Partition-controllable modular porous medium burner

Abstract

The invention belongs to the technical field of combustors, and discloses a modularized porous medium combustor with controllable partitions. The system comprises an area division module, a control module and a control module, wherein the area division module is used for dividing the combustor into R independent control subareas; the first acquisition module is used for acquiring combustion operation data of the R independent control zones in unit time, and the combustion operation data comprises air flow, fuel flow and combustion temperature; the first processing module is used for performing feature extraction on the combustion operation data to obtain operation feature data corresponding to the R independent control partitions; the second acquisition module is used for acquiring monitoring gas data of R independent control zones in unit time, and the monitoring gas data comprises oxygen concentration, carbon monoxide concentration, nitrogen oxide concentration and hydrocarbon concentration; and through zone control and self-adaptive adjustment, each independent control zone maintains the optimal air-fuel ratio and combustion state, and the more efficient, safer and environment-friendly combustion process is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of burners, and more particularly to a modular porous medium burner with controllable zones. Background Art

[0002] With the continuous improvement of energy efficiency and environmental protection requirements, porous media combustion technology, as a new type of combustion method, has received widespread attention in recent years. By burning in porous materials, porous media burners can significantly improve combustion efficiency and reduce pollutant emissions. They also have the advantages of good flame stability and high combustion intensity. With the improvement of industrial production requirements for energy conservation and emission reduction, how to achieve more flexible and precise combustion control has become an important technical challenge. In recent years, with the rapid development of control technology and data analysis technology, combustion systems that use intelligent control methods and perform dynamic adjustments based on real-time sensor data have gradually attracted attention. Therefore, the development of a porous media burner technology that can overcome the limitations of existing technologies, has higher efficiency and wider applicability, is of great significance to promoting the advancement of clean and efficient combustion technologies and meeting increasingly stringent energy and environmental requirements.

[0003] In order to improve the combustion efficiency of the burner, the existing Chinese patent with publication number CN113494708A proposes a modular gas burner and its operation method. The burner includes: an air flow nozzle composed of an air pipe, a positioning pipe, and a fuel pipe. Bellows barrel, bellows air inlet pipe and connection system, positioning round steel and porous medium body assembly. The combustion-supporting air is connected to the bellows through the bellows air inlet pipe and the external air introduction pipe and the output flange, and then sent into the bellows through the bellows end plate, enters the positioning pipe and is further mixed with the gas from the fuel pipe and burned. The bellows barrel and the combustion chamber are connected by the bellows outer flange, and the combustion chamber is connected by the mixing chamber and the pre-combustion chamber through the positioning pins and pins.

[0004] Although the above method can meet most scenarios, research and practical application of the above method and existing technology have found that the above method and existing technology have at least the following defects:

[0005] Most traditional burners adopt an overall control mode, that is, uniform fuel supply and air conditioning for the entire combustion area, and cannot perform adaptive control according to the heat load demand, temperature change or combustion efficiency of different areas. When the load changes or the combustion state changes, it makes it difficult for the burner to maintain the optimal air-fuel ratio and combustion state in each area, thus affecting the combustion efficiency and environmental performance of the overall system.

[0006] In view of this, the present invention proposes a modular porous medium burner with controllable partitions to solve the above problems. Summary of the invention

[0007] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a modular porous medium burner with controllable partitions, comprising:

[0008] The zone division module is used to divide the burner into R independent control zones;

[0009] A first acquisition module is used to acquire combustion operation data of R independent control partitions per unit time, wherein the combustion operation data includes air flow, fuel flow and combustion temperature;

[0010] The first processing module is used to extract features from the combustion operation data to obtain operation feature data corresponding to R independent control partitions;

[0011] A second acquisition module is used to collect monitoring gas data of R independent control partitions per unit time, wherein the monitoring gas data includes oxygen concentration, carbon monoxide concentration, nitrogen oxide concentration and hydrocarbon concentration;

[0012] The second processing module is used to extract features from the monitored gas data to obtain gas feature data corresponding to R independent control partitions;

[0013] A combustion diagnosis module, used for inputting the operation characteristic data and gas characteristic data of the R independent control partitions into the combustion diagnosis model to obtain the combustion diagnosis results corresponding to the R independent control partitions;

[0014] The automatic adjustment module is used to adaptively adjust the burners of the R independent control partitions according to the combustion diagnosis results corresponding to the R independent control partitions.

[0015] Furthermore, the method for adaptively adjusting the combustion systems of the R independently controlled partitions includes:

[0016] S121: Preset the initial value of r to 1, and the value range of r is 1 to R;

[0017] S122: Obtain the combustion diagnosis result of the rth independent control partition. If the combustion diagnosis result is normal combustion, execute S124; otherwise, execute S123;

[0018] S123: if the combustion diagnosis result is incomplete combustion, the burner is adaptively adjusted for the incomplete combustion; if the combustion diagnosis result is an air-fuel ratio imbalance, the burner is adaptively adjusted for the air-fuel ratio imbalance; if the combustion diagnosis result is local overheating, the burner is adaptively adjusted for the local overheating; if the combustion diagnosis result is a flameout risk, the burner is adaptively adjusted for the flameout risk; after the adaptive adjustment, S124 is executed;

[0019] S124: Let r=r+1. If r is less than or equal to R, continue to execute S122; if r is greater than R, end the current process.

[0020] Furthermore, the method for adaptively adjusting the incomplete combustion includes:

[0021] Inputting the operation characteristic data and the gas characteristic data into the first parameter model to obtain predicted first parameter data, wherein the first parameter data includes air supply, fuel supply and burner power; sending a control instruction to the burner to adjust the current air supply to the predicted air supply, the current fuel supply to the predicted fuel supply, and the current burner power to the predicted burner power;

[0022] The training method of the first parameter model includes:

[0023] A first training data set is collected in advance, the first training data set includes H groups of first training data, and first parameter data corresponding to the H groups of first training data, where H is a positive integer greater than 0; the first training data includes operation characteristic data and gas characteristic data; the first training data set is divided into a first training set and a first test set, the first training data in the first training set is used as the input of the first parameter model, and the first parameter data in the first training set is used as the output of the first parameter model, with minimizing the sum of the prediction accuracies of all predicted first parameter data as the training goal; the training is stopped when the sum of the prediction accuracies converges; the first parameter model is a decision tree model or a support vector machine model.

