Method and system for rapidly predicting concentration of nitric oxide in hearth by using DT modeling based on numerical simulation
Through the DT modeling method based on numerical simulation, the nitrogen oxide concentration in the boiler furnace is quickly predicted, which solves the problems of insufficient accuracy and large calculation workload in the existing technology, and achieves efficient and accurate nitrogen oxide concentration prediction, which improves the The safety and environmental protection of the boiler.
Patent Information
- Application Number
- CN202411528032.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-30
- Publication Date
- 2025-02-21
AI Technical Summary
In the prior art, the measurement and calculation methods for nitrogen oxide concentration in the furnace have problems such as insufficient accuracy, poor convergence, large calculation workload, incomplete model, slow speed and low efficiency.
Using DT modeling method based on numerical simulation, by collecting real-time or offline data, establishing a physical model of the boiler furnace, performing simulation calculations and preprocessing data, and using DT modeling software to clean and machine learning prediction, realizing under different working conditions Rapid prediction of nitrogen oxide concentration.
It significantly improves the calculation efficiency and accuracy of nitrogen oxide concentration, saves the calculation time of numerical simulation, shortens from three days to one minute, meets the requirements of timely prediction, and effectively reduces the concentration of nitrogen oxides, and improves the The safety and environmental protection of the boiler.
Smart Images
Figure CN119495379A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of prediction of nitrogen oxide concentration in a furnace, and in particular to a method and system for rapidly predicting the nitrogen oxide concentration in a furnace by using DT modeling based on numerical simulation. Background Art
[0002] In order to maintain the safety and environmental protection of boiler operation, it is very important to effectively control and reduce the concentration of nitrogen oxides in the boiler. By monitoring the combustion conditions in the furnace, predicting the concentration of nitrogen oxides based on the oxygen content, temperature and time of the combustion process is a method to control nitrogen oxide emissions. At present, the traditional measurement and calculation method is to use CFD (Computational Fluid Dynamics) numerical simulation technology to simulate fluid flow, heat transfer, combustion and other processes, but it has the problems of insufficient accuracy, poor convergence, large calculation workload, incomplete model, slow speed and low efficiency. For industrial applications of boilers, new methods need to be proposed to quickly and accurately predict the concentration of nitrogen oxides in the furnace. Summary of the invention
[0003] The present invention aims to solve the problems in the prior art of the method for measuring and calculating the nitrogen oxide concentration in the furnace, such as insufficient accuracy, poor convergence, large calculation workload, incomplete model, slow speed and low efficiency.
[0004] To solve the above technical problems, the present invention is achieved through the following technical solutions:
[0005] Solution 1: The present invention proposes a method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation, the method comprising the following steps:
[0006] S1. Collect real-time or offline data and use numerical simulation software to establish a physical model of the boiler furnace;
[0007] S2, performing simulation calculation on the physical model of the boiler furnace established in S1, and preprocessing and pre-analyzing the obtained simulation data;
[0008] S3. Import the preprocessed data set of the simulation data in S2 as input data into the DT modeling software for data cleaning, perform machine learning prediction on the cleaned data, and realize rapid prediction of nitrogen oxide concentration under different working conditions.
[0009] Furthermore, a preferred implementation is provided, in which the input of the physical model of the boiler furnace established by numerical simulation software in S1 is the amount of oxygen burned in the furnace under various operating conditions, heat load, position, and amount of coal; and the output is the concentration of nitrogen oxides under various operating conditions.
[0010] Furthermore, a preferred embodiment is provided, in which the method for preprocessing the obtained simulation data in S2 is: the preprocessing is achieved by integrating and segmenting the simulation data.
[0011] Furthermore, a preferred implementation is provided, in which S3 imports the data set after the simulation data preprocessing in S2 as input data into the DT modeling software for data cleaning through the intelligent module.
[0012] Furthermore, a preferred implementation is provided, wherein S3 also includes a step of performing correlation analysis on the cleaned data.
[0013] Furthermore, a preferred implementation is provided, wherein the correlation analysis is implemented by a gated residual module GRA.
[0014] Furthermore, a preferred implementation manner is provided, in which the cleaned data is subjected to machine learning prediction using the Linear model, the CART model, the GBDT model, and the RandomForest model in S4, and the four models are used to predict the generated nitrogen oxide concentration.
[0015] Solution 2: A system for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation, the system comprising:
[0016] Data acquisition module, used to collect real-time or offline data and build a physical model of the boiler furnace using numerical simulation software;
[0017] The data processing module is used to simulate the physical model of the boiler furnace established by the data acquisition module, and to pre-process and pre-analyze the obtained simulation data;
[0018] The prediction module is used to import the data set preprocessed by the simulation data in the data processing module as input data into the DT modeling software for data cleaning, and perform machine learning prediction on the cleaned data to achieve rapid prediction of nitrogen oxide concentration under different working conditions.
