Optimized combustion control system and method for W-type flame boiler based on wind powder balance
Through real-time data acquisition and intelligent proportion control, the optimal air powder ratio is determined using SAGAFS algorithm and optimization algorithm, which solves the problem of low combustion efficiency of W-type flame boilers, realizes an efficient, stable and environmentally friendly combustion process, and improves combustion Efficiency and stability.
Patent Information
- Application Number
- CN202411278993.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-12
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The combustion efficiency of W-type flame boiler is lower than the design standard, mainly due to the precise control of coal and air volume, which leads to low combustion efficiency and strict requirements on flame form. Any small fluctuations will affect the combustion effect.
Adoption of an optimized combustion control system based on air powder balance, and through real-time data acquisition, accurate model prediction and intelligent proportion control, the dynamic balance between coal powder and wind speed is achieved, including data acquisition module, proportion determination module and feedback adjustment module, using SAGAFS Algorithms and optimization algorithms determine the optimal air powder ratio and adjust the combustion process in real time.
It significantly improves the combustion efficiency of the boiler, reduces pollutant emissions, and enhances combustion stability, achieving dual improvements in economic and environmental benefits.
Smart Images

Figure BDA0005041070980000031 
Figure BDA0005041070980000051 
Figure BDA0005041070980000091
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of boiler combustion, and in particular to an optimized combustion control system and method for a W-type flame boiler based on air-powder balance. Background Art
[0002] W-type flame boiler is a large coal-fired boiler designed for burning low-volatile coal. Its unique furnace structure arranges the burner on the boiler arch, forming a unique W-type flame combustion method, which aims to improve the combustion efficiency and burnout rate by extending the residence time of coal powder in the furnace. However, in actual operation, the combustion efficiency of W-type flame boilers is generally lower than the design standard, usually around 90%, and its combustion efficiency is generally 1 to 2 percentage points lower than other types of pulverized coal boilers.
[0003] The main reason for the low combustion efficiency of W-type flame boilers is not the boiler design itself, but the problem of precise control of coal and air volume. Since W-type flame boilers have strict requirements on flame shape, any small fluctuation in coal or air volume may significantly affect the combustion effect. Therefore, how to achieve precise and real-time control of coal and air volume becomes the key to improving combustion efficiency. Summary of the invention
[0004] In view of the defects in the prior art, the present invention provides a W-type flame boiler optimized combustion control system and method based on air-powder balance, which realizes the dynamic balance of coal powder and wind speed through real-time data acquisition, accurate model prediction and intelligent ratio control, thereby optimizing the combustion process of the boiler.
[0005] On the one hand, an embodiment of the present invention provides a W-type flame boiler optimized combustion control system based on air-powder balance, the system comprising:
[0006] A data acquisition module is used to obtain the combustion parameters at the current moment, wherein the combustion parameters include boiler operation status parameters, environmental parameters and equipment parameters;
[0007] A ratio determination module is used to input the combustion parameters into a pre-built pulverized coal quantity prediction model and an air volume prediction model to obtain a pulverized coal quantity prediction value and an air volume prediction value, and determine the air-pulverized coal ratio according to the pulverized coal quantity prediction value and the air volume prediction value; fine-tune the air-pulverized coal ratio according to the boiler load and coal quality parameters to obtain a target air-pulverized coal ratio;
[0008] A feedback regulation module is used to perform combustion control according to the target air-powder ratio.
[0009] Preferably, the ratio determination module includes a model building unit, and the model building unit includes:
[0010] A data acquisition unit, used to collect historical combustion parameters and their corresponding wind speed and coal powder amount at future moments, wherein the historical combustion parameters include historical boiler operation state parameters, historical environmental parameters and historical equipment parameters;
[0011] A data preprocessing unit, used for performing data cleaning and normalization processing on the historical combustion parameters;
[0012] A feature selection unit, used for performing feature selection on the historical combustion parameters by using a SAGAFS algorithm, obtaining a sample set, and dividing the sample set into a training set and a validation set;
[0013] A model building unit, used for obtaining an initial air volume prediction model and an initial powder volume prediction model;
[0014] The model training unit is used to train the model using the training set, update the model parameters through the back propagation algorithm, minimize the prediction error, and obtain the air volume prediction model and the powder volume prediction model.
