On-line calculation method and system for corrosion state in furnace based on simulation data

Abstract

The invention discloses an in-furnace corrosion state online calculation method and system based on simulation data, and relates to the field of boiler combustion operation control. The problems that in the prior art, sulfide in coal and the pipe wall temperature are not considered in high-temperature corrosion, high-temperature corrosion in some areas is likely to be intensified, and then the safety and reliability of equipment are affected are solved. The method comprises the following steps: constructing a boiler typical operation database according to boiler operation data; establishing a sulfide reaction mechanism model; calculating sulfide concentration distribution conditions and thermal load data under all working conditions according to the sulfide reaction mechanism model; dividing hydraulic calculation loops, and calculating hydraulic distribution and metal wall temperature distribution of each loop according to structural resistance and thermal load data; calculating the H2S concentration of the wall surface of the water cooling wall; and calculating the sulfide concentration and the metal wall temperature of the near wall surface in the furnace by using the neural network model to obtain the corrosion state in the furnace. The method can adapt to boilers and fuels of different types, and has universality and flexibility.

Description

Technical Field

[0001] The invention relates to boiler combustion operation control, and in particular to an online calculation method for corrosion state in a furnace based on simulation data. Background Art

[0002] In the process of monitoring the wall temperature and corrosion during boiler operation, the actual measurement points are relatively few, and the high-temperature area cannot be measured. The high-temperature corrosion problem has always been one of the major hidden dangers affecting the safe operation of power plants. The occurrence of high-temperature corrosion mainly refers to the chemical reaction between the volatile matter and non-silicate impurities in the sediment during the combustion of coal powder and the iron in the metal pipe wall under high-temperature environment to cause slagging. This phenomenon is particularly obvious on the water-cooled wall and superheater of the boiler burning high-sulfur coal. Patent CN116776770A discloses "a method for alleviating high-temperature corrosion of boilers based on CFD numerical simulation coupled with BP neural network", which establishes a method for alleviating high-temperature corrosion by adjusting the opening of the secondary air door according to the distribution of flue gas temperature and CO concentration. However, high-temperature corrosion is not only related to CO concentration and high flue gas temperature, but also to sulfides in coal and pipe wall temperature.

[0003] Relying solely on monitoring flue gas temperature and CO concentration may not effectively identify the direct corrosion effect of sulfides produced during the combustion of high-sulfur coal on the pipe wall. This will lead to increased high-temperature corrosion in certain areas, forming localized corrosion that is difficult to detect, which in turn affects the safety and reliability of the equipment. At the same time, ignoring the presence of sulfides in coal will reduce the accuracy of the model, leading to deviations in the prediction of corrosion conditions, which in turn affects the effectiveness of preventive measures. For example, higher corrosion rates may occur in actual operation, but the model fails to reflect this change, making it impossible to take necessary intervention measures in advance.

[0004] High-temperature corrosion is particularly evident in the water-cooled wall and superheater areas. If changes in tube wall temperature are not taken into account, the life of these key components may be shortened, the frequency of shutdowns for maintenance may increase, and the economic benefits of the power plant may be affected. Summary of the invention

[0005] The present invention aims at the problem that high temperature corrosion in the prior art does not take into account the sulfide in the coal and the temperature of the pipe wall, which easily leads to the aggravation of high temperature corrosion in certain areas, forming local corrosion that is not easy to be discovered, and further affecting the safety and reliability of the equipment. An online calculation method for the corrosion state in the furnace based on simulation data is proposed. The method includes:

[0006] Step 1: Building a typical boiler operation database based on boiler operation data;

[0007] Step 2: Establish a sulfide reaction mechanism model based on boiler boundary conditions and the relationship between sulfide and heat load;

[0008] Step 3, calculating the sulfide concentration distribution and heat load data of the whole working condition according to the sulfide reaction mechanism model;

[0009] Step 4: Divide the hydraulic calculation loops according to the boiler structure, and calculate the hydraulic distribution and metal wall temperature distribution of each loop according to the structural resistance and heat load data;

[0010] Step 5, calculate the H2S concentration on the water-cooled wall surface;

[0011] Step 6: Use the neural network model to calculate the sulfide concentration near the wall and the metal wall temperature in the furnace to obtain the corrosion status in the furnace.