[0024] Furthermore, the method for adaptively adjusting the air-fuel ratio imbalance includes:

[0025] Inputting the operation characteristic data and the gas characteristic data into the second parameter model to obtain predicted second parameter data, wherein the second parameter data includes an air supply amount and a fuel supply amount; dividing the predicted air supply amount by the predicted fuel supply amount to obtain a predicted air-fuel ratio, and sending a control instruction to the burner to adjust the current air supply amount to the predicted air supply amount, the current fuel supply amount to the predicted fuel supply amount, and the current air-fuel ratio to the predicted air-fuel ratio;

[0026] The training method of the second parameter model includes:

[0027] A second training data set is collected in advance, the second training data set includes J groups of second training data, and second parameter data corresponding to the J groups of second training data, where J is a positive integer greater than 0; the second training data includes operation characteristic data and gas characteristic data; the second training data set is divided into a second training set and a second test set, the second training data in the second training set is used as the input of the second parameter model, and the second parameter data in the second training set is used as the output of the second parameter model, with minimizing the sum of the prediction accuracies of all predicted second parameter data as the training goal; the training is stopped when the sum of the prediction accuracies converges; the second parameter model is a decision tree model or a support vector machine model.

[0028] Furthermore, the method for adaptively adjusting local overheating includes:

[0029] Inputting the operation characteristic data and the gas characteristic data into the third parameter model to obtain predicted third parameter data, wherein the third parameter data includes an air supply amount and a fuel supply amount; sending a control instruction to the burner to adjust the current air supply amount to the predicted air supply amount, and adjust the current fuel supply amount to the predicted fuel supply amount;

[0030] The training method of the third parameter model includes:

[0031] A third training data set is collected in advance, the third training data set includes K groups of third training data, and third parameter data corresponding to the K groups of third training data, K is a positive integer greater than 0; the third training data includes operation characteristic data and gas characteristic data; the third training data set is divided into a third training set and a third test set, the third training data in the third training set is used as the input of the third parameter model, and the third parameter data in the third training set is used as the output of the third parameter model, so as to minimize the sum of the prediction accuracies of all predicted third parameter data as the training goal; the training is stopped when the sum of the prediction accuracies converges; the third parameter model is a random forest model.

[0032] Furthermore, the method for adaptively adjusting the flameout risk includes:

[0033] Inputting the operation characteristic data and the gas characteristic data into the fourth parameter model to obtain predicted fourth parameter data, wherein the fourth parameter data includes air supply and burner power; sending a control instruction to the burner to adjust the current air supply to the predicted air supply and the current burner power to the predicted burner power;

[0034] The training method of the fourth parameter model includes:

[0035] A fourth training data set is collected in advance, the fourth training data set includes L groups of fourth training data, and fourth parameter data corresponding to the L groups of fourth training data, where L is a positive integer greater than 0; the fourth training data includes operation characteristic data and gas characteristic data; the fourth training data set is divided into a fourth training set and a fourth test set, the fourth training data in the fourth training set is used as the input of a fourth parameter model, and the fourth parameter data in the fourth training set is used as the output of the fourth parameter model, so as to minimize the sum of the prediction accuracies of all predicted fourth parameter data as a training goal; the training is stopped when the sum of the prediction accuracies converges; the fourth parameter model is a naive Bayes model or a gradient boosting tree model.

[0036] Furthermore, the method for acquiring the operation characteristic data of the R independently controlled partitions includes:

[0037] S101: Preset an initial value of r to 1, and the value range of r is 1 to R; divide the unit time into M time points, preset an initial value of m to 2, and the value range of m is 2 to M; preset an initial value of n to 1, and the value range of n is M-1;

[0038] S102: The time interval between the m-1th and mth time points constitutes the nth monitoring time interval; the air flow rate change rate of the nth monitoring time interval is calculated based on the air flow rates at the m-1th and mth time points; the fuel flow rate change rate of the nth monitoring time interval is calculated based on the fuel flow rates at the m-1th and mth time points; the combustion temperature change rate of the nth monitoring time interval is calculated based on the combustion temperatures at the m-1th and mth time points;

[0039] S103: adding the air flow rate change rate to the air flow rate change rate set corresponding to the r-th independent control partition, adding the fuel flow rate change rate to the fuel flow rate change rate set corresponding to the r-th independent control partition, and adding the combustion temperature change rate to the combustion temperature change rate set corresponding to the r-th independent control partition;

[0040] S104: Let m=m+1. If m is less than or equal to M, execute S102 to S103. If m is greater than M, execute S105 to S106.

[0041] S105: The air flow rate change rate set, the fuel flow rate change rate set and the combustion temperature change rate set constitute the operation characteristic data of the rth independent control partition;

[0042] S106: Let r=r+1. If r is less than or equal to R, then let m=2 and execute S102 to S104. If r is greater than R, then end the current process.

[0043] Furthermore, the method for calculating the air flow rate change rate includes:

[0044]

[0045] QUR rn KQLL is the air flow rate change rate of the rth independent control partition in the nth monitoring time interval, rm KQLL is the air flow rate corresponding to the rth independent control partition at the mth time point, r(m-1 is the air flow rate corresponding to the rth independent control partition at the m-1th time point, T m is the mth time point, T m-1 is the m-1th time point;

[0046] The method for calculating the fuel flow rate change rate includes:

[0047]

[0048] RLBH rn RLL is the fuel flow rate change rate of the rth independent control partition in the nth monitoring time interval. rm is the fuel flow rate corresponding to the rth independent control partition at the mth time point, RLL r(m-1 is the fuel flow rate corresponding to the rth independent control partition at the m-1th time point;

[0049] The method for calculating the combustion temperature change rate includes:

[0050]

[0051] RSBH rn RSWD is the combustion temperature change rate of the rth independent control partition in the nth monitoring time interval. rm RSWD is the combustion temperature corresponding to the rth independent control partition at the mth time point, r(m-1) is the combustion temperature corresponding to the rth independent control partition at the m-1th time point.

[0052] Furthermore, the method for acquiring the gas characteristic data of the R independently controlled partitions includes:

[0053] S111: Preset an initial value of r to 1, and the value range of r is 1 to R; divide the unit time into M time points, preset an initial value of m to 2, and the value range of m is 2 to M; preset an initial value of n to 1, and the value range of n is M-1;

[0054] S112: The time interval between the m-1th and mth time points constitutes the nth monitoring time interval; the oxygen concentration change rate of the nth monitoring time interval is calculated based on the oxygen concentrations at the m-1th and mth time points; the carbon monoxide concentration change rate of the nth monitoring time interval is calculated based on the carbon monoxide concentrations at the m-1th and mth time points; the nitrogen oxide concentration change rate of the nth monitoring time interval is calculated based on the nitrogen oxide concentrations at the m-1th and mth time points; the hydrocarbon concentration change rate of the nth monitoring time interval is calculated based on the hydrocarbon concentrations at the m-1th and mth time points;

[0055] S113: adding the oxygen concentration change rate to the first concentration change rate set of the rth independent control partition, adding the carbon monoxide concentration change rate to the second concentration change rate set of the rth independent control partition, adding the nitrogen oxide concentration change rate to the third concentration change rate set of the rth independent control partition; adding the hydrocarbon concentration change rate to the fourth concentration change rate set of the rth independent control partition;

[0056] S114: Let m=m+1. If m is less than or equal to M, execute S112 to S113. If m is greater than M, execute S115 to S116.