[0015] Preferably, the feature selection unit is specifically used for:
[0016] Based on historical combustion parameters, an initial feature subset population is randomly generated, each individual represents a feature combination, and the initial temperature, cooling rate, and genetic algorithm parameters are set; at the same time, an evaluation function is defined to measure the impact of each feature subset on the model performance;
[0017] In the current population, simulated annealing selection is applied to each individual; the operation of genetic algorithm is applied to the population after simulated annealing selection;
[0018] The temperature is reduced according to a predetermined cooling rate, and simulated annealing selection and genetic operations are repeatedly performed until the stopping condition is met and the maximum number of iterations is reached, or the temperature drops sufficiently low or the population converges;
[0019] The individuals with the best evaluation function value in the final population are selected as the selected feature subset to construct the sample set.
[0020] Preferably, the ratio determination module also includes a ratio optimization unit, which is specifically used to calculate the target air-powder ratio based on the air-powder ratio using an optimization algorithm to minimize the loss function of coal powder unburnt loss, pollutant emission and combustion stability.
[0021] Preferably, the optimization algorithm is specifically as follows:
[0022]
[0023] in, is the target air-coal ratio of the i-th burner, R iis the air-coal ratio of the i-th burner, a, β, γ are the weight coefficients of coal powder unburned loss, pollutant emission and combustion stability loss, respectively. Loss1 is the coal powder unburned loss function, C r is the residual carbon content of pulverized coal at the outlet of the i-th burner, H v is the calorific value of coal powder, m c is the coal powder mass flow rate of the i-th burner, m t is the total coal powder mass flow rate; Loss2 is the pollutant emission loss function, E f is the emission of the jth pollutant, E f is the environmental penalty coefficient related to the jth pollutant, C t is the cost of treating the jth pollutant; Loss3 is the combustion stability loss function, Var(F f,k is the variance of the flame intensity in the kth combustion area, F o,k is the flameout event frequency of the kth combustion zone, Var(E c,k ) is the variance of the combustion efficiency of the kth combustion zone.
[0024] On the other hand, an embodiment of the present invention further provides a W-type flame boiler optimized combustion control method based on air-powder balance, the method comprising:
[0025] Obtaining combustion parameters at the current moment, wherein the combustion parameters include boiler operation state parameters, environmental parameters and equipment parameters;
[0026] Input the combustion parameters into a pre-built pulverized coal amount prediction model and an air volume prediction model to obtain a pulverized coal amount prediction value and an air volume prediction value, and determine the air-pulverized coal ratio according to the pulverized coal amount prediction value and the air volume prediction value; fine-tune the air-pulverized coal ratio according to the boiler load and coal quality parameters to obtain a target air-pulverized coal ratio;
[0027] Combustion control is performed according to the target air-powder ratio.
[0028] Preferably, the process of acquiring the powder quantity prediction model and the air volume prediction model is as follows:
[0029] Collecting historical combustion parameters and their corresponding wind speed and coal powder amount at future moments, wherein the historical combustion parameters include historical boiler operation state parameters, historical environmental parameters and historical equipment parameters;
[0030] Performing data cleaning and normalization processing on the historical combustion parameters;
[0031] Using the SAGAFS algorithm to perform feature selection on the historical combustion parameters, obtaining a sample set, and dividing the sample set into a training set and a validation set;
[0032] Obtaining an initial air volume prediction model and an initial powder volume prediction model;
[0033] The model is trained using the training set, and the model parameters are updated through the back propagation algorithm to minimize the prediction error, thus obtaining the air volume prediction model and the powder volume prediction model.