[0012] Furthermore, a preferred method is proposed, in which the boiler operation data in step 1 includes: coal quality data, mill operation mode, metal material, and corrosion status historical data.

[0013] Furthermore, a preferred method is proposed, wherein step 3 comprises:

[0014] Grid the computational area;

[0015] Input reactant concentration, temperature, and pressure;

[0016] The reaction equations were calculated using a numerical solver to obtain the distribution of sulfide concentration over time and space;

[0017] Calculate the heat load of the system based on reaction heat and mass transfer.

[0018] Furthermore, a preferred method is proposed, wherein the calculation of the hydraulic distribution and metal wall temperature distribution of each circuit in step 4 includes:

[0019] Q=c*m*△t

[0020] t=t0+△t

[0021] Among them, Q is the total heat, Q=q*A, q is the heat load data, A is the loop area, C is the constant pressure specific heat of the working fluid, m is the working fluid flow rate; t0 is the known inlet temperature, △t is the temperature rise, and t is the temperature of the calculation point.

[0022] Furthermore, a preferred embodiment is proposed, wherein step 5 comprises:

[0023] Obtain the water wall structure, fuel characteristics and reaction conditions, including temperature and pressure;

[0024] Determine the coal quantity, mill operation mode, air door opening and coal characteristic parameters;

[0025] Use CFD software to build a three-dimensional geometric model of the water-cooled wall;

[0026] Select appropriate physical models, including fluid dynamics models, heat transfer models, and chemical reaction models;

[0027] Input initial conditions and boundary conditions, wherein the initial conditions include temperature, pressure, and concentration, and the boundary conditions include air inlet and air outlet conditions;

[0028] Run CFD simulation to obtain three-dimensional data of H2S concentration distribution on the wall.

[0029] Furthermore, a preferred embodiment is proposed, wherein step 6 comprises:

[0030] Extract wall H2S concentration data from CFD results, integrate the extracted CFD results and input conditions into a data set, and use the constructed data set to train the limit vector machine neural network model;

[0031] The average prediction ability of the EVM neural network model was measured using correlation coefficient, mean absolute error, root mean square error and mean relative error;

[0032] The sulfide concentration near the wall and the metal wall temperature in the furnace are calculated through the trained limit vector machine neural network model to determine the corrosion state in the furnace.

[0033] Furthermore, a preferred method is proposed, wherein the average prediction ability of the extreme vector machine neural network model is measured by using the correlation coefficient, the mean absolute error, the root mean square error and the mean relative error, including:

[0034]

[0035]

[0036]

[0037]

[0038] Among them, R(X,Y) is the correlation coefficient, MAE is the mean absolute error, RMSE is the root mean square error, MRE is the mean relative error, X, Y represent two variables, N represents the number of samples, y t represents the tth actual value, represents the tth predicted value.

[0039] Based on the same inventive concept, the present invention also proposes an online calculation system for the corrosion state in a furnace based on simulation data, the system comprising:

[0040] A database construction unit, used for constructing a typical boiler operation database according to boiler operation data;

[0041] The mechanism model building unit is used to establish the sulfide reaction mechanism model according to the boiler boundary conditions and the relationship between sulfide and heat load;

[0042] A calculation unit is used to calculate the sulfide concentration distribution and heat load data of all working conditions according to the sulfide reaction mechanism model;

[0043] The hydraulic and metal wall temperature distribution acquisition unit is used to divide the hydraulic calculation loop according to the boiler structure, and calculate the hydraulic distribution and metal wall temperature distribution of each loop according to the structural resistance and heat load data;

[0044] H2S concentration calculation unit, used to calculate H2S concentration on the water wall surface;

[0045] The furnace corrosion state acquisition unit is used to calculate the sulfide concentration near the wall and the metal wall temperature in the furnace by using a neural network model to obtain the corrosion state in the furnace.

[0046] Based on the same inventive concept, the present invention also proposes a computer device, including a memory and a processor, wherein the memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the online calculation method for the corrosion state in the furnace based on simulation data as described in any one of the above items.

[0047] Based on the same inventive concept, the present invention also proposes a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the online calculation method of the corrosion state in the furnace based on simulation data as described in any one of the above are executed.

