calorific value and high calorific value, combining the coal feed amount of the coal-fired boiler to change the blending ratio from 10% to 60% to form multiple groups of working conditions, and conducting multi-condition numerical simulation; S202. Obtain the combustion characteristics of coal-fired boilers mixed with different types and amounts of biomass under different operating conditions, analyze the heating surface and furnace temperature, and flue gas composition, and extract and save information data based on the cross section and each node on the cross section as characteristics.

4. The method for intelligent optimization of biomass blending conditions in a coal-fired boiler according to claim 1, characterized in that: The S300 includes: S301, building based on CFD simulation data, standardizing CFD simulation results, and generating a training set for deep learning, including: input variables and output targets; S302, using the graph convolutional network GCN to extract the characteristics of different areas of the boiler combustion process, capturing the spatial distribution characteristics of the combustion process, and using the bidirectional long short-term memory network to extract the temperature distribution, pressure, and distribution changes of the flue gas outlet; S303, using an adaptive learning rate optimization algorithm to improve model accuracy by minimizing prediction errors.

5. The method for intelligent optimization of biomass blending conditions in a coal-fired boiler according to claim 4, characterized in that: The S301 includes: S301-1. Extract key data sets of different cross sections of coal-fired boilers mixed with biomass from CFD simulations, including: Temperature field T(x,y,z,t); Flow field velocity v(x,y,z,t) pollutant; Biomass blending ratio r and calorific value Q; S301-2. Discretize CFD data into a space-time grid and construct a feature matrix X∈R of different cross-section nodes N×F , where N is the number of grid nodes and F is the feature dimension; construct the spatial topological relationship A∈R N×N , reflecting the spatial adjacency between grid nodes.

6. The method for intelligent optimization of biomass blending conditions in a coal-fired boiler according to claim 4, characterized in that: The S302 includes: S302-1. Spatial feature extraction based on GCN, H (l+1) =σ(AH (l) W (l) ); Among them, H (l) is the node feature matrix of the lth layer, H (0) =X represents the characteristic matrix X of the initial test section state, A=D -1 / 2 A'D -1 / 2 represents the symmetric normalized adjacency matrix, D is the degree matrix, which reflects the adjacency pointing relationship between grid nodes in different sections; W (l) represents the weight matrix of the lth layer grid, σ represents the activation function, which is used to stimulate the training of grid data; S302-2, based on the time dynamic characteristics extraction of Bi-LSTM, its dynamic behavior is described by the following formula, the gate structure and unit under the LSTM unit state update: A1. Forget gate: f t =σ(W f ·[h t-1 ,x t ]+b f ); A2. Input gate: i t =σ(W i ·[h t-1 ,x t ]+b i ); C' t =tanh(W C ·[h t-1 ,x t ]+b C ); A3. Memory unit update: C t =f t ⊙C t-1 +i t ⊙C' t ; A4, output gate: i t =σ(W o ·[h t-1 ,x t ]+b o ); h t =o t ⊙tanh(C t ); Among them, σ(*) represents the sigmoid function, W f , W i , W C and W o Both represent the weight matrix of the gate structure, [h t-1 ,x t ] represents the hidden state h at the previous moment t-1 and the current input x t connection, b f , b i , b C , b o Both represent the bias vector of the gate structure, C' t represents candidate memory cells, h t represents the hidden state of the output gate; S302-3, bidirectional structure, the bidirectional memory network forms two hidden states, the front and the back, for information selection. In the optimization of coal-fired boiler operating conditions, the future change trend of boiler parameters may have important guiding significance for current predictions. In CFD simulation, the furnace temperature at certain moments is not only related to the previous combustion process, but may also be affected by future changes in fuel flow rate. The bidirectional structure can comprehensively consider the information in the upper and lower time periods and more accurately construct complex time dynamic characteristics.

7. The method for intelligent optimization of biomass blending conditions in a coal-fired boiler according to claim 4, characterized in that: The S303 includes: The spatial feature H extracted by GCN space The temporal feature H extracted by Bi-LSTM time Fusion, through the fully connected layer to generate the prediction output: H output =FC(concat(H space ,H time )); Among them, the output values ​​include thermal efficiency η, combustion stability index and pollutant emission index. Finally, the optimization objective function is obtained in the form of product and addition of the sum of the three outputs, and the boiler operation parameters are predicted and optimized based on the prediction results.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method for intelligently optimizing the biomass blending combustion conditions of a coal-fired boiler as described in any one of claims 1 to 7 are implemented.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for intelligent optimization of biomass blending conditions in a coal-fired boiler as described in any one of claims 1-8 are implemented.

Images