[0057] S115: The first concentration change rate set, the second concentration change rate set, the third concentration change rate set and the fourth concentration change rate set constitute gas characteristic data of the rth independent control partition;

[0058] S116: Let r=r+1. If r is less than or equal to R, then let m=2 and execute S112 to S114. If r is greater than R, then end the current process.

[0059] Furthermore, the method for calculating the oxygen concentration change rate includes:

[0060]

[0061] OND rn is the oxygen concentration change rate of the rth independent control partition in the nth monitoring time interval, O rm is the oxygen concentration corresponding to the rth independent control partition at the mth time point, O r ( m-1 ) is the oxygen concentration corresponding to the rth independent control partition at the m-1th time point, T m is the mth time point, T m-1 is the m-1th time point;

[0062] The method for calculating the carbon monoxide concentration change rate includes:

[0063]

[0064] COND rn is the carbon monoxide concentration change rate of the rth independent control partition in the nth monitoring time interval, CO rm is the carbon monoxide concentration corresponding to the rth independent control partition at the mth time point, CO r(m-1) is the carbon monoxide concentration corresponding to the rth independent control partition at the m-1th time point.

[0065] Furthermore, the method for calculating the nitrogen oxide concentration change rate includes:

[0066]

[0067] NOND rn is the nitrogen oxide concentration change rate of the rth independent control partition in the nth monitoring time interval, NO rm is the nitrogen oxide concentration corresponding to the rth independent control partition at the mth time point, NO r(m-1) is the nitrogen oxide concentration corresponding to the rth independent control partition at the m-1th time point;

[0068] The method for calculating the hydrocarbon concentration change rate includes:

[0069]

[0070] HCND rn is the hydrocarbon concentration change rate of the rth independent control partition in the nth monitoring time interval, HC rm is the hydrocarbon concentration corresponding to the rth independent control partition at the mth time point, HC r(m-1) is the hydrocarbon concentration corresponding to the rth independent control partition at the m-1th time point.

[0071] Furthermore, the training method of the combustion diagnosis model includes:

[0072] A combustion diagnosis data set is collected in advance, wherein the combustion diagnosis data set includes G groups of combustion diagnosis data and combustion diagnosis results corresponding to the G groups of combustion diagnosis data, where G is a positive integer greater than 0; the combustion diagnosis data includes operation characteristic data and gas characteristic data; the combustion diagnosis data set is divided into a training set and a test set, the combustion diagnosis data in the training set is used as the input of a combustion diagnosis model, and the combustion diagnosis results in the training set is used as the output of the combustion diagnosis model, with minimizing the sum of prediction accuracies of all predicted combustion diagnosis results as a training goal; training is stopped when the sum of prediction accuracies converges; the combustion diagnosis model is a random forest model or a support vector machine model.

[0073] The technical effects and advantages of the modular porous medium burner with controllable partitions of the present invention are as follows:

[0074] The modular porous media burner solution with controllable partitions provided by the present invention is achieved by dividing the burner into a plurality of independently controlled partitions, each partition being equipped with an independent fuel and air supply system, sensor, and controller; using a variety of sensors to monitor air flow, fuel flow, temperature, and gas concentrations such as oxygen and carbon monoxide in real time, combined with machine learning models, so as to timely detect problems such as incomplete combustion, local overheating, air-fuel ratio imbalance, or flameout risk. In addition, by operating characteristic data and gas characteristic data, combined with machine learning models, corresponding adjustment parameters are obtained, and the air supply, fuel supply, and burner power are adjusted through the burner control system, thereby achieving precise adaptive adjustment and ensuring stable operation of the system. Through the partition control and adaptive adjustment technology of the present invention, each independently controlled partition maintains the optimal air-fuel ratio and combustion state, thereby effectively overcoming the limitations of the prior art and achieving a more efficient, safe, and environmentally friendly combustion process. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is a schematic diagram of a modular porous medium burner system with controllable partitions according to Example 1 of the present invention;

[0076] Figure 2 A flow chart for adaptively adjusting the burners of R independently controlled partitions;

[0077] Figure 3 A flow chart of a method for obtaining operating characteristic data of R independent control partitions;

[0078] Figure 4 Flow chart of a method for acquiring gas characteristic data of R independently controlled partitions. DETAILED DESCRIPTION

[0079] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0080] Example 1

[0081] See also Figure 1 As shown, a system described in this embodiment includes a region division module, a first acquisition module, a first processing module, a second acquisition module, a second processing module, a combustion diagnosis module and an automatic adjustment module; each module realizes data transmission through wired and / or wireless connections.

[0082] The zone division module is used to divide the burner into R independent control zones.

[0083] The method for dividing the R independent control partitions includes:

[0084] The burner adopts a modular structure. The internal modules of the burner are distributed according to actual operational requirements (for example, by area or function). The internal area of ​​the burner is divided into R independent control partitions. The area of ​​each independent control partition is determined according to operational requirements.

[0085] Each independent control partition is equipped with an independent fuel pipeline and fuel control valve. By individually controlling the fuel input, it is ensured that each area can adaptively adjust the fuel flow according to its combustion conditions; each independent control partition is equipped with an independent air channel with a control valve and a fan to adjust the air flow. The control system can adjust the air flow according to the combustion requirements of the partition to ensure an ideal air-fuel ratio; each independent control partition is equipped with an independent temperature sensor, gas sensor, air flow sensor and fuel flow sensor to monitor the combustion temperature data, combustion gas data, air flow data and fuel flow data inside the partition in real time, and feed back the combustion temperature data, air flow data and fuel flow data to the main control system for analysis and adjustment.

[0086] The first acquisition module is used to acquire the combustion operation data of R independent control partitions within a unit time, wherein the combustion operation data includes air flow, fuel flow and combustion temperature. The air flow is acquired by an air flow sensor, the fuel flow is acquired by a fuel flow sensor, and the combustion temperature is acquired by a temperature sensor.