[0048] The benefits of the present invention are:

[0049] The present invention proposes an online calculation method for the corrosion state in a furnace based on simulation data, which aims to calculate the high-temperature corrosion problem and avoids the limitation of the existing method that does not consider the influence of sulfide in coal and pipe wall temperature on corrosion.

[0050] First, the operation data of the boiler under different operating conditions are collected to form a database. These data provide a basis for the establishment of subsequent models. By analyzing the boundary conditions of the boiler (such as temperature, pressure, etc.) and the relationship between sulfide and heat load, the corresponding chemical reaction mechanism model is established. This makes it possible to more accurately understand the behavior of sulfide in a high temperature environment. Using the established mechanism model, the sulfide concentration distribution of the boiler under different operating conditions and the corresponding heat load data are calculated. This step provides important parameters for subsequent hydraulic calculations. The boiler structure is divided into multiple hydraulic calculation loops, and the hydraulic distribution and metal wall temperature distribution are calculated according to the structural resistance and heat load data of each loop. This allows a more detailed understanding of the changes in the internal environment of the boiler. For the water-cooled wall, the H2S concentration on its surface is calculated. This data is an important indicator for evaluating high-temperature corrosion. Finally, the neural network model is used to combine the above data to calculate the sulfide concentration and metal wall temperature near the wall of the furnace, so as to obtain more comprehensive and real-time information on the corrosion status of the furnace.

[0051] The present invention proposes an online calculation method for the corrosion state in a furnace based on simulation data. Through three-dimensional online calculation of the atmosphere distribution and wall temperature of the entire area, the corrosion state of the boiler can be monitored in real time, potential corrosion problems can be discovered in time, and the safety of the equipment can be improved. Furthermore, the corrosion analysis of the coupled atmosphere field and temperature field in this embodiment combines multiple factors such as sulfide concentration and tube wall temperature, which can more accurately reflect the corrosion situation in the boiler and has higher accuracy than traditional methods. It can identify local high-temperature corrosion phenomena, so that targeted measures can be taken in advance to avoid equipment failures. By reducing the maintenance frequency and improving equipment reliability, maintenance costs can be reduced and the economic benefits of the power plant can be improved. This method can adapt to different types of boilers and fuels and has good versatility and flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a flow chart of the online calculation method of the corrosion state in the furnace based on simulation data described in Implementation Method 1. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0054] Implementation method 1, see Figure 1 The method for online calculation of corrosion state in a furnace based on simulation data described in this embodiment includes:

[0055] Step 1: Building a typical boiler operation database based on boiler operation data;

[0056] Step 2: Establish a sulfide reaction mechanism model based on boiler boundary conditions and the relationship between sulfide and heat load;

[0057] Step 3, calculating the sulfide concentration distribution and heat load data of the whole working condition according to the sulfide reaction mechanism model;

[0058] Step 4: Divide the hydraulic calculation loops according to the boiler structure, and calculate the hydraulic distribution and metal wall temperature distribution of each loop according to the structural resistance and heat load data;

[0059] Step 5, calculate the H2S concentration on the water-cooled wall surface;

[0060] Step 6: Use the neural network model to calculate the sulfide concentration near the wall and the metal wall temperature in the furnace to obtain the corrosion status in the furnace.

[0061] In practical applications, the calculated results of sulfide and metal wall temperature can be visualized through the program to show the corrosion situation, and the corrosion situation can be controlled by adjusting the air damper.

[0062] This embodiment proposes an online calculation method for the corrosion state in the furnace based on simulation data, which aims to calculate the high-temperature corrosion problem and avoids the limitation of the existing method that does not consider the influence of sulfide in coal and pipe wall temperature on corrosion.

[0063] First, the operation data of the boiler under different operating conditions are collected to form a database. These data provide the basis for the establishment of subsequent models. By analyzing the boundary conditions of the boiler (such as temperature, pressure, etc.) and the relationship between sulfide and heat load, the corresponding chemical reaction mechanism model is established. This makes it possible to more accurately understand the behavior of sulfide in a high temperature environment. Using the established mechanism model, the sulfide concentration distribution of the boiler under different operating conditions and the corresponding heat load data are calculated. This step provides important parameters for subsequent hydraulic calculations. The boiler structure is divided into multiple hydraulic calculation loops, and the hydraulic distribution and metal wall temperature distribution are calculated according to the structural resistance and heat load data of each loop. This allows a more detailed understanding of the changes in the internal environment of the boiler. For the water-cooled wall, the H2S concentration on its surface is calculated. This data is an important indicator for evaluating high-temperature corrosion. Finally, the neural network model is used to combine the above data to calculate the sulfide concentration and metal wall temperature near the wall of the furnace, so as to obtain more comprehensive and real-time information on the corrosion status of the furnace.