[0019] Solution three: A computer device includes 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 any one of the methods described in Solution one.
[0020] Solution 4: A computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the method described in any one of Solution 1.
[0021] The present invention is beneficial in that:
[0022] In order to maintain the safety and environmental protection of boiler operation, the present invention effectively controls and reduces the concentration of nitrogen oxides in the boiler. A method for quickly predicting the concentration of nitrogen oxides in the furnace using DT modeling based on numerical simulation is now proposed to improve the calculation efficiency and accuracy of nitrogen oxides in the furnace. Based on the operating data collected on site, the concentration of nitrogen oxides is predicted and adjusted according to the oxygen content, temperature, time and other conditions in the combustion process in the furnace of the thermal power unit of the power plant, thereby maintaining and reducing the emission of nitrogen oxides, reducing pollution, and improving the environmental protection of boiler operation.
[0023] The method for rapidly predicting the concentration of nitrogen oxides in the furnace by using DT modeling based on numerical simulation described in the present invention can save the calculation time of numerical simulation, compressing the time from three days to within one minute, effectively improving the calculation speed and meeting the requirements of timely prediction. Using DT modeling for regression prediction can accurately and efficiently obtain the concentration of nitrogen oxides in the furnace under various working conditions during the combustion process, and make timely adjustments to reduce the concentration of nitrogen oxides and maintain the safety and environmental protection of the boiler.
[0024] The DT modeling method described in the present invention has intelligent and rich algorithm model theory, can conveniently and efficiently realize interactive modeling and data visualization, and can handle complex problems and accurately predict the future based on data-driven compatible physical models in the industrial field. Therefore, a method for quickly predicting the concentration of nitrogen oxides in the furnace using DT modeling based on numerical simulation is proposed to improve calculation efficiency and accuracy.
[0025] The invention is also applicable to the field of monitoring the concentration of nitrogen oxides in a furnace. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 This is a schematic diagram of the DT model prediction process in a method for quickly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation as described in Implementation Method 1.
[0027] Figure 2 This is a schematic diagram of the prediction results of the Linear model described in Implementation Method Seven.
[0028] Figure 3 This is a schematic diagram of the prediction results of the CART model described in Implementation Method Seven.
[0029] Figure 4 This is a schematic diagram of the prediction results of the GBDT model described in Implementation Method 7.
[0030] Figure 5 This is a schematic diagram of the prediction results of the RandomForest model described in Implementation Method 7.
[0031] Figure 6This is a flow chart of a method for quickly predicting the concentration of nitrogen oxides in a furnace using DT modeling based on numerical simulation as described in Implementation Method 1. DETAILED DESCRIPTION
[0032] In order to make the purpose, technical solutions and advantages of the implementation methods of the present application clearer, the technical solutions in the implementation methods of the present application will be clearly and completely described below in conjunction with the drawings in the implementation methods of the present application. Obviously, the described implementation methods are only part of the implementation methods of the present application, not all of the implementation methods.
[0033] Embodiment 1: This embodiment provides a method for quickly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation, and the method comprises the following steps:
[0034] S1. Collect real-time or offline data and use numerical simulation software to establish a physical model of the boiler furnace;
[0035] S2, performing simulation calculation on the physical model of the boiler furnace established in S1, and preprocessing and pre-analyzing the obtained simulation data;
[0036] S3. Import the preprocessed data set of the simulation data in S2 as input data into the DT modeling software for data cleaning, perform machine learning prediction on the cleaned data, and realize rapid prediction of nitrogen oxide concentration under different working conditions.
[0037] Implementation method 2. This implementation method is a further limitation of the cabin flooding flow field prediction method based on adaptive physical neural network described in implementation method 1. In S1, the physical model of the boiler furnace is established using numerical simulation software, and the input is the amount of oxygen burned in the furnace under various operating conditions, heat load, position, and coal quantity; the output is the concentration of nitrogen oxides under various operating conditions.
[0038] Implementation method three: This implementation method is a further limitation of the method for quickly predicting the nitrogen oxide concentration in the furnace by using DT modeling based on numerical simulation as described in implementation method one. The method for preprocessing the obtained simulation data in S2 is: the preprocessing is achieved by integrating and selecting the simulation data in sections.
[0039] Implementation method 4. This implementation method is a further limitation of the method for quickly predicting the concentration of nitrogen oxides in the furnace by using DT modeling based on numerical simulation as described in implementation method 1. In S3, the data set after preprocessing the simulation data in S2 is imported as input data into the DT modeling software for data cleaning, which is implemented through an intelligent module.
