Method for controlling the combustion process of a to furnace

Abstract

The present application relates to the technical field of data preprocessing, in particular to a TO furnace combustion process control method; when the present application preprocesses the TO furnace combustion state data, the state vector of coupling Kalman filtering is established; the state equation and the observation equation of coupling Kalman filtering are established; the combustion chamber temperature data, the fuel supply data and the combustion air flow data are subjected to coupling Kalman filtering according to the state vector, the state equation and the observation equation; the physical coupling relationship of the state equation is utilized to realize multivariable joint filtering, which not only utilizes the optimal estimation advantage of Kalman filtering, but also solves the problem of ignoring process linkage in traditional single parameter filtering through physical coupling.

Description

Technical Field

[0001] This invention relates to the field of data preprocessing technology, specifically a method for controlling the combustion process of a TO furnace. Background Technology

[0002] TO furnaces are environmental protection devices that decompose VOCs and other organic pollutants through high-temperature oxidation. Their core principle is to heat waste gas containing organic matter to 700-1000℃, causing the organic matter to react with oxygen to produce carbon dioxide and water. Simultaneously, the heat generated during combustion is recovered to achieve energy savings. The stability and precision of the TO furnace combustion process directly affect treatment efficiency (typically requiring a VOCs removal rate ≥95%), energy consumption (natural gas / fuel usage), and pollutant emissions (such as the control of secondary pollutants like NOx and CO).

[0003] In the combustion process of a TO furnace, combustion process control is generally achieved by monitoring the combustion operation status data of the TO furnace. Since it is necessary to monitor multiple operating status data, data preprocessing is required for the TO furnace combustion operation status data. However, for the TO furnace operation status data, the various data are interdependent. For example, the combustion chamber temperature is closely related to the fuel supply and combustion air flow rate, and there is a complex nonlinear relationship between the combustion chamber temperature and VOCs concentration. Various types of data may change significantly in a short period of time. For example, the combustion chamber temperature may change rapidly due to changes in fuel composition or fluctuations in combustion air flow rate, and the exhaust gas flow rate and VOCs concentration may fluctuate rapidly due to changes in upstream processes. Traditional data preprocessing operations cannot achieve good results. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for controlling the combustion process of a TO furnace, which solves the problems existing in the prior art.

[0005] This invention provides a method for controlling the combustion process of a TO furnace, comprising the following steps:

[0006] S1: Real-time acquisition of the TO furnace operating status data via sensor array;

[0007] S2: Perform data preprocessing on the TO furnace operating status data; S2 includes: S2.1: Perform coupled Kalman filtering on the combustion chamber temperature data, fuel supply data, and combustion air flow data; specifically:

[0008] S211: Establish the state vector of the coupled Kalman filter;

[0009] S212: Establish the state equation and observation equation of the coupled Kalman filter;

[0010] S213: Perform coupled Kalman filtering on combustion chamber temperature data, fuel supply data, and combustion air flow data based on the state vector, state equation, and observation equation;

[0011] S2.2: Perform transient time detection based on the TO furnace operating status data, and dynamically synchronize the TO furnace operating status data in time;

[0012] S3: Control the combustion process of the TO furnace based on the pre-processed TO furnace operating status data.

[0013] Preferably, the state vector X(k) of the coupled Kalman filter is:

[0014] ;

[0015] In the formula, k represents time, F(k) is the fuel supply at time k, A(k) is the combustion air flow at time k, and T(k) is the combustion chamber temperature at time k.

[0016] Preferably, the specific formula of the state equation of the coupled Kalman filter is as follows:

[0017]

[0018] In the formula, F( k -1) represents the fuel supply at time k-1, and A(k-1) represents the combustion air flow rate at time k-1. opt (k-1) represents the optimal combustion air quantity at time k-1, and T(k-1) represents the combustion chamber temperature at time k-1. For process noise in fuel supply data, The process noise for combustion airflow data, For combustion chamber temperature data, process noise, This refers to the nominal calorific value of the fuel. For the theoretical air-fuel ratio, α β is the thermal efficiency conversion coefficient, and β is the excess air correction term coefficient;

[0019] The observation equation of the coupled Kalman filter Z ( k The specific formula is:

[0020] ;

[0021] In the formula, H is the observation matrix, F sensor (k) represents the sensor measurement of fuel supply data, A sensor (k) represents the sensor measurement value of the combustion air flow data, T sensor (k) represents the sensor measurement value of the combustion chamber temperature data, vk For observation noise, X(k) is the state vector of the coupled Kalman filter, v k To observe noise.