[0034] Preferably, the SAGAFS algorithm is used to perform feature selection on the historical combustion parameters to obtain a sample set, including:
[0035] Based on historical combustion parameters, an initial feature subset population is randomly generated, each individual represents a feature combination, and the initial temperature, cooling rate, and genetic algorithm parameters are set; at the same time, an evaluation function is defined to measure the impact of each feature subset on the model performance;
[0036] In the current population, simulated annealing selection is applied to each individual; the operation of genetic algorithm is applied to the population after simulated annealing selection;
[0037] The temperature is reduced according to a predetermined cooling rate, and simulated annealing selection and genetic operations are repeatedly performed until the stopping condition is met and the maximum number of iterations is reached, or the temperature drops sufficiently low or the population converges;
[0038] The individuals with the best evaluation function value in the final population are selected as the selected feature subset to construct the sample set.
[0039] Preferably, the air-powder ratio is fine-tuned according to boiler load and coal quality parameters to obtain a target air-powder ratio, including: based on the air-powder ratio, an optimization algorithm is used to calculate the target air-powder ratio to minimize the loss function of unburned coal powder loss, pollutant emissions and combustion stability.
[0040] Preferably, the optimization algorithm is specifically as follows:
[0041]
[0042] in, is the target air-coal ratio of the i-th burner, R i is the air-coal ratio of the i-th burner, a, β, γ are the weight coefficients of coal powder unburned loss, pollutant emission and combustion stability loss, respectively. Loss1 is the coal powder unburned loss function, C r is the residual carbon content of pulverized coal at the outlet of the i-th burner, H v is the calorific value of coal powder, m c is the coal powder mass flow rate of the i-th burner, m t is the total coal powder mass flow rate; Loss2 is the pollutant emission loss function, E f is the emission of the jth pollutant, E f is the environmental penalty coefficient related to the jth pollutant, C tis the cost of treating the jth pollutant; Loss3 is the combustion stability loss function, Var(F f,k is the variance of the flame intensity in the kth combustion area, F o,k is the flameout event frequency of the kth combustion zone, Var(E c,k ) is the variance of the combustion efficiency of the kth combustion zone.
[0043] 1) The beneficial effects of the present invention are as follows: The embodiment of the present invention provides a W-type flame boiler optimized combustion control system based on air-powder balance, which realizes precise control of the boiler combustion process by integrating three modules: data acquisition, ratio determination and feedback regulation. The system first collects various real-time combustion parameters of the boiler, and then uses the pre-built powder quantity prediction model and air volume prediction model, combined with the SAGAFS algorithm (to preliminarily determine and fine-tune the air-powder ratio. This process not only takes into account key factors such as coal powder concentration, coal feed, primary air volume and wind speed, boiler load, and coal mill operating status, but also further minimizes the comprehensive loss function of unburned coal powder loss, pollutant emissions, and combustion stability loss through the optimization algorithm to obtain the optimal air-powder ratio. Finally, the feedback regulation module adjusts the combustion process in real time according to the target air-powder ratio to ensure efficient, stable and environmentally friendly operation of the boiler. The advantages of the present invention are that it can significantly improve the combustion efficiency of the boiler, reduce pollutant emissions, and enhance combustion stability, thereby achieving a dual improvement in economic and environmental benefits.
[0044] In addition, an embodiment of the present invention also provides a W-type flame boiler optimized combustion control method based on air-powder balance. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.
[0046] Figure 1 A system block diagram of a W-type flame boiler optimized combustion control system based on air-powder balance is provided for an embodiment of the present invention;
[0047] Figure 2 A flow chart of a W-type flame boiler optimized combustion control method based on air-powder balance is provided for an embodiment of the present invention. DETAILED DESCRIPTION
[0048] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.
[0049] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.
[0050] In order to better understand the technical solution of the present invention, the basic principle of the Spinal code involved in the present invention is described in detail below.
[0051] Example 1
[0052] like Figure 1 As shown, an embodiment of the present invention provides a W-type flame boiler optimized combustion control system based on air-powder balance, the system comprising:
[0053] The data acquisition module is used to obtain the combustion parameters at the current moment, and the combustion parameters include boiler operation status parameters, environmental parameters and equipment parameters.