[0040] Embodiment 5: This embodiment further limits the method for rapidly predicting the concentration of nitrogen oxides in the furnace by using DT modeling based on numerical simulation as described in Embodiment 1. S3 also includes the step of performing correlation analysis on the cleaned data.
[0041] Embodiment 6: This embodiment further limits the method for rapidly predicting the nitrogen oxide concentration in the furnace by using DT modeling based on numerical simulation as described in Embodiment 5, and the correlation analysis is implemented by a gated residual module GRA.
[0042] Implementation method seven. This implementation method is a further limitation of the method for quickly predicting the concentration of nitrogen oxides in the furnace by using DT modeling based on numerical simulation as described in implementation method one. In S4, machine learning prediction is performed on the cleaned data using the Linear model, CART model, GBDT model, and RandomForest model. The four models predict the generated nitrogen oxide concentration.
[0043] Embodiment 8: This embodiment proposes a system for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation, and the system includes:
[0044] Data acquisition module, used to collect real-time or offline data and build a physical model of the boiler furnace using numerical simulation software;
[0045] The data processing module is used to simulate the physical model of the boiler furnace established by the data acquisition module, and to pre-process and pre-analyze the obtained simulation data;
[0046] The prediction module is used to import the data set preprocessed by the simulation data in the data processing module as input data into the DT modeling software for data cleaning, and perform machine learning prediction on the cleaned data to achieve rapid prediction of nitrogen oxide concentration under different working conditions.
[0047] Embodiment 9. This embodiment proposes a computer device, including 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 method described in any one of embodiments 1 to 7.
[0048] Embodiment 10: This embodiment proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method described in any one of embodiments 1 to 7 are implemented.
[0049] Implementation eleven: This implementation provides an example, which is used to explain the above implementations one to eight. The specific example is as follows:
[0050] See also Figures 1 to 6 To illustrate this embodiment, this embodiment proposes a method for quickly predicting the concentration of nitrogen oxides in the furnace by using DT modeling based on numerical simulation, so as to improve the calculation efficiency and accuracy of nitrogen oxides in the furnace.
[0051] Based on the real-time or offline data collected, the physical model of the boiler furnace is established using numerical simulation software, and after simulation calculation, the data is pre-processed and pre-analyzed. Using DT modeling software, machine learning prediction is performed on the processed data to achieve rapid prediction of nitrogen oxide concentration under different working conditions, providing theoretical and practical technical results support for boiler operation and maintenance.
[0052] The prediction method comprises the following steps:
[0053] (I) Numerical simulation: Use numerical simulation software to establish the physical model of the boiler furnace and perform simulation calculations based on the combustion mechanism.
[0054] (II) Data processing: Process the numerical simulation results.
[0055] (III) Algorithm prediction: Four training methods are used to perform data modeling and predict the concentration of nitrogen oxides in the furnace under different working conditions.
[0056] (IV) Application of results: Select the best algorithm from four neural networks to predict the nitrogen oxide concentration under specified working conditions.
[0057] (I) The specific steps of numerical simulation are as follows:
[0058] The real-time or offline data collected by the power plant boiler is used as the boundary condition, and the combustion state parameters of the furnace are simulated in combination with the combustion mechanism, flow model and heat transfer model. The input objects are the oxygen content, heat load, position, coal quantity, etc. of the furnace combustion under various working conditions, and the output object is the concentration of nitrogen oxides under various working conditions.
[0059] (2) Data processing.
[0060] The huge amount of data will cause problems such as jamming during data processing. Therefore, for the huge amount of simulation data, a diluted sampling method is adopted, that is, the simulation data is integrated and segmented, and real data is used for analysis based on the original samples.
[0061] The data set after data preprocessing is imported as input data into the DT modeling software for data cleaning. The AIOD intelligent module is used to screen out outliers and data that is obviously inconsistent with reality, so that the selected data is more in line with the research scope and the training accuracy is improved.
[0062] (3) Algorithm prediction.
[0063] In order to train influencing factors that are more suitable for practical applications, the data is analyzed for correlation, using the GRA module. GRA is based on gray matrix analysis and gives the correlation between variables. This training selects a correlation threshold of 0.7 or above and selects 4 variables from 12 input variables: operating conditions, temperature, y coordinate, and coal mill status.
[0064] Using the processed data, four training methods, Linear, CART, GBDT, and RandomForest, were selected on the DT modeling platform for training. The concentration of nitrogen oxides generated during the combustion process in the furnace was predicted through these four models. The input data of the test set was passed through these four networks to obtain the prediction results. Finally, the model was evaluated and selected by comparing the degree of fit, root mean square error and other evaluation indicators.
[0065] (IV) Application of results.
[0066] Design a single or multiple operating conditions to be predicted, take the operating conditions, temperature, position, and pulverizer status as input data, and use the optimal data modeling model to predict the concentration of nitrogen oxides generated during the combustion process in the furnace.