[0022] Preferably, the coupled Kalman filter achieves joint estimation of state variables (F, A, T) through an extended Kalman filter framework, and the specific steps are as follows:

[0023] Based on the state estimate from the previous moment Covariance The prior estimate at the current time is calculated using the state equation. with prior covariance ;

[0024] The sensor measurements of fuel supply, combustion air flow, and combustion chamber temperature at the current moment are obtained using the Kalman gain K(k) and the prior estimate. with prior covariance After making corrections, the posterior estimate is obtained. and posterior covariance Wherein, the posterior estimate is filtered data of the combustion chamber temperature data, fuel supply data, and combustion air flow data.

[0025] Preferably, the calculation process of the Kalman gain K(k) is as follows:

[0026] The formula for calculating the Kalman gain K(k) is based on the covariance propagation of the state equation and the noise statistics of the observation equation, specifically:

[0027] ;

[0028] In the formula, Let H(k) be the prior covariance, and H(k) be the Jacobian matrix of the observation equation. R ( k ) represents the observation noise covariance.

[0029] Preferably, S2.2 specifically includes:

[0030] S221: Detect the transient state of combustion in the TO furnace;

[0031] S222: Dynamically synchronize the operating status data of the TO furnace.

[0032] Preferably, in step S221, the transient state start time and duration of combustion chamber temperature data, exhaust gas VOCs concentration data, and exhaust gas flow data are identified in real time through short-time change rate monitoring, absolute change amount monitoring, and dynamic threshold determination.

[0033] Preferably, the monitoring data selected for transient states include: combustion chamber temperature, exhaust gas VOCs concentration, and exhaust gas flow rate; the monitoring indicators include short-term change rate and absolute change amount;

[0034] The short-time rate of change Slope i The formula for calculating (k) is:

[0035] ;

[0036] In the formula, This refers to the value of monitoring data i at time k, where i=1 indicates the monitoring data is the combustion chamber temperature, i=2 indicates the monitoring data is the VOCs concentration in the exhaust gas, and i=3 indicates the monitoring data is the exhaust gas flow rate. n Δ is the size of the sliding window. t The sampling interval;

[0037] The absolute change Δ i ( k The formula for calculating ) is:

[0038] ;

[0039] The dynamic threshold is adaptively determined based on the historical quantiles of the monitoring data; specifically, it is calculated based on the steady-state operating data of the past 24 hours, showing the 95th percentile change rate for each monitoring data point. ) and the absolute change of the 95th percentile The dynamic threshold is defined as:

[0040] ;

[0041] ;

[0042] In the formula, Threshold Slope,i The short-term rate of change dynamic threshold, The dynamic threshold for absolute change. a and b This is a conservative coefficient;

[0043] The transient state is determined based on the dynamic threshold.

[0044] Specifically, when the short-time rate of change Slope i (k), the absolute change Δ i ( k If either of the two indicators meets the following conditions, it is considered a transient state;

[0045] ;

[0046] ;

[0047] After the transient state is identified, the start time of the transient state is automatically recorded. t start The system continues to monitor the indicators until they fall back to within the steady-state threshold, at which point the transient ends. t end This yields the transient time period.

[0048] Preferably, the dynamic time synchronization of the TO furnace operating status data specifically involves:

[0049] When the combustion state of the TO furnace is in a steady state, the low-frequency operating state data is linearly interpolated based on the highest sampling frequency to generate a time series aligned with the highest frequency parameter.

[0050] When the combustion state of the TO furnace is in a transient state, the operating state data involving this transient state is interpolated using a high-frequency interpolation method, and the synchronization rhythm is dynamically adjusted based on causal time delay.

[0051] The high-frequency interpolation method is as follows: the sampling frequency of the monitoring data during the transient period is increased from the original frequency to 10Hz through cubic spline interpolation; the dynamic adjustment of the synchronization rhythm based on causal delay is as follows: the physical coupling delay between the running status data is analyzed, and the time axis of the delayed running status data is fine-tuned during synchronization.