[0087] The first processing module is used to extract features from the combustion operation data to obtain operation feature data corresponding to R independent control partitions, wherein the operation feature data includes an air flow rate change rate set, a fuel flow rate change rate set and a combustion temperature change rate set.

[0088] The method for acquiring the operation characteristic data of the R independently controlled partitions includes:

[0089] S101: Preset an initial value of r to 1, and the value range of r is 1 to R; divide the unit time into M time points, preset an initial value of m to 2, and the value range of m is 2 to M; preset an initial value of n to 1, and the value range of n is M-1;

[0090] S102: The time interval between the m-1th and mth time points constitutes the nth monitoring time interval; the air flow rate change rate of the nth monitoring time interval is calculated based on the air flow rates at the m-1th and mth time points; the fuel flow rate change rate of the nth monitoring time interval is calculated based on the fuel flow rates at the m-1th and mth time points; the combustion temperature change rate of the nth monitoring time interval is calculated based on the combustion temperatures at the m-1th and mth time points;

[0091] S103: adding the air flow rate change rate to the air flow rate change rate set corresponding to the r-th independent control partition, adding the fuel flow rate change rate to the fuel flow rate change rate set corresponding to the r-th independent control partition, and adding the combustion temperature change rate to the combustion temperature change rate set corresponding to the r-th independent control partition;

[0092] S104: Let m=m+1. If m is less than or equal to M, execute S102 to S103. If m is greater than M, execute S105 to S106.

[0093] S105: The air flow rate change rate set, the fuel flow rate change rate set and the combustion temperature change rate set constitute the operation characteristic data of the rth independent control partition;

[0094] S106: Let r=r+1. If r is less than or equal to R, then let m=2 and execute S102 to S104. If r is greater than R, then end the current process.

[0095] The method for calculating the air flow rate change rate includes:

[0096]

[0097] Among them, KQBH rn KQLL is the air flow rate change rate of the rth independent control partition in the nth monitoring time interval, rm KQLL is the air flow rate corresponding to the rth independent control partition at the mth time point, r(m-1) is the air flow rate corresponding to the rth independent control partition at the m-1th time point, T m is the mth time point, T m-1 is the m-1th time point.

[0098] The method for calculating the fuel flow rate change rate includes:

[0099]

[0100] Among them, RLBH rn RLL is the fuel flow rate change rate of the rth independent control partition in the nth monitoring time interval. rmis the fuel flow rate corresponding to the rth independent control partition at the mth time point, RLL r(m-1) is the fuel flow rate corresponding to the rth independent control partition at the m-1th time point.

[0101] The method for calculating the combustion temperature change rate includes:

[0102]

[0103] Among them, RSBH rn RSWD is the combustion temperature change rate of the rth independent control partition in the nth monitoring time interval. rm RSWD is the combustion temperature corresponding to the rth independent control partition at the mth time point, r(m-1) is the combustion temperature corresponding to the rth independent control partition at the m-1th time point.

[0104] It should be noted that the air flow rate is the time derivative of the air flow change, which describes the amplitude of the air flow change. When the air flow rate is positive, it indicates that the air flow is increasing, and the data value of the air flow rate is related to the amplitude of the air flow change. The larger the data value of the air flow rate is, the greater the increase in the air flow is. Conversely, when the air flow rate is negative, it indicates that the air flow is decreasing, and the larger the data value of the air flow rate is, the greater the decrease in the air flow is. The fuel flow rate is the time derivative of the fuel flow change, which describes the amplitude of the fuel flow change. When the fuel flow rate is positive, it indicates that the fuel flow is increasing, and the data value of the fuel flow rate is related to the fuel flow change. The combustion temperature change rate is the time derivative of the combustion temperature change, which describes the amplitude of the combustion temperature change. When the combustion temperature change rate is positive, it indicates that the combustion temperature is increasing. The data value of the combustion temperature change rate is related to the amplitude of the combustion temperature change. The larger the data value of the combustion temperature change rate, the larger the amplitude of the combustion temperature increase. Conversely, when the combustion temperature change rate is negative, it indicates that the combustion temperature is decreasing. The larger the data value of the combustion temperature change rate, the larger the amplitude of the combustion temperature increase. Conversely, when the combustion temperature change rate is negative, it indicates that the combustion temperature is decreasing. The larger the data value of the combustion temperature change rate, the larger the amplitude of the combustion temperature decrease.

[0105] The second acquisition module is used to collect the monitoring gas data of R independent control partitions within a unit time, wherein the monitoring gas data includes oxygen concentration, carbon monoxide concentration, nitrogen oxide concentration and hydrocarbon concentration. The monitoring gas data is obtained through a gas sensor.

[0106] The second processing module is used to extract features from the monitored gas data to obtain gas feature data corresponding to R independent control partitions, wherein the gas feature data includes a first concentration change rate set, a second concentration change rate set, a third concentration change rate set, and a fourth concentration change rate set. The first concentration change rate set is a change rate set of oxygen concentration, the second concentration change rate set is a change rate set of carbon monoxide concentration, the third concentration change rate set is a change rate set of nitrogen oxide concentration, and the fourth concentration change rate set is a change rate set of hydrocarbons.

[0107] The method for acquiring the gas characteristic data of the R independently controlled partitions includes:

[0108] S111: Preset an initial value of r to 1, and the value range of r is 1 to R; divide the unit time into M time points, preset an initial value of m to 2, and the value range of m is 2 to M; preset an initial value of n to 1, and the value range of n is M-1;

[0109] S112: The time interval between the m-1th and mth time points constitutes the nth monitoring time interval; the oxygen concentration change rate of the nth monitoring time interval is calculated based on the oxygen concentrations at the m-1th and mth time points; the carbon monoxide concentration change rate of the nth monitoring time interval is calculated based on the carbon monoxide concentrations at the m-1th and mth time points; the nitrogen oxide concentration change rate of the nth monitoring time interval is calculated based on the nitrogen oxide concentrations at the m-1th and mth time points; the hydrocarbon concentration change rate of the nth monitoring time interval is calculated based on the hydrocarbon concentrations at the m-1th and mth time points;

[0110] S113: adding the oxygen concentration change rate to the first concentration change rate set of the rth independent control partition, adding the carbon monoxide concentration change rate to the second concentration change rate set of the rth independent control partition, adding the nitrogen oxide concentration change rate to the third concentration change rate set of the rth independent control partition; adding the hydrocarbon concentration change rate to the fourth concentration change rate set of the rth independent control partition;

[0111] S114: Let m=m+1. If m is less than or equal to M, execute S112 to S113. If m is greater than M, execute S115 to S116.