[0067] After the dataset is trained by four regression models, Linear, CART, GBDT, and RandomForest, the training results are as follows: Figures 2 to 5 As shown, where:
[0068] This study set 25% of the data as the test set for model evaluation. According to the results of model evaluation and model comparison, the prediction results of the four regression models are basically consistent with the true value, with an error of less than 1%. The results of linear regression show that the linear regression has a good fitting effect, and it is determined that the selected input variables have a good explanation for the output variables, and the prediction accuracy is high. After a comprehensive analysis of the fitting effects of mape, mse, and r2, the CART model has the best prediction effect, so the CART model is selected to predict the concentration of nitrogen oxides in the furnace.
[0069] The method of using DT modeling based on numerical simulation to quickly predict the concentration of nitrogen oxides in the furnace can save the calculation time of numerical simulation, compressing the time from three days to within one minute, effectively improving the calculation speed and meeting the requirements of timely prediction. Using DT modeling for regression prediction can accurately and efficiently obtain the concentration of nitrogen oxides in the furnace under various working conditions during the combustion process, and make timely adjustments to reduce the concentration of nitrogen oxides and maintain the safety and environmental protection of the boiler.
[0070] Figure 6Any process or method description in the flowchart described in or otherwise described herein can be understood as representing a module, fragment or portion of a code including one or more executable instructions for implementing the steps of a custom logic function or process, and the scope of the preferred embodiment of the present invention includes other implementations, in which the functions may be performed in a substantially simultaneous manner or in a reverse order according to the functions involved, which should be understood by a person skilled in the art of the present invention. The logic and / or steps represented in the flowchart or otherwise described herein illustrate the possible implementation architecture, functions and operations of the apparatus and methods according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It is also noted that each block in the block diagram or flowchart, and combinations of blocks in the block diagram or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions. For example, what may be considered as an ordered list of executable instructions for implementing a logical function may be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other system that can fetch instructions from and execute instructions in an instruction execution system, apparatus, or device).
[0071] Those skilled in the art will appreciate that the above are only preferred embodiments of the present invention, and the various embodiments of the present disclosure and / or the features described in the claims may be combined or combined in various ways, even if such combinations or combinations are not explicitly described in the present disclosure. It 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 may still modify the technical solutions described in the aforementioned embodiments, or perform equivalent substitutions on some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.
[0072] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.
Claims
1. A method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation, characterized in that: The method comprises the following steps: S1. Collect real-time or offline data and use numerical simulation software to establish a physical model of the boiler furnace; S2, performing simulation calculation on the physical model of the boiler furnace established in S1, and preprocessing and pre-analyzing the obtained simulation data; S3. Import the preprocessed data set of the simulation data in S2 as input data into the DT modeling software for data cleaning, perform machine learning prediction on the cleaned data, and realize rapid prediction of nitrogen oxide concentration under different working conditions.
2. The method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation according to claim 1 is characterized in that: The input of the physical model of the boiler furnace established by numerical simulation software in S1 is the oxygen content, heat load, position, and coal quantity of the furnace combustion under various operating conditions; the output is the concentration of nitrogen oxides under various operating conditions.
3. The method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation according to claim 1 is characterized in that: The method for preprocessing the obtained simulation data in S2 is: the preprocessing is achieved by integrating and selecting the simulation data.
4. The method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation according to claim 1 is characterized in that: In S3, the dataset preprocessed by the simulation data in S2 is imported as input data into the DT modeling software for data cleaning through the intelligent module.
5. The method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation according to claim 1 is characterized in that: S3 also includes a step of performing correlation analysis on the cleaned data.
6. The method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation according to claim 5 is characterized in that: The correlation analysis is implemented by a gated residual module GRA.
7. The method for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation according to claim 1 is characterized in that: In S4, machine learning predictions are performed on the cleaned data using the Linear model, CART model, GBDT model, and RandomForest model. Four models are used to predict the generated nitrogen oxide concentrations.
8. A system for rapidly predicting the concentration of nitrogen oxides in a furnace by using DT modeling based on numerical simulation, characterized in that: The system comprises: Data acquisition module, used to collect real-time or offline data and build a physical model of the boiler furnace using numerical simulation software; The data processing module is used to simulate the physical model of the boiler furnace established by the data acquisition module, and to pre-process and pre-analyze the obtained simulation data; The prediction module is used to import the data set preprocessed by the simulation data in the data processing module as input data into the DT modeling software for data cleaning, and perform machine learning prediction on the cleaned data to achieve rapid prediction of nitrogen oxide concentration under different working conditions.
9. A computer device comprising a memory and a processor, characterized in that A computer program is stored in the memory. When the processor runs the computer program stored in the memory, the processor executes the method 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, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.