[0052] The embodiments of the present invention have the following technical effects:

[0053] In this embodiment, when preprocessing the combustion state data of the TO furnace, a state vector for coupled Kalman filtering is established; a state equation and an observation equation for coupled Kalman filtering are established; coupled Kalman filtering is performed on the combustion chamber temperature data, fuel supply data, and combustion air flow data based on the state vector, state equation, and observation equation; multi-variable joint filtering is achieved by utilizing the physical coupling relationship of the state equation, which not only takes advantage of the optimal estimation of Kalman filtering, but also solves the problem of neglecting process linkage in traditional single-parameter filtering through physical coupling.

[0054] Meanwhile, by detecting transient states in real time, dynamically adjusting synchronization strategies, and preserving key change details, high-fidelity, rhythm-matched multi-source data is provided for subsequent feature extraction and control strategy generation.

[0055] Meanwhile, by using historical quantile adaptive thresholds and multi-parameter correlation analysis, the transient start and end times under different operating conditions can be accurately identified, avoiding misjudgments based on fixed thresholds. At the same time, the steady-state and transient periods are distinguished, and strategies such as low-frequency steady-state interpolation, high-frequency transient interpolation, and causal delay compensation are adopted to preserve the details of rapid changes in monitoring data, while ensuring the alignment of the physical meaning of multi-source data. Attached Figure Description

[0056] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0057] Figure 1 This is a flowchart of a TO furnace combustion process control method provided in an embodiment of the present invention;

[0058] Figure 2 This is a flowchart of a data preprocessing operation for TO furnace operating status data provided in an embodiment of the present invention;

[0059] Figure 3 This is a flowchart of coupled Kalman filtering of combustion chamber temperature data, fuel supply data, and combustion air flow data provided in an embodiment of the present invention;

[0060] Figure 4 This is a flowchart provided by an embodiment of the present invention for performing transient time detection based on the TO furnace operating status data and dynamically synchronizing the TO furnace operating status data in time. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0062] Example 1, Appendix Figure 1 A flowchart of a TO furnace combustion process control method is shown in the attached diagram. Figure 1 As shown, a method for controlling the combustion process of a TO furnace includes the following steps:

[0063] S1: Real-time acquisition of the TO furnace operating status data via sensor array;

[0064] The sensor group includes an infrared thermometer, a differential pressure flow meter, a PID sensor, a gas flow meter, an air flow meter, and a flue gas analyzer; the operating status data includes combustion chamber temperature data, exhaust gas flow data, exhaust gas VOCs concentration data, fuel supply data, combustion air flow data, and flue gas composition analysis data; the flue gas composition analysis data includes nitrogen oxide data, carbon monoxide data, and carbon dioxide data.

[0065] Furthermore, the infrared thermometer is arranged in the combustion chamber, with 3-5 measuring points positioned axially at the upper, middle, and lower levels, covering the preheating zone, main reaction zone, and afterburning zone, to acquire combustion chamber temperature data; the differential pressure flow meter is arranged in the upstream straight pipe section to acquire exhaust gas flow data; the PID sensor is a photoionization detector with a range of 0-5000ppm and a response time of less than 3s, to acquire the VOCs concentration data of the exhaust gas input to the TO furnace; the gas flow meter is a turbine gas flow meter with a built-in temperature / pressure sensor for automatic temperature and pressure compensation, to acquire fuel supply data; the air flow meter is used to acquire combustion air flow data; and the flue gas analyzer is a chemiluminescence analyzer to acquire flue gas composition analysis data.

[0066] S2: Perform data preprocessing on the TO furnace operating status data;

[0067] For TO furnace operating status data, the data are interconnected, which leads to poor preprocessing results in existing data preprocessing methods because they do not take into account the strong physical coupling and transient change characteristics of TO furnace operating status data.

[0068] Based on the above situation, this embodiment proposes a preprocessing method for TO furnace operating status data to improve the quality and usability of TO furnace operating status data; specifically, as follows... Figure 2 As shown, S2 includes:

[0069] S2.1: Perform coupled Kalman filtering on combustion chamber temperature data, fuel supply data, and combustion air flow data;

[0070] In traditional filtering, combustion chamber temperature data, fuel supply data, and combustion air flow data are generally treated as independent parameters and filtered separately, ignoring the dynamic relationship between the three. Therefore, this embodiment is based on the energy balance relationship between fuel supply data (F), combustion air flow data (A), and combustion chamber temperature data (T), and performs coupled Kalman filter (CKF) processing on the fuel supply data (F), combustion air flow data (A), and combustion chamber temperature data (T).