[0112] S115: The first concentration change rate set, the second concentration change rate set, the third concentration change rate set and the fourth concentration change rate set constitute gas characteristic data of the rth independent control partition;

[0113] S116: Let r=r+1. If r is less than or equal to R, then let m=2 and execute S112 to S114. If r is greater than R, then end the current process.

[0114] The method for calculating the oxygen concentration change rate includes:

[0115]

[0116] Among them, OND rn is the oxygen concentration change rate of the rth independent control partition in the nth monitoring time interval, O rm is the oxygen concentration corresponding to the rth independent control partition at the mth time point, O r(m-1 is the oxygen concentration corresponding to the rth independent control partition at the m-1th time point, T m is the mth time point, T m-1 is the m-1th time point.

[0117] The method for calculating the carbon monoxide concentration change rate includes:

[0118]

[0119] Among them, COND rn is the carbon monoxide concentration change rate of the rth independent control partition in the nth monitoring time interval, CO rm is the carbon monoxide concentration corresponding to the rth independent control partition at the mth time point, CO r(m-1) is the carbon monoxide concentration corresponding to the rth independent control partition at the m-1th time point.

[0120] The method for calculating the nitrogen oxide concentration change rate includes:

[0121]

[0122] Among them, NOND rn is the nitrogen oxide concentration change rate of the rth independent control partition in the nth monitoring time interval, NO rm is the nitrogen oxide concentration corresponding to the rth independent control partition at the mth time point, NO r(m-1) is the nitrogen oxide concentration corresponding to the rth independent control partition at the m-1th time point.

[0123] The method for calculating the hydrocarbon concentration change rate includes:

[0124]

[0125] Among them, HCND rn is the hydrocarbon concentration change rate of the rth independent control partition in the nth monitoring time interval, HC rm is the hydrocarbon concentration corresponding to the rth independent control partition at the mth time point, HC r(m-1)is the hydrocarbon concentration corresponding to the rth independent control partition at the m-1th time point.

[0126] The combustion diagnosis module is used to input the operating characteristic data and gas characteristic data of R independent control partitions into the combustion diagnosis model to obtain the combustion diagnosis results corresponding to the R independent control partitions, wherein the combustion diagnosis results include normal combustion and abnormal combustion; the abnormal combustion includes incomplete combustion, air-fuel ratio imbalance, local overheating and flameout risk.

[0127] The training method of the combustion diagnosis model comprises:

[0128] A combustion diagnosis data set is collected in advance, wherein the combustion diagnosis data set includes G groups of combustion diagnosis data and combustion diagnosis results corresponding to the G groups of combustion diagnosis data, where G is a positive integer greater than 0; the combustion diagnosis data includes operation characteristic data and gas characteristic data; the combustion diagnosis data set is divided into a training set and a test set, the combustion diagnosis data in the training set is used as the input of a combustion diagnosis model, and the combustion diagnosis results in the training set is used as the output of the combustion diagnosis model, with minimizing the sum of prediction accuracies of all predicted combustion diagnosis results as a training goal; training is stopped when the sum of prediction accuracies converges; the combustion diagnosis model is a random forest model or a support vector machine model.

[0129] It should be noted that early problems such as incomplete combustion, air-fuel ratio imbalance, local overheating and flameout can be diagnosed through operation characteristic data and gas characteristic data. The biggest advantage of early diagnosis is that it can prevent problems before they happen, avoid the occurrence of serious problems, extend equipment life, and improve operational efficiency.

[0130] For example, when the type of abnormal combustion is incomplete combustion, the air flow rate change rate and the fuel flow rate change rate remain stable. Incomplete combustion is usually accompanied by incomplete combustion and temperature drop. The value of the combustion temperature change rate is negative, showing a slowly decreasing trend. Insufficient oxygen cannot support complete combustion. The value of the oxygen concentration change rate is negative, and the oxygen concentration shows a decreasing trend. Incomplete combustion will produce more carbon monoxide. The value of the carbon monoxide concentration change rate is positive, and the carbon monoxide concentration shows a rapid increasing trend. Nitrogen oxides are generated at high temperatures. The nitrogen oxide change rate remains stable, and the nitrogen oxide concentration has no obvious changing trend. The amount of unburned fuel residue increases, the value of the hydrocarbon concentration change rate is positive, and the hydrocarbon concentration shows an increasing trend.

[0131] For example, when the type of abnormal combustion is air-fuel ratio imbalance, the changes in the air flow rate and the fuel flow rate are in an unstable state, and the changes in the fuel flow rate and the air flow rate are inconsistent. For example, after the air flow rate adjustment and the fuel flow rate adjustment, the air flow rate and the fuel flow rate cannot be well adapted, that is, the air-fuel ratio is unbalanced; the value of the combustion temperature change rate fluctuates, sometimes positive and sometimes negative, reflecting the instability of the combustion process; the oxygen concentration change rate is positively correlated with the air flow rate change rate; the air-fuel ratio imbalance also leads to a certain degree of incomplete combustion, which will produce carbon monoxide. The value of the carbon monoxide concentration change rate is positive, and the carbon monoxide concentration shows an unstable upward trend; the nitrogen oxide concentration change rate is positively correlated with the air flow rate change rate; if the value of the air flow rate change rate is positive, the value of the air flow rate change rate is large, and the nitrogen oxide concentration will also increase with the increase in combustion temperature; on the contrary, if the value of the air flow rate change rate is negative, and it does not increase with the increase in combustion temperature, the nitrogen oxide concentration will show a downward trend; due to the air-fuel ratio imbalance, the concentration of hydrocarbons fluctuates.

[0132] For example, when the type of abnormal combustion is local overheating, the change in the air flow rate is in a stable state, the fuel flow rate change value is positive, and the fuel flow increases slowly; the local fuel supply in the combustion area increases, resulting in overheating, the combustion temperature change rate value is positive, and the value is relatively large, and the combustion temperature rises rapidly; because the combustion reaction intensifies, more oxygen is consumed and the oxygen concentration decreases; the overheated area leads to incomplete combustion and the concentration of carbon monoxide increases; high temperature will intensify the formation of nitrogen oxides, that is, the concentration of nitrogen oxides shows a significant upward trend; the local overheated area will cause the concentration of hydrocarbons to fluctuate.

[0133] For example, when the type of abnormal combustion is the risk of flameout, the air flow rate decreases rapidly or fluctuates greatly, making it difficult to maintain the combustion process, losing the dynamic balance between the air and fuel flows, and the heat in the combustion area is insufficient to maintain continuous combustion, and the temperature change rate decreases rapidly; when the combustion weakens, it is accompanied by a decrease in oxygen consumption, that is, the oxygen concentration shows a downward trend; the fuel does not have enough oxygen to burn completely, resulting in an increase in the concentration of carbon monoxide, that is, the concentration of carbon monoxide shows an upward trend; the concentration of nitrogen oxides shows a rapid downward trend; unburned fuel gradually accumulates, causing the concentration of hydrocarbons to show an upward trend.