[0054] The data acquisition module is responsible for obtaining various combustion parameters of the boiler in real time, including but not limited to boiler operating status parameters (such as coal powder amount, wind speed, air volume, etc.), environmental parameters (such as ambient temperature, humidity) and equipment parameters (such as burner angle, etc.). These parameters are the basis for subsequent air-powder ratio calculation and combustion control.
[0055] The ratio determination module is used to input the combustion parameters into a pre-built powder quantity prediction model and an air volume prediction model to obtain a coal powder quantity prediction value and an air volume prediction value, and determine the air-powder ratio according to the coal powder quantity prediction value and the air volume prediction value; fine-tune the air-powder ratio according to the boiler load and coal quality parameters to obtain a target air-powder ratio.
[0056] When constructing and training a prediction model for W-type flame boiler combustion optimization, a series of detailed steps need to be taken, including data collection, feature selection, model structure design, training and verification. Specifically, the ratio determination module includes a model construction unit, which includes: a data acquisition unit for collecting historical combustion parameters and their corresponding wind speed and coal powder amount at future moments, the historical combustion parameters include historical boiler operating state parameters, historical environmental parameters and historical equipment parameters; a data preprocessing unit for data cleaning and normalization of the historical combustion parameters; a feature selection unit for selecting features of the historical combustion parameters using the SAGAFS algorithm, obtaining a sample set, and dividing the sample set into a training set and a verification set; a model construction unit for obtaining an initial air volume prediction model and an initial powder amount prediction model; a model training unit for training the model using the training set, updating the model parameters through the back propagation algorithm, minimizing the prediction error, and obtaining an air volume prediction model and a powder amount prediction model.
[0057] It should be noted that the feature selection unit is specifically used to: randomly generate an initial feature subset population based on historical combustion parameters, each individual represents a feature combination, set the initial temperature, cooling rate, and genetic algorithm parameters; at the same time, define an evaluation function to measure the impact of each feature subset on the model performance; in the current population, apply simulated annealing selection to each individual; apply the genetic algorithm operation to the population after simulated annealing selection; reduce the temperature according to a predetermined cooling rate, and repeat the simulated annealing selection and genetic operation until the stopping condition is met and the maximum number of iterations is reached or the temperature drops to a sufficiently low level or the population converges; select the individual with the best evaluation function value in the final population as the selected feature subset to construct a sample set.
[0058] Specifically, the annealing process includes: for each individual, randomly select a "neighbor" (i.e., a similar feature subset) and compare the evaluation function values of the two. According to the Metropolis criterion, accept a worse solution with a certain probability in order to jump out of the local optimum. The acceptance probability is related to the current temperature and the difference between the evaluation function values of the two. The operations of the genetic algorithm include: selection (based on fitness), crossover (exchanging some features of two individuals) and mutation (randomly changing certain features in an individual).
[0059] Simulated Annealing (SA) is a general probabilistic algorithm used to find the optimal solution to a proposition in a large search space. It simulates the annealing process of solid matter in physics, and by giving the search process a time-varying and eventually zero probability jump, it can effectively avoid falling into the local optimum and eventually tend to the global optimum. Genetic Algorithm (GA) is a search heuristic algorithm that simulates natural selection and genetic mechanisms. It searches for the optimal solution by simulating the natural evolution process. Genetic algorithms continuously iterate and update individuals in the population through operations such as selection, crossover and mutation, in order to achieve the global optimum. The SAGAFS provided by the present invention combines the global search capability of the simulated annealing algorithm and the iterative optimization characteristics of the genetic algorithm, aiming to perform feature selection more efficiently.
[0060] Specifically, the parameters determined by the SAGAFS algorithm include: historical data of coal powder concentration: the measured value of coal powder concentration in the past period of time; coal feed rate: the amount of coal powder entering the boiler, which is the direct factor affecting the coal powder concentration; primary air volume and wind speed: the primary air is responsible for transporting coal powder to the boiler, and its air volume and wind speed will affect the distribution and concentration of coal powder; boiler load: under different loads, the boiler's demand for coal powder and combustion conditions may be different, thus affecting the coal powder concentration; pulverizer operating status: the output and efficiency of the pulverizer directly affect the fineness and output of coal powder.