[0071] like Figure 3 As shown, the coupled Kalman filter is applied to the combustion chamber temperature data, fuel supply data, and combustion air flow data as follows:

[0072] S211: Establish the state vector of the coupled Kalman filter;

[0073] The state vector of the coupled Kalman filter is:

[0074] ;

[0075] In the formula, k represents time, F(k) is the fuel supply at time k, A(k) is the combustion air flow at time k, and T(k) is the combustion chamber temperature at time k.

[0076] S212: Establish the state equation and observation equation of the coupled Kalman filter;

[0077] The dynamic evolution of combustion chamber temperature data, fuel supply data, and combustion air flow data is described based on the principles of energy balance and mass conservation. The state equation and observation equation of the coupled Kalman filter are used to demonstrate the coupled logic that changes in fuel flow directly affect the releaseable energy, and that air flow regulates energy release efficiency through the air-fuel ratio, ultimately determining temperature changes.

[0078] The specific formula for the state equation of the coupled Kalman filter is as follows:

[0079]

[0080] In the formula, F( k -1) represents the fuel supply at time k-1, and A(k-1) represents the combustion air flow rate at time k-1. opt (k-1) represents the optimal combustion air quantity at time k-1, and T(k-1) represents the combustion chamber temperature at time k-1. For process noise in fuel supply data, The process noise for combustion airflow data, For combustion chamber temperature data, process noise, This refers to the nominal calorific value of the fuel. For the theoretical air-fuel ratio, α β is the thermal efficiency conversion coefficient, and β is the excess air correction term coefficient;

[0081] The observation equation of the coupled Kalman filter Z ( k The specific formula is:

[0082] ;

[0083] In the formula, H is the observation matrix. In this embodiment, the observation matrix is ​​a 3×a matrix, where... a This is the dimension of the state vector X(k). Each row of the observation matrix corresponds to an observation value from a sensor, and each column corresponds to the weight of a state variable. The function of the observation matrix is ​​to map the state vector X(k) to the observation space, thereby relating it to the sensor observation equation. Z ( kEstablish contact, F sensor (k) represents the sensor measurement of fuel supply data, A sensor (k) represents the sensor measurement value of the combustion air flow data, T sensor (k) represents the sensor measurement value of the combustion chamber temperature data, v k Observation noise represents the random factor that causes the observed value to deviate from the true value due to sensor measurement errors and other reasons.

[0084] S213: Perform coupled Kalman filtering on combustion chamber temperature data, fuel supply data, and combustion air flow data based on the state vector, state equation, and observation equation;

[0085] The coupled Kalman filter achieves joint estimation of state variables (F, A, T) through an extended Kalman filter framework. The specific steps are as follows:

[0086] Based on the state estimate from the previous moment Covariance The prior estimate at the current time is calculated using the state equation. with prior covariance In this step, the dynamic coupling relationship between F, A, and T is preserved through the above state equations;

[0087] The sensor measurements of fuel supply, combustion air flow, and combustion chamber temperature at the current moment are obtained using the Kalman gain K(k) and the prior estimate. with prior covariance After making corrections, the posterior estimate is obtained. and posterior covariance The posterior estimate is the final state estimate that integrates the model prior and the measured values, which is the filtered data of the combustion chamber temperature data, fuel supply data, and combustion air flow data.

[0088] The calculation process for the Kalman gain K(k) is as follows:

[0089] The formula for calculating the Kalman gain K(k) is based on the covariance propagation of the state equation and the noise statistics of the observation equation, specifically:

[0090] ;

[0091] In the formula, Let H(k) be the prior covariance, and H(k) be the Jacobian matrix of the observation equation. R ( k ) represents the observation noise covariance.

[0092] Traditional Kalman filtering often only considers "self-noise" when processing single-parameter or few-parameter data. However, in actual industrial processes, combustion chamber temperature data, fuel supply data, and combustion air flow data are strongly coupled. For example, if the fuel changes, the air may need to be adjusted, and the temperature will also change accordingly. Therefore, the coupled Kalman filtering in this step utilizes the physical coupling relationship of the state equation to achieve multi-variable joint filtering. This not only takes advantage of the optimal estimation of Kalman filtering, but also solves the problem of neglecting process linkage in traditional single-parameter filtering through physical coupling.

[0093] S2.1 further includes performing conventional filtering operations on other data.