[0134] The automatic adjustment module is used to adaptively adjust the burners of the R independent control partitions according to the combustion diagnosis results corresponding to the R independent control partitions.

[0135] The method for adaptively adjusting the combustion systems of the R independently controlled partitions includes:

[0136] S121: Preset the initial value of r to 1, and the value range of r is 1 to R;

[0137] S122: Obtain the combustion diagnosis result of the rth independent control partition. If the combustion diagnosis result is normal combustion, execute S124; otherwise, execute S123;

[0138] S123: if the combustion diagnosis result is incomplete combustion, the burner is adaptively adjusted for the incomplete combustion; if the combustion diagnosis result is an air-fuel ratio imbalance, the burner is adaptively adjusted for the air-fuel ratio imbalance; if the combustion diagnosis result is local overheating, the burner is adaptively adjusted for the local overheating; if the combustion diagnosis result is a flameout risk, the burner is adaptively adjusted for the flameout risk; after the adaptive adjustment, S124 is executed;

[0139] S124: Let r=r+1. If r is less than or equal to R, continue to execute S122; if r is greater than R, end the current process.

[0140] Methods for adaptive adjustment for incomplete combustion include:

[0141] The operation characteristic data and the gas characteristic data are input into the first parameter model to obtain the predicted first parameter data, wherein the first parameter data includes the air supply, the fuel supply and the burner power; a control instruction is sent to the burner to adjust the current air supply to the predicted air supply, adjust the current fuel supply to the predicted fuel supply, and adjust the current burner power to the predicted burner power.

[0142] The training method of the first parameter model includes:

[0143] A first training data set is collected in advance, the first training data set includes H groups of first training data, and first parameter data corresponding to the H groups of first training data, where H is a positive integer greater than 0; the first training data includes operation characteristic data and gas characteristic data; technical personnel in this field adjust the air supply, fuel supply and burner power of the burner through historical operation data and actual operation conditions to obtain optimal adjustment values, thereby obtaining the first parameter data in the training data; the first training data set is divided into a first training set and a first test set, the first training data in the first training set is used as the input of the first parameter model, and the first parameter data in the first training set is used as the output of the first parameter model, so as to minimize the sum of the prediction accuracies of all predicted first parameter data as the training goal; the training is stopped until the sum of the prediction accuracies converges; the first parameter model is a decision tree model or a support vector machine model.

[0144] It should be noted that when incomplete combustion occurs, the adaptive adjustment strategy is that the burner system can increase the air supply, increase the burner power, and reduce the fuel supply, thereby increasing the combustion temperature, improving the combustion efficiency, promoting complete combustion, and reducing the occurrence of incomplete combustion, thereby improving the combustion effect.

[0145] The methods for adaptively adjusting the air-fuel ratio imbalance include:

[0146] The operation characteristic data and the gas characteristic data are input into the second parameter model to obtain predicted second parameter data, wherein the second parameter data includes air supply and fuel supply; the predicted air supply is divided by the predicted fuel supply to obtain a predicted air-fuel ratio, and a control instruction is sent to the burner to adjust the current air supply to the predicted air supply, adjust the current fuel supply to the predicted fuel supply, and adjust the current air-fuel ratio to the predicted air-fuel ratio.

[0147] The method for obtaining the air-fuel ratio includes:

[0148]

[0149] Among them, KRB is the air-fuel ratio, KQGJ is the air supply amount, and RLGJ is the fuel supply amount.

[0150] The training method of the second parameter model includes:

[0151] A second training data set is collected in advance, the second training data set includes J groups of second training data, and second parameter data corresponding to the J groups of second training data, where J is a positive integer greater than 0; the second training data includes operation characteristic data and gas characteristic data; technical personnel in this field adjust the air supply and fuel supply of the burner through historical operation data and actual operation conditions to obtain optimal adjustment values, thereby obtaining the second parameter data in the training data; the second training data set is divided into a second training set and a second test set, the second training data in the second training set is used as the input of the second parameter model, and the second parameter data in the second training set is used as the output of the second parameter model, so as to minimize the sum of the prediction accuracies of all predicted second parameter data as the training goal; the training is stopped until the sum of the prediction accuracies converges; the second parameter model is a decision tree model or a support vector machine model.

[0152] It should be noted that air-fuel ratio imbalance is caused by changes in air supply or fuel supply, which leads to a decrease in combustion efficiency. When air-fuel ratio imbalance occurs, the adaptive adjustment strategy is to adjust the intake valve or fuel supply system according to the predicted air-fuel ratio, thereby adjusting the ratio of air supply to fuel supply, so that the air-fuel ratio returns to the ideal range.

[0153] Methods for adaptive regulation of local overheating include:

[0154] The operation characteristic data and the gas characteristic data are input into the third parameter model to obtain predicted third parameter data, wherein the third parameter data includes air supply and fuel supply; a control instruction is sent to the burner to adjust the current air supply to the predicted air supply, and the current fuel supply to the predicted fuel supply.

[0155] The training method of the third parameter model includes:

[0156] A third training data set is collected in advance, the third training data set includes K groups of third training data, and third parameter data corresponding to the K groups of third training data, K is a positive integer greater than 0; the third training data includes operation characteristic data and gas characteristic data; technical personnel in this field adjust the air supply and fuel supply of the burner through historical operation data and actual operation conditions to obtain optimal adjustment values, thereby obtaining the third parameter data in the training data; the third training data set is divided into a third training set and a third test set, the third training data in the third training set is used as the input of the third parameter model, and the third parameter data in the third training set is used as the output of the third parameter model, so as to minimize the sum of the prediction accuracies of all predicted third parameter data as the training goal; the training is stopped until the sum of the prediction accuracies converges; the third parameter model is a random forest model.

[0157] It should be noted that local overheating is caused by excessive combustion intensity in certain areas of the burner, and the heat cannot be evenly dispersed in time, causing the local temperature to exceed the normal range. Local overheating will not only damage the burner components, but may also cause unstable combustion, reduce system efficiency, and even bring safety hazards. When local overheating occurs, the adaptive adjustment strategy is to increase the air supply to help reduce the combustion temperature, improve combustion efficiency, adjust the fuel supply system, reduce the fuel supply, and ensure uniform fuel distribution.