[0061] The SAGAFS algorithm of the embodiment of the present invention combines the global search capability of simulated annealing and the iterative optimization capability of genetic algorithms, and is likely to find a global optimal solution; by introducing probabilistic jumps, it is helpful to jump out of a local optimal solution.
[0062] In an embodiment of the present invention, the ratio determination module further includes a ratio optimization unit, which is specifically used to calculate the target air-powder ratio based on the air-powder ratio using an optimization algorithm to minimize the loss function of unburned coal loss, pollutant emissions and combustion stability. The optimization algorithm is specifically:
[0063]
[0064] in, is the target air-coal ratio of the i-th burner, R i is the air-coal ratio of the i-th burner, a , β, γ are the weight coefficients of coal powder unburned loss, pollutant emission and combustion stability loss, respectively. Loss1 is the coal powder unburned loss function, C r is the residual carbon content of pulverized coal at the outlet of the i-th burner, H v is the calorific value of coal powder, m c is the coal powder mass flow rate of the i-th burner, mt is the total coal powder mass flow rate; Loss2 is the pollutant emission loss function, E f is the emission of the jth pollutant, E f is the environmental penalty coefficient related to the jth pollutant, C t is the cost of treating the jth pollutant; Loss3 is the combustion stability loss function, Var(F f,k is the variance of the flame intensity in the kth combustion area, F o,k is the flameout event frequency of the kth combustion zone, Var(E c,k ) is the variance of the combustion efficiency of the kth combustion zone.
[0065] Based on the initially determined air-powder ratio, the ratio optimization unit further calculates the target air-powder ratio using an optimization algorithm to minimize the comprehensive loss function.
[0066] Specifically, a comprehensive loss function is constructed that includes coal powder unburned loss, pollutant emission loss and combustion stability loss. This function strives to achieve the best balance between the three by adjusting the air-powder ratio. The specific implementation process of the optimization algorithm is: set the initial value of the air-powder ratio, learning rate and other parameters; calculate the gradient of the loss function (i.e., the partial derivative of the loss function with respect to each parameter) according to the current air-powder ratio; update the air-powder ratio parameter value in the opposite direction of the gradient to reduce the value of the loss function. The update formula is usually: new parameter value = old parameter value - learning rate * gradient; repeat the gradient calculation and parameter update steps until the loss function value converges to a stable range or reaches the preset number of iterations. ; The air-powder ratio obtained after the iteration is the optimal solution output by the optimization algorithm, which is used to guide the real-time adjustment of the combustion control system.
[0067] The feedback adjustment module is used to perform combustion control according to the target air-powder ratio. Specifically, the combustion control includes: outputting control instructions to each burner to adjust the air volume, wind speed, etc. of the primary air, as well as the coal powder feeding rate, etc., to achieve the best combustion effect.
[0068] In summary, the embodiment of the present invention provides a W-type flame boiler optimized combustion control system based on air-powder balance, which realizes precise control of the boiler combustion process by integrating three modules: data acquisition, ratio determination and feedback regulation. The system first collects various real-time combustion parameters of the boiler, and then uses the pre-built powder quantity prediction model and air volume prediction model, combined with the SAGAFS algorithm (to perform preliminary determination and fine-tuning optimization of the air-powder ratio. This process not only takes into account key factors such as coal powder concentration, coal feed, primary air volume and wind speed, boiler load, and coal mill operating status, but also further minimizes the comprehensive loss function of unburned coal powder loss, pollutant emissions, and combustion stability loss through the optimization algorithm to obtain the optimal air-powder ratio. Finally, the feedback regulation module adjusts the combustion process in real time according to the target air-powder ratio to ensure efficient, stable and environmentally friendly operation of the boiler. The advantages of the present invention are that it can significantly improve the combustion efficiency of the boiler, reduce pollutant emissions, and enhance combustion stability, thereby achieving a dual improvement in economic and environmental benefits.