[0094] S2.2: Perform transient time detection based on the TO furnace operating status data, and dynamically synchronize the TO furnace operating status data in time.

[0095] During the combustion process in a TO furnace, combustion chamber temperature, exhaust gas VOCs concentration, and exhaust gas flow rate data may change significantly within a short period (e.g., 10-30 seconds) due to factors such as upstream process fluctuations and changes in fuel composition. For example, the temperature may change by ±80℃ / minute, or the VOCs concentration may suddenly increase from 500 mg / m³ to 2000 mg / m³. Traditional multi-source data preprocessing methods typically employ fixed sampling frequencies and interpolation strategies, such as linearly interpolating all parameters based on the highest frequency of 2 Hz. However, these methods do not consider the transient coupling characteristics and differences in the dynamic change rhythms between parameters. This leads to problems such as the transient processes of sudden increases in VOCs concentration or abrupt changes in fuel composition being treated as noise and filtered out, or the loss of details due to low-frequency interpolation, and the asynchronous transient change rhythms of different parameters.

[0096] This embodiment proposes a transient state detection and dynamic time synchronization method. By detecting transient states in real time, dynamically adjusting the synchronization strategy, and preserving key change details, it provides high-fidelity, rhythm-matched multi-source data for subsequent feature extraction and control strategy generation.

[0097] Specifically, such as Figure 4 As shown, S2.2 specifically includes:

[0098] S221: Detect the transient state of combustion in the TO furnace;

[0099] The core characteristic of transient states is that the TO furnace combustion monitoring data undergoes drastic changes that exceed the normal range within a short period of time. This step identifies the start time and duration of transient states in combustion chamber temperature data, exhaust gas VOCs concentration data, exhaust gas flow rate data, etc., in real time through short-term change rate monitoring, absolute change amount monitoring, and dynamic threshold determination.

[0100] The transient monitoring data selected include: combustion chamber temperature, exhaust gas VOCs concentration, and exhaust gas flow rate; for each of the monitoring data, two monitoring indicators are set, including short-term change rate and absolute change amount;

[0101] The short-time rate of change Slope i (k) is used to reflect the instantaneous rate of change of the monitoring data, and the calculation formula is:

[0102] ;

[0103] In the formula, This refers to the value of monitoring data i at time k, where i=1 indicates the monitoring data is the combustion chamber temperature, i=2 indicates the monitoring data is the VOCs concentration in the exhaust gas, and i=3 indicates the monitoring data is the exhaust gas flow rate. n Δ is the size of the sliding window. t The sampling interval;

[0104] The absolute change Δ i ( k This is used to reflect the cumulative change in the parameter, and the calculation formula is:

[0105] .

[0106] The dynamic threshold is adaptively determined based on the historical quantiles of the monitoring data; specifically, it is calculated based on the steady-state operating data of the past 24 hours, using the 95th quantile change rate of each monitoring data point. ) and the absolute change of the 95th percentile The dynamic threshold is defined as:

[0107] ;

[0108] ;

[0109] In the formula, Threshold Slope,i The short-term rate of change dynamic threshold, The dynamic threshold for absolute change. a and b This is a conservative coefficient used to ensure that only significant anomalous changes are captured.

[0110] The transient state is determined based on the dynamic threshold.

[0111] Specifically, when the short-time rate of change Slope i (k), the absolute change Δ i ( k If either of the two indicators meets the following conditions, it is considered a transient state;

[0112] ;

[0113] ;

[0114] After the transient state is identified, the start time of the transient state is automatically recorded. t start The system continues to monitor the indicators until they fall back to within the steady-state threshold, at which point the transient ends. t end This yields the transient time period, which is used for subsequent adjustments to the dynamic synchronization strategy.

[0115] S222: Dynamically synchronize the operating status data of the TO furnace over time;

[0116] Traditional methods use the highest sampling frequency as a benchmark to perform linear interpolation on all parameters. However, this forced synchronization ignores the rhythm differences in transient changes between parameters. This embodiment proposes a dynamic synchronization strategy of steady-state low-frequency interpolation + transient high-frequency interpolation to ensure that the key change details of each parameter are preserved, while realizing the physical meaning alignment of multi-source data.

[0117] Specifically, the dynamic time synchronization of the TO furnace operating status data involves:

[0118] When the combustion state of the TO furnace is in a steady state, the low-frequency operating state data is linearly interpolated based on the highest sampling frequency to generate a time series aligned with the highest frequency parameter.