[0064] This embodiment proposes an online calculation method for the corrosion state in the furnace based on simulation data. Through online calculation, the corrosion state of the boiler can be monitored in real time, potential corrosion problems can be discovered in time, and the safety of the equipment can be improved. Furthermore, the corrosion analysis of the coupled atmosphere field and temperature field in this embodiment combines multiple factors such as sulfide concentration and tube wall temperature, which can more accurately reflect the corrosion situation in the boiler and has higher accuracy than traditional methods. It can identify local high-temperature corrosion phenomena, so that targeted measures can be taken in advance to avoid equipment failures. By reducing the maintenance frequency and improving equipment reliability, maintenance costs can be reduced and the economic benefits of the power plant can be improved. This method can adapt to different types of boilers and fuels and has good versatility and flexibility.

[0065] Implementation method 2: This implementation method further limits the online calculation method of the furnace corrosion state based on simulation data described in implementation method 1. The boiler operation data in step 1 includes: coal quality data, mill operation mode, metal material, and corrosion status historical data.

[0066] In practical applications, coal quality data includes information such as coal composition, sulfur content, ash content, etc., which is crucial for predicting the occurrence of high-temperature corrosion. Accurate coal quality data can better understand the concentration of sulfides and other corrosive components generated during coal combustion, thereby improving the accuracy of corrosion prediction.

[0067] The operation mode of the mill directly affects the fineness of the coal powder and the combustion efficiency. Different operation modes may lead to changes in the temperature distribution and gas flow inside the boiler, thereby affecting the degree of corrosion. By analyzing the operating data of different mill operation modes, the operation strategy of the boiler can be optimized and the corrosion risk can be reduced.

[0068] Embodiment 3: This embodiment further limits the online calculation method of the corrosion state in the furnace based on simulation data described in embodiment 1, and step 3 includes:

[0069] Grid the computational area;

[0070] Input reactant concentration, temperature, and pressure;

[0071] The reaction equations were calculated using a numerical solver to obtain the distribution of sulfide concentration over time and space;

[0072] Calculate the heat load of the system based on reaction heat and mass transfer.

[0073] In this embodiment, the sulfide concentration distribution is calculated by the sulfide reaction mechanism model, which can simulate the actual reaction process more accurately. The specific grid division and input parameters (such as reactant concentration, temperature, and pressure) make the calculation results closer to the actual working conditions. The use of time and space distribution calculations can monitor the corrosion state in the furnace in real time, detect potential corrosion problems in time, and improve the safety and reliability of the equipment. Through the calculation of reaction heat and material transfer, the heat load of the system can be effectively evaluated, which helps to optimize the operating conditions in the furnace and prevent equipment damage caused by excessive heat load.

[0074] Embodiment 4: This embodiment further limits the online calculation method of the corrosion state in the furnace based on simulation data described in Embodiment 1. The calculation of the hydraulic distribution and metal wall temperature distribution of each circuit in step 4 includes:

[0075] Q=c*m*△t

[0076] t=t0+△t

[0077] Among them, Q is the total heat, Q=q*A, q is the heat load data, A is the loop area, C is the constant pressure specific heat of the working fluid, m is the working fluid flow rate; t0 is the known inlet temperature, △t is the temperature rise, and t is the temperature of the calculation point.

[0078] The total heat passing through each circuit can be accurately calculated by the formula Q = c*m*△t, providing a reliable data basis for subsequent corrosion analysis. This accuracy is crucial to understanding the impact of heat on corrosion.

[0079] By calculating the temperature rise t=t0+△t, it is possible to dynamically predict the metal wall temperature at different locations, making it easier to identify possible high-temperature corrosion areas. This information helps to make timely adjustments in design and operation.

[0080] By calculating Q=q*A, the heat load distribution of each circuit can be effectively analyzed. This is of great significance to optimizing the operating conditions of the boiler and improving the thermal efficiency, and can reduce the risk of corrosion caused by local overheating.