[0069] Example 2
[0070] Based on the same inventive concept, Figure 2 As shown, the embodiment of the present invention also provides a W-type flame boiler optimized combustion control method based on air-powder balance, the method comprising:
[0071] Obtaining combustion parameters at the current moment, wherein the combustion parameters include boiler operation state parameters, environmental parameters and equipment parameters;
[0072] Input the combustion parameters into a pre-built pulverized coal amount prediction model and an air volume prediction model to obtain a pulverized coal amount prediction value and an air volume prediction value, and determine the air-pulverized coal ratio according to the pulverized coal amount prediction value and the air volume prediction value; fine-tune the air-pulverized coal ratio according to the boiler load and coal quality parameters to obtain a target air-pulverized coal ratio;
[0073] Combustion control is performed according to the target air-powder ratio.
[0074] In the embodiment of the present invention, the process of acquiring the powder quantity prediction model and the air volume prediction model is as follows:
[0075] Collecting historical combustion parameters and their corresponding wind speed and coal powder amount at future moments, wherein the historical combustion parameters include historical boiler operation state parameters, historical environmental parameters and historical equipment parameters;
[0076] Performing data cleaning and normalization processing on the historical combustion parameters;
[0077] Using the SAGAFS algorithm to perform feature selection on the historical combustion parameters, obtaining a sample set, and dividing the sample set into a training set and a validation set;
[0078] Obtaining an initial air volume prediction model and an initial powder volume prediction model;
[0079] The model is trained using the training set, and the model parameters are updated through the back propagation algorithm to minimize the prediction error, thus obtaining the air volume prediction model and the powder volume prediction model.
[0080] In an embodiment of the present invention, the SAGAFS algorithm is used to perform feature selection on the historical combustion parameters to obtain a sample set, including:
[0081] Based on historical combustion parameters, an initial feature subset population is randomly generated, each individual represents a feature combination, and the initial temperature, cooling rate, and genetic algorithm parameters are set; at the same time, an evaluation function is defined to measure the impact of each feature subset on the model performance;
[0082] In the current population, simulated annealing selection is applied to each individual; the operation of genetic algorithm is applied to the population after simulated annealing selection;
[0083] The temperature is reduced according to a predetermined cooling rate, and simulated annealing selection and genetic operations are repeatedly performed until the stopping condition is met and the maximum number of iterations is reached, or the temperature drops sufficiently low or the population converges;
[0084] The individuals with the best evaluation function value in the final population are selected as the selected feature subset to construct the sample set.
[0085] In an embodiment of the present invention, the air-powder ratio is fine-tuned according to the boiler load and coal quality parameters to obtain the target air-powder ratio, including: based on the air-powder ratio, an optimization algorithm is used to calculate the target air-powder ratio to minimize the loss function of unburned coal powder loss, pollutant emissions and combustion stability.
[0086] In the embodiment of the present invention, the optimization algorithm is specifically:
[0087]
[0088] in, is the target air-coal ratio of the i-th burner, R i is the air-coal ratio of the i-th burner, a, β, γ are the weight coefficients of coal powder unburned loss, pollutant emission and combustion stability loss, respectively. Loss1 is the coal powder unburned loss function, C r is the residual carbon content of pulverized coal at the outlet of the i-th burner, H v is the calorific value of coal powder, m c is the coal powder mass flow rate of the i-th burner, m t is the total coal powder mass flow rate; Loss2 is the pollutant emission loss function, E f is the emission of the jth pollutant, E fis the environmental penalty coefficient related to the jth pollutant, C t is the cost of treating the jth pollutant; Loss3 is the combustion stability loss function, Var(F f,k is the variance of the flame intensity in the kth combustion area, F o,k is the flameout event frequency of the kth combustion zone, Var(E c,k ) is the variance of the combustion efficiency of the kth combustion zone.