[0158] The methods for adaptively adjusting the risk of flameout include:

[0159] The operation characteristic data and the gas characteristic data are input into the fourth parameter model to obtain predicted fourth parameter data, wherein the fourth parameter data includes air supply and burner power; a control instruction is sent to the burner to adjust the current air supply to the predicted air supply and the current burner power to the predicted burner power.

[0160] The training method of the fourth parameter model includes:

[0161] A fourth training data set is collected in advance, the fourth training data set includes L groups of fourth training data, and fourth parameter data corresponding to the L groups of fourth training data, where L is a positive integer greater than 0; the fourth training data includes operation characteristic data and gas characteristic data; technical personnel in this field adjust the air supply and burner power of the burner through historical operation data and actual operation conditions to obtain optimal adjustment values, thereby obtaining the fourth parameter data in the training data; the fourth training data set is divided into a fourth training set and a fourth test set, the fourth training data in the fourth training set is used as the input of the fourth parameter model, and the fourth parameter data in the fourth training set is used as the output of the fourth parameter model, so as to minimize the sum of the prediction accuracies of all predicted fourth parameter data as the training goal; the training is stopped until the sum of the prediction accuracies converges; the fourth parameter model is a naive Bayes model or a gradient boosting tree model.

[0162] It should be noted that when the fault type is flameout risk, it is usually because the combustion conditions in the burner cannot maintain the minimum standard required for stable combustion. This fault will affect the continuous operation of the equipment and may even cause safety problems. When flameout risk occurs, the adaptive adjustment strategy is to adjust the combustion temperature by adjusting the air supply and adjusting the burner power to maintain stable combustion.

[0163] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0164] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A modular porous medium burner with controllable partitions, characterized in that: include: The zone division module is used to divide the burner into R independent control zones; A first acquisition module is used to acquire combustion operation data of R independent control partitions per unit time, wherein the combustion operation data includes air flow, fuel flow and combustion temperature; The first processing module is used to extract features from the combustion operation data to obtain operation feature data corresponding to R independent control partitions; A second acquisition module is used to collect monitoring gas data of R independent control partitions per unit time, wherein the monitoring gas data includes oxygen concentration, carbon monoxide concentration, nitrogen oxide concentration and hydrocarbon concentration; The second processing module is used to extract features from the monitored gas data to obtain gas feature data corresponding to R independent control partitions; A combustion diagnosis module, used for inputting the operation characteristic data and gas characteristic data of the R independent control partitions into the combustion diagnosis model to obtain the combustion diagnosis results corresponding to the R independent control partitions; The automatic adjustment module is used to adaptively adjust the burners of the R independent control partitions according to the combustion diagnosis results corresponding to the R independent control partitions.

2. A modular porous medium burner with controllable zones according to claim 1, characterized in that: The method for adaptively adjusting the combustion systems of the R independently controlled partitions includes: S121: Preset the initial value of r to 1, and the value range of r is 1 to R; S122: Obtain the combustion diagnosis result of the rth independent control partition. If the combustion diagnosis result is normal combustion, execute S124; otherwise, execute S123; S123: if the combustion diagnosis result is incomplete combustion, the burner is adaptively adjusted for the incomplete combustion; if the combustion diagnosis result is an air-fuel ratio imbalance, the burner is adaptively adjusted for the air-fuel ratio imbalance; if the combustion diagnosis result is local overheating, the burner is adaptively adjusted for the local overheating; if the combustion diagnosis result is a flameout risk, the burner is adaptively adjusted for the flameout risk; after the adaptive adjustment, S124 is executed; S124: Let r=r+1. If r is less than or equal to R, continue to execute S122; if r is greater than R, end the current process.

3. A modular porous medium burner with controllable zones according to claim 2, characterized in that: Methods for adaptive adjustment for incomplete combustion include: Inputting the operation characteristic data and the gas characteristic data into the first parameter model to obtain predicted first parameter data, wherein the first parameter data includes air supply, fuel supply and burner power; sending a control instruction to the burner to adjust the current air supply to the predicted air supply, the current fuel supply to the predicted fuel supply, and the current burner power to the predicted burner power; The training method of the first parameter model includes: A first training data set is collected in advance, the first training data set includes H groups of first training data, and first parameter data corresponding to the H groups of first training data, where H is a positive integer greater than 0; the first training data includes operation characteristic data and gas characteristic data; the first training data set is divided into a first training set and a first test set, the first training data in the first training set is used as the input of the first parameter model, and the first parameter data in the first training set is used as the output of the first parameter model, with minimizing the sum of the prediction accuracies of all predicted first parameter data as the training goal; the training is stopped when the sum of the prediction accuracies converges; the first parameter model is a decision tree model or a support vector machine model.

4. A modular porous medium burner with controllable zones according to claim 3, characterized in that: The methods for adaptively adjusting the air-fuel ratio imbalance include: Inputting the operation characteristic data and the gas characteristic data into the second parameter model to obtain predicted second parameter data, wherein the second parameter data includes an air supply amount and a fuel supply amount; dividing the predicted air supply amount by the predicted fuel supply amount to obtain a predicted air-fuel ratio, and sending a control instruction to the burner to adjust the current air supply amount to the predicted air supply amount, the current fuel supply amount to the predicted fuel supply amount, and the current air-fuel ratio to the predicted air-fuel ratio; The training method of the second parameter model includes: A second training data set is collected in advance, the second training data set includes J groups of second training data, and second parameter data corresponding to the J groups of second training data, where J is a positive integer greater than 0; the second training data includes operation characteristic data and gas characteristic data; the second training data set is divided into a second training set and a second test set, the second training data in the second training set is used as the input of the second parameter model, and the second parameter data in the second training set is used as the output of the second parameter model, with minimizing the sum of the prediction accuracies of all predicted second parameter data as the training goal; the training is stopped when the sum of the prediction accuracies converges; the second parameter model is a decision tree model or a support vector machine model.

5. A modular porous medium burner with controllable zones according to claim 4, characterized in that: Methods for adaptive regulation of local overheating include: Inputting the operation characteristic data and the gas characteristic data into the third parameter model to obtain predicted third parameter data, wherein the third parameter data includes an air supply amount and a fuel supply amount; sending a control instruction to the burner to adjust the current air supply amount to the predicted air supply amount, and adjust the current fuel supply amount to the predicted fuel supply amount; The training method of the third parameter model includes: A third training data set is collected in advance, the third training data set includes K groups of third training data, and third parameter data corresponding to the K groups of third training data, K is a positive integer greater than 0; the third training data includes operation characteristic data and gas characteristic data; the third training data set is divided into a third training set and a third test set, the third training data in the third training set is used as the input of the third parameter model, and the third parameter data in the third training set is used as the output of the third parameter model, so as to minimize the sum of the prediction accuracies of all predicted third parameter data as the training goal; the training is stopped when the sum of the prediction accuracies converges; the third parameter model is a random forest model.