[0081] Calculating the hydraulic distribution of each circuit helps to understand the flow of fluid in the system and ensure that the fluid is evenly distributed.

[0082] Embodiment 5: This embodiment further limits the online calculation method of the corrosion state in the furnace based on simulation data described in embodiment 1, and step 5 includes:

[0083] Obtain the water wall structure, fuel characteristics and reaction conditions, including temperature and pressure;

[0084] Determine the coal quantity, mill operation mode, air door opening and coal characteristic parameters;

[0085] Use CFD software to build a three-dimensional geometric model of the water-cooled wall;

[0086] Select appropriate physical models, including fluid dynamics models, heat transfer models, and chemical reaction models;

[0087] Input initial conditions and boundary conditions, wherein the initial conditions include temperature, pressure, and concentration, and the boundary conditions include air inlet and air outlet conditions;

[0088] Run CFD simulation to obtain three-dimensional data of H2S concentration distribution on the wall.

[0089] In this embodiment, by obtaining the structure, fuel characteristics and reaction conditions (temperature, pressure) of the water-cooled wall, comprehensive basic data is provided for the subsequent CFD simulation, ensuring the authenticity and reliability of the model. Determining the coal quantity, mill operation mode, air door opening and characteristic parameters of coal can deeply analyze the heat generation in the combustion process and its impact on the furnace environment. This helps to optimize the use of fuel, improve combustion efficiency and reduce energy consumption. Using CFD software to establish a three-dimensional geometric model of the water-cooled wall can accurately reflect the complex fluid flow and heat transfer process in the furnace. This three-dimensional modeling provides a basis for more detailed analysis and can reveal local corrosion hotspots. By selecting fluid dynamics models, heat transfer models and chemical reaction models, the dynamic process of fluid flow, heat transfer and chemical reactions can be comprehensively analyzed. This multi-physics field coupling analysis method improves the accuracy and reliability of the simulation. Clearly inputting initial conditions (temperature, pressure, concentration) and boundary conditions (inlet and outlet conditions) ensure the effectiveness of CFD simulation, so that the simulation results can more realistically reflect the actual operating environment.

[0090] Furthermore, the three-dimensional data of H2S concentration distribution on the wall surface obtained through CFD simulation can clearly show the distribution of corrosive media on the water-cooled wall surface, providing an important basis for corrosion monitoring and evaluation. This helps to timely identify high-risk areas and take corresponding preventive measures.

[0091] Embodiment 6: This embodiment further limits the online calculation method of the corrosion state in the furnace based on simulation data described in embodiment 1. Step 6 includes:

[0092] Extract wall H2S concentration data from CFD results as training and testing data;

[0093] The extracted CFD results and input conditions are integrated into a data set, and the constructed data set is used to train the limit vector machine (ELM) neural network model;

[0094] The average prediction ability of the EVM neural network model was measured using correlation coefficient, mean absolute error, root mean square error and mean relative error;

[0095] The sulfide concentration near the wall and the metal wall temperature in the furnace are calculated through the trained limit vector machine neural network model to determine the corrosion state in the furnace.

[0096] This embodiment can obtain more realistic and high-quality input by extracting H2S concentration data from CFD (computational fluid dynamics) results. This accurate data source helps to improve the prediction accuracy of the model and reduce errors. Integrating CFD results and input conditions into a data set helps to comprehensively analyze the operating environment in the furnace. This systematic data processing method can improve the training efficiency of the neural network model and ensure that the model can capture the changing characteristics under different conditions.

[0097] Furthermore, the extreme vector machine (ELM) is used as a neural network model to calculate the sulfide concentration near the wall and the metal wall temperature in the furnace because the extreme vector machine has a fast training speed and good generalization ability. ELM can effectively improve the generalization ability of the model by randomly selecting input weights and biases and then optimizing only the output weights. This method can better adapt to the changing furnace environment when dealing with complex nonlinear problems, thereby accurately predicting the sulfide concentration and metal wall temperature. Compared with traditional neural networks, ELM has a simpler training process and can quickly process large-scale data sets. The prediction ability of the model is evaluated by multiple indicators such as correlation coefficient, mean absolute error, root mean square error and mean relative error, providing a more comprehensive model performance evaluation. This multi-dimensional evaluation system can help developers quickly identify the advantages and disadvantages of the model and perform targeted optimization. Combined with the online calculation method, real-time monitoring of the corrosion state in the furnace can be achieved. Timely acquisition of sulfide concentration and metal wall temperature data helps to quickly respond to changes in furnace corrosion, and then formulate corresponding maintenance and operation strategies to reduce accident risks.