[0089] It should be understood that the W-type flame boiler optimized combustion control method based on air-powder balance provided in the embodiment of the present invention is based on the same inventive concept as the above-mentioned embodiment. For more specific working principles of each step in the embodiment of the present invention, please refer to the above-mentioned embodiment and will not be repeated in this embodiment.
[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.
Claims
1. W-type flame boiler optimized combustion control system based on air-powder balance, characterized in that: include: A data acquisition module is used to obtain the combustion parameters at the current moment, wherein the combustion parameters include boiler operation status parameters, environmental parameters and equipment parameters; A ratio determination module is used to input the combustion parameters into a pre-built pulverized coal quantity prediction model and an air volume prediction model to obtain a pulverized coal quantity prediction value and an air volume prediction value, and determine the air-pulverized coal ratio according to the pulverized coal quantity prediction value and the air volume prediction value; fine-tune the air-pulverized coal ratio according to the boiler load and coal quality parameters to obtain a target air-pulverized coal ratio; A feedback regulation module is used to perform combustion control according to the target air-powder ratio.
2. The W-type flame boiler optimized combustion control system based on air-powder balance according to claim 1 is characterized in that: The ratio determination module includes a model building unit, and the model building unit includes: A data acquisition unit, used to collect historical combustion parameters and their corresponding wind speed and coal powder amount at future moments, wherein the historical combustion parameters include historical boiler operation state parameters, historical environmental parameters and historical equipment parameters; A data preprocessing unit, used for performing data cleaning and normalization processing on the historical combustion parameters; A feature selection unit, used for performing feature selection on the historical combustion parameters by using a SAGAFS algorithm, obtaining a sample set, and dividing the sample set into a training set and a validation set; A model building unit, used for obtaining an initial air volume prediction model and an initial powder volume prediction model; The model training unit is used to train the model using the training set, update the model parameters through the back propagation algorithm, minimize the prediction error, and obtain the air volume prediction model and the powder volume prediction model.
3. The W-type flame boiler optimized combustion control system based on air-powder balance according to claim 2 is characterized in that: The feature selection unit is specifically used for: Based on historical combustion parameters, an initial feature subset population is randomly generated, each individual represents a feature combination, and the initial temperature, cooling rate, and genetic algorithm parameters are set; at the same time, an evaluation function is defined to measure the impact of each feature subset on the model performance; In the current population, simulated annealing selection is applied to each individual; the operation of genetic algorithm is applied to the population after simulated annealing selection; The temperature is reduced according to a predetermined cooling rate, and simulated annealing selection and genetic operations are repeatedly performed until the stopping condition is met and the maximum number of iterations is reached, or the temperature drops sufficiently low or the population converges; The individuals with the best evaluation function value in the final population are selected as the selected feature subset to construct the sample set.
4. The W-type flame boiler optimized combustion control system based on air-powder balance according to claim 1 is characterized in that: The ratio determination module also includes a ratio optimization unit, which is specifically used to calculate the target air-powder ratio based on the air-powder ratio using an optimization algorithm to minimize the loss function of coal powder unburned loss, pollutant emissions and combustion stability.
5. The W-type flame boiler optimized combustion control system based on air-powder balance according to claim 4 is characterized in that: The optimization algorithm is specifically: in, is the target air-coal ratio of the i-th burner, R i is the air-coal ratio of the i-th burner, a, β, γ are the weight coefficients of coal powder unburned loss, pollutant emission and combustion stability loss, respectively. Loss1 is the coal powder unburned loss function, C r is the residual carbon content of pulverized coal at the outlet of the i-th burner, H v is the calorific value of coal powder, m c is the coal powder mass flow rate of the i-th burner, m t is the total coal powder mass flow rate; Loss2 is the pollutant emission loss function, E f is the emission of the jth pollutant, E f is the environmental penalty coefficient related to the jth pollutant, C t is the cost of treating the jth pollutant; Loss3 is the combustion stability loss function, Var(F f,k is the variance of the flame intensity in the kth combustion area, F o,k is the flameout event frequency of the kth combustion zone, Var(E c,k ) is the variance of the combustion efficiency of the kth combustion zone.