6. A modular porous medium burner with controllable zones according to claim 5, characterized in that: The methods for adaptively adjusting the risk of flameout include: Inputting the operation characteristic data and the gas characteristic data into the fourth parameter model to obtain predicted fourth parameter data, wherein the fourth parameter data includes air supply and burner power; sending a control instruction to the burner to adjust the current air supply to the predicted air supply and the current burner power to the predicted burner power; The training method of the fourth parameter model includes: A fourth training data set is collected in advance, the fourth training data set includes L groups of fourth training data, and fourth parameter data corresponding to the L groups of fourth training data, where L is a positive integer greater than 0; the fourth training data includes operation characteristic data and gas characteristic data; the fourth training data set is divided into a fourth training set and a fourth test set, the fourth training data in the fourth training set is used as the input of a fourth parameter model, and the fourth parameter data in the fourth training set is used as the output of the fourth parameter model, so as to minimize the sum of the prediction accuracies of all predicted fourth parameter data as a training goal; the training is stopped when the sum of the prediction accuracies converges; the fourth parameter model is a naive Bayes model or a gradient boosting tree model.

7. The modular porous medium burner with controllable zones according to claim 1, characterized in that: The method for acquiring the operation characteristic data of the R independently controlled partitions includes: S101: Preset an initial value of r to 1, and the value range of r is 1 to R; divide the unit time into M time points, preset an initial value of m to 2, and the value range of m is 2 to M; preset an initial value of n to 1, and the value range of n is M-1; S102: The time interval between the m-1th and mth time points constitutes the nth monitoring time interval; the air flow rate change rate of the nth monitoring time interval is calculated based on the air flow rates at the m-1th and mth time points; the fuel flow rate change rate of the nth monitoring time interval is calculated based on the fuel flow rates at the m-1th and mth time points; the combustion temperature change rate of the nth monitoring time interval is calculated based on the combustion temperatures at the m-1th and mth time points; S103: adding the air flow rate change rate to the air flow rate change rate set corresponding to the r-th independent control partition, adding the fuel flow rate change rate to the fuel flow rate change rate set corresponding to the r-th independent control partition, and adding the combustion temperature change rate to the combustion temperature change rate set corresponding to the r-th independent control partition; S104: Let m=m+1. If m is less than or equal to M, execute S102 to S103. If m is greater than M, execute S105 to S106. S105: The air flow rate change rate set, the fuel flow rate change rate set and the combustion temperature change rate set constitute the operation characteristic data of the rth independent control partition; S106: Let r=r+1. If r is less than or equal to R, then let m=2 and execute S102 to S104. If r is greater than R, then end the current process.

8. A modular porous medium burner with controllable zones according to claim 7, characterized in that: The method for calculating the air flow rate change rate includes: QUR rn KQLL is the air flow rate change rate of the rth independent control partition in the nth monitoring time interval, rm is the air flow rate corresponding to the rth independent control partition at the mth time point, KQLL r(m-1 is the air flow rate corresponding to the rth independent control partition at the m-1th time point, T m is the mth time point, T m-1 is the m-1th time point; The method for calculating the fuel flow rate change rate includes: RLBH rn RLL is the fuel flow rate change rate of the rth independent control partition in the nth monitoring time interval. rm is the fuel flow rate corresponding to the rth independent control partition at the mth time point, RLL r(m-1 is the fuel flow rate corresponding to the rth independent control partition at the m-1th time point; The method for calculating the combustion temperature change rate includes: RSBH rn RSWD is the combustion temperature change rate of the rth independent control partition in the nth monitoring time interval. rm RSWD is the combustion temperature corresponding to the rth independent control partition at the mth time point, r(m-1) is the combustion temperature corresponding to the rth independent control partition at the m-1th time point.

9. The modular porous medium burner with controllable zones according to claim 1, characterized in that: The method for acquiring the gas characteristic data of the R independently controlled partitions includes: S111: Preset an initial value of r to 1, and the value range of r is 1 to R; divide the unit time into M time points, preset an initial value of m to 2, and the value range of m is 2 to M; preset an initial value of n to 1, and the value range of n is M-1; S112: The time interval between the m-1th and mth time points constitutes the nth monitoring time interval; the oxygen concentration change rate of the nth monitoring time interval is calculated based on the oxygen concentrations at the m-1th and mth time points; the carbon monoxide concentration change rate of the nth monitoring time interval is calculated based on the carbon monoxide concentrations at the m-1th and mth time points; the nitrogen oxide concentration change rate of the nth monitoring time interval is calculated based on the nitrogen oxide concentrations at the m-1th and mth time points; the hydrocarbon concentration change rate of the nth monitoring time interval is calculated based on the hydrocarbon concentrations at the m-1th and mth time points; S113: adding the oxygen concentration change rate to the first concentration change rate set of the rth independent control partition, adding the carbon monoxide concentration change rate to the second concentration change rate set of the rth independent control partition, adding the nitrogen oxide concentration change rate to the third concentration change rate set of the rth independent control partition; adding the hydrocarbon concentration change rate to the fourth concentration change rate set of the rth independent control partition; S114: Let m=m+1. If m is less than or equal to M, execute S112 to S113. If m is greater than M, execute S115 to S116. S115: The first concentration change rate set, the second concentration change rate set, the third concentration change rate set and the fourth concentration change rate set constitute gas characteristic data of the rth independent control partition; S116: Let r=r+1. If r is less than or equal to R, then let m=2 and execute S112 to S114. If r is greater than R, then end the current process.

10. The modular porous medium burner with controllable zones according to claim 1, characterized in that: The training method of the combustion diagnosis model comprises: A combustion diagnosis data set is collected in advance, wherein the combustion diagnosis data set includes G groups of combustion diagnosis data and combustion diagnosis results corresponding to the G groups of combustion diagnosis data, where G is a positive integer greater than 0; the combustion diagnosis data includes operation characteristic data and gas characteristic data; the combustion diagnosis data set is divided into a training set and a test set, the combustion diagnosis data in the training set is used as the input of a combustion diagnosis model, and the combustion diagnosis results in the training set is used as the output of the combustion diagnosis model, with minimizing the sum of prediction accuracies of all predicted combustion diagnosis results as a training goal; training is stopped when the sum of prediction accuracies converges; the combustion diagnosis model is a random forest model or a support vector machine model.

Images