[0098] Embodiment 7: This embodiment further limits the online calculation method of the furnace corrosion state based on simulation data described in Embodiment 5. The average absolute error, root mean square error and average relative error measure the average prediction ability of the limit vector machine neural network model, including:

[0099]

[0100]

[0101]

[0102]

[0103] Among them, R(X,Y) is the correlation coefficient, MAE is the mean absolute error, RMSE is the root mean square error, MRE is the mean relative error, X, Y represent two variables, N represents the number of samples, y t represents the tth actual value, represents the tth predicted value.

[0104] In this embodiment, by introducing the three indicators of mean absolute error (MAE), root mean square error (RMSE) and mean relative error (MRE), the prediction performance of the model can be evaluated more comprehensively. These indicators provide a systematic analysis of the prediction results, so that the model can more accurately reflect the corrosion state in the furnace in practical applications. The introduction of the correlation coefficient makes it possible to quantify the linear relationship between the predicted value and the actual value, providing an important basis for the reliability of the model. High correlation indicates that the model can better capture the relationship between the data, thereby enhancing the credibility of the prediction results. This method emphasizes the importance of sample size. Through the analysis of a large sample size, the influence of random errors can be effectively reduced, the applicability and stability of the model under different operating conditions can be improved, and online monitoring can be made more effective. Combined with the characteristics of online calculation, real-time monitoring of the corrosion state in the furnace can be achieved, and potential safety hazards can be quickly responded to. Real-time feedback can provide timely decision support for operators, thereby improving overall safety and operating efficiency. By using a variety of error indicators in combination, the model performance can be evaluated not only from the perspective of absolute error, but also the influence of relative error can be considered. This multi-dimensional evaluation method enables the model to be flexibly adjusted and optimized in different application scenarios.

[0105] Embodiment 8: The online calculation system for corrosion state in a furnace based on simulation data described in this embodiment comprises:

[0106] A database construction unit, used for constructing a typical boiler operation database according to boiler operation data;

[0107] The mechanism model building unit is used to establish the sulfide reaction mechanism model according to the boiler boundary conditions and the relationship between sulfide and heat load;

[0108] A calculation unit is used to calculate the sulfide concentration distribution and heat load data of all working conditions according to the sulfide reaction mechanism model;

[0109] The hydraulic and metal wall temperature distribution acquisition unit is used to divide the hydraulic calculation loop according to the boiler structure, and calculate the hydraulic distribution and metal wall temperature distribution of each loop according to the structural resistance and heat load data;

[0110] H2S concentration calculation unit, used to calculate H2S concentration on the water wall surface;

[0111] The furnace corrosion state acquisition unit is used to calculate the sulfide concentration near the wall and the metal wall temperature in the furnace by using a neural network model to obtain the corrosion state in the furnace.

[0112] Embodiment 9. A computer device described in this embodiment includes a memory and a processor, wherein the memory stores a computer program. When the processor runs the computer program stored in the memory, the processor executes the online calculation method for the corrosion state in the furnace based on simulation data described in any one of embodiments 1 to 7.

[0113] Embodiment 10. A computer-readable storage medium described in this embodiment stores a computer program, and when the computer program is executed by a processor, the steps of the online calculation method of the corrosion state in the furnace based on simulation data as described in any one of embodiments 1 to 7 are executed.

[0114] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0115] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, devices (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram and the combination of 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 produce 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 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0116] 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 in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0117] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure rather than to limit its protection scope. Although the present disclosure has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that after reading the present disclosure, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the disclosed claims to be approved.