6. A W-type flame boiler optimized combustion control method based on air-powder balance, characterized in that: include: Obtaining combustion parameters at the current moment, wherein the combustion parameters include boiler operation state parameters, environmental parameters and equipment parameters; Input the combustion parameters into a pre-built pulverized coal amount prediction model and an air volume prediction model to obtain a pulverized coal amount prediction value and an air volume prediction value, and determine the air-pulverized coal ratio according to the pulverized coal amount prediction value and the air volume prediction value; fine-tune the air-pulverized coal ratio according to the boiler load and coal quality parameters to obtain a target air-pulverized coal ratio; Combustion control is performed according to the target air-powder ratio.
7. The W-type flame boiler optimized combustion control method based on air-powder balance according to claim 6 is characterized in that: The process of obtaining the powder quantity prediction model and the air volume prediction model is as follows: Collecting historical combustion parameters and their corresponding wind speed and coal powder amount at future moments, wherein the historical combustion parameters include historical boiler operation state parameters, historical environmental parameters and historical equipment parameters; Performing data cleaning and normalization processing on the historical combustion parameters; Using the SAGAFS algorithm to perform feature selection on the historical combustion parameters, obtaining a sample set, and dividing the sample set into a training set and a validation set; Obtaining an initial air volume prediction model and an initial powder volume prediction model; The model is trained using the training set, and the model parameters are updated through the back propagation algorithm to minimize the prediction error, thus obtaining the air volume prediction model and the powder volume prediction model.
8. The W-type flame boiler optimized combustion control method based on air-powder balance according to claim 7 is characterized in that: The SAGAFS algorithm is used to perform feature selection on the historical combustion parameters to obtain a sample set, including: Based on historical combustion parameters, an initial feature subset population is randomly generated, each individual represents a feature combination, and the initial temperature, cooling rate, and genetic algorithm parameters are set; at the same time, an evaluation function is defined to measure the impact of each feature subset on the model performance; In the current population, simulated annealing selection is applied to each individual; the operation of genetic algorithm is applied to the population after simulated annealing selection; The temperature is reduced according to a predetermined cooling rate, and simulated annealing selection and genetic operations are repeatedly performed until the stopping condition is met and the maximum number of iterations is reached, or the temperature drops sufficiently low or the population converges; The individuals with the best evaluation function value in the final population are selected as the selected feature subset to construct the sample set.
9. The W-type flame boiler optimized combustion control method based on air-powder balance according to claim 1 is characterized in that: The air-powder ratio is fine-tuned according to boiler load and coal quality parameters to obtain a target air-powder ratio, including: based on the air-powder ratio, an optimization algorithm is used to calculate the target air-powder ratio to minimize the loss function of unburned coal powder, pollutant emissions and combustion stability.
10. The W-type flame boiler optimized combustion control method based on air-powder balance according to claim 9 is characterized in that: The optimization algorithm is specifically: in, is the target air-coal ratio of the i-th burner, R i is the air-coal ratio of the i-th burner, a, β, γ are the weight coefficients of coal powder unburned loss, pollutant emission and combustion stability loss, respectively. Loss1 is the coal powder unburned loss function, C r is the residual carbon content of pulverized coal at the outlet of the i-th burner, H v is the calorific value of coal powder, m c is the coal powder mass flow rate of the i-th burner, m t is the total coal powder mass flow rate; Loss2 is the pollutant emission loss function, E f is the emission of the jth pollutant, E f is the environmental penalty coefficient related to the jth pollutant, C t is the cost of treating the jth pollutant; Loss3 is the combustion stability loss function, Var(F f,k is the variance of the flame intensity in the kth combustion area, F o,k is the flameout event frequency of the kth combustion zone, Var(E c,k ) is the variance of the combustion efficiency of the kth combustion zone.
Citation Information
Cited By
Boiler pulverized coal monitoring method
CN120908373A