[0119] The interpolation formula is as follows: For the missing value at time t, it is calculated using two sampling points (t1, x1) and (t2, x2) before and after the missing value, and the formula is:

[0120] ;

[0121] In the formula, t1 and t2 are the two sampling times before and after the missing time, and x1 and x2 are the two sampled values ​​before and after the missing value.

[0122] When the combustion state of the TO furnace is in a transient state, the operating state data involving this transient state is interpolated using a high-frequency interpolation method, and the synchronization rhythm is dynamically adjusted based on causal time delay.

[0123] The high-frequency interpolation method involves increasing the sampling frequency of the transient monitoring data from the original frequency to 10Hz using cubic spline interpolation. Cubic spline interpolation can generate smooth midpoints, better preserve the nonlinear characteristics of the transient curve, and avoid inflection point distortion caused by linear interpolation.

[0124] Specifically, the dynamic adjustment of synchronization rhythm based on causal delay involves analyzing the physical coupling delay between operating status data and fine-tuning the time axis of delayed operating status data during synchronization. For example, if the VOCs concentration in the exhaust gas starts to rise sharply at t=0 seconds, while the temperature only responds at t=10 seconds, then during transient synchronization, the starting time of the temperature is marked as t=10 seconds to ensure that the causal logic of the change rhythm of the two is consistent.

[0125] By using historical quantile adaptive thresholds and multi-parameter correlation analysis, the transient start and end times under different operating conditions are accurately identified, avoiding misjudgments based on fixed thresholds. At the same time, the steady-state and transient periods are distinguished, and strategies such as low-frequency steady-state interpolation, high-frequency transient interpolation, and causal delay compensation are adopted to preserve the details of rapid changes in monitoring data, while ensuring the alignment of the physical meaning of multi-source data.

[0126] S3: Control the combustion process of the TO furnace based on the pre-processed TO furnace operating status data.

[0127] Specifically, S3 is as follows: the combustion process of the TO furnace is controlled by adopting a dual-loop control and feedforward compensation architecture; wherein, the dual-loop control includes a temperature control loop and an air-fuel ratio control loop; the temperature control loop takes the combustion chamber temperature (T) as the main control target and maintains the temperature stable near the set value through feedback adjustment; the air-fuel ratio control loop takes maintaining the optimal air-fuel ratio (λ) as the target and adjusts the fuel and air ratio through a combination of feedforward and feedback.

[0128] Example 2: The present invention also provides an electronic device, including one or more processors and a memory.

[0129] A processor can be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and can control other components in an electronic device to perform desired functions.

[0130] The memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and a processor may execute the program instructions to implement a TO furnace combustion process control method described above in any embodiment of this application, and / or other desired functions. Various contents such as initial external parameters and thresholds may also be stored in the computer-readable storage medium.

[0131] In addition to the methods and devices described above, embodiments of this application may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the function of a TO furnace combustion process control method provided in any embodiment of this application.

[0132] Furthermore, embodiments of this application may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to implement a TO furnace combustion process control method provided in any embodiment of this application.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling the combustion process of a TO furnace, characterized in that, Includes the following steps: S1: Real-time acquisition of the TO furnace operating status data via sensor array; S2: Perform data preprocessing on the TO furnace operating status data; S2 includes: S2.1: Perform coupled Kalman filtering on combustion chamber temperature data, fuel supply data, and combustion air flow data; specifically: S211: Establish the state vector of the coupled Kalman filter; S212: Establish the state equation and observation equation of the coupled Kalman filter; S213: Perform coupled Kalman filtering on combustion chamber temperature data, fuel supply data, and combustion air flow data based on the state vector, state equation, and observation equation; S2.2: Perform transient time detection based on the TO furnace operating status data, and dynamically synchronize the TO furnace operating status data in time; S3: Control the combustion process of the TO furnace based on the pre-processed TO furnace operating status data; Specifically, S3 involves using a dual-loop control and feedforward compensation architecture to control the combustion process of the TO furnace. The dual-loop control includes a temperature control loop and an air-fuel ratio control loop. The temperature control loop takes the combustion chamber temperature (T) as the main control target and maintains the temperature stable near the set value through feedback adjustment. The air-fuel ratio control loop aims to maintain the optimal air-fuel ratio (λ) and adjusts the fuel-air ratio through a combination of feedforward and feedback.

2. The TO furnace