Claims

1. An online calculation method for the corrosion state in a furnace based on simulation data, characterized in that: The method comprises: Step 1: Building a typical boiler operation database based on boiler operation data; Step 2: Establish a sulfide reaction mechanism model based on boiler boundary conditions and the relationship between sulfide and heat load; Step 3, calculating the sulfide concentration distribution and heat load data of the whole working condition according to the sulfide reaction mechanism model; Step 4: Divide the hydraulic calculation loops according to the boiler structure, and calculate the hydraulic distribution and metal wall temperature distribution of each loop according to the structural resistance and heat load data; Step 5, calculate the H2S concentration on the water-cooled wall surface; Step 6: Use the neural network model to calculate the sulfide concentration near the wall and the metal wall temperature in the furnace to obtain the corrosion status in the furnace.

2. The method for online calculation of furnace corrosion state based on simulation data according to claim 1 is characterized in that: The boiler operation data in step 1 includes: coal quality data, mill operation mode, metal material, and corrosion status historical data.

3. The method for online calculation of corrosion state in a furnace based on simulation data according to claim 1 is characterized in that: The step 3 comprises: Grid the computational area; Input reactant concentration, temperature, and pressure; The reaction equations were calculated using a numerical solver to obtain the distribution of sulfide concentration over time and space; Calculate the heat load of the system based on reaction heat and mass transfer.

4. The method for online calculation of furnace corrosion state based on simulation data according to claim 1 is characterized in that: The calculation of the hydraulic distribution and metal wall temperature distribution of each circuit in step 4 includes: Q=c*m*△t t=t0+△t Among them, Q is the total heat, Q=q*A, q is the heat load data, A is the loop area, C is the constant pressure specific heat of the working fluid, m is the working fluid flow rate; t0 is the known inlet temperature, △t is the temperature rise, and t is the temperature of the calculation point.

5. The method for online calculation of corrosion state in a furnace based on simulation data according to claim 1 is characterized in that: The step 5 comprises: Obtain the water wall structure, fuel characteristics and reaction conditions, including temperature and pressure; Determine the coal quantity, mill operation mode, air door opening and coal characteristic parameters; Use CFD software to build a three-dimensional geometric model of the water-cooled wall; Select appropriate physical models, including fluid dynamics models, heat transfer models, and chemical reaction models; Input initial conditions and boundary conditions, wherein the initial conditions include temperature, pressure, and concentration, and the boundary conditions include air inlet and air outlet conditions; Run CFD simulation to obtain three-dimensional data of H2S concentration distribution on the wall.

6. The method for online calculation of corrosion state in a furnace based on simulation data according to claim 1 is characterized in that: The step 6 comprises: Extract wall H2S concentration data from CFD results, integrate the extracted CFD results and input conditions into a data set, and use the constructed data set to train the limit vector machine neural network model; The average prediction ability of the EVM neural network model was measured using correlation coefficient, mean absolute error, root mean square error and mean relative error; The sulfide concentration near the wall and the metal wall temperature in the furnace are calculated through the trained limit vector machine neural network model to determine the corrosion state in the furnace.

7. The method for online calculation of corrosion state in a furnace based on simulation data according to claim 6 is characterized in that: The correlation coefficient, mean absolute error, root mean square error and mean relative error are used to measure the average prediction ability of the extreme vector machine neural network model, including: Among them, R(X,Y) is the correlation coefficient, MAE is the mean absolute error, RMSE is the root mean square error, MRE is the mean relative error, X, Y represent two variables, N represents the number of samples, y t represents the tth actual value, represents the tth predicted value.

8. The online calculation system of furnace corrosion state based on simulation data is characterized by: The system comprises: A database construction unit, used for constructing a typical boiler operation database according to boiler operation data; The mechanism model building unit is used to establish the sulfide reaction mechanism model according to the boiler boundary conditions and the relationship between sulfide and heat load; A calculation unit is used to calculate the sulfide concentration distribution and heat load data of all working conditions according to the sulfide reaction mechanism model; The hydraulic and metal wall temperature distribution acquisition unit is used to divide the hydraulic calculation loop according to the boiler structure, and calculate the hydraulic distribution and metal wall temperature distribution of each loop according to the structural resistance and heat load data; H2S concentration calculation unit, used to calculate H2S concentration on the water wall surface; The furnace corrosion state acquisition unit is used to calculate the sulfide concentration near the wall and the metal wall temperature in the furnace by using a neural network model to obtain the corrosion state in the furnace.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the online calculation method for the corrosion state in the furnace based on simulation data according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, executes the steps of the method for online calculation of the corrosion state in a furnace based on simulation data as described in any one of claims 1 to 7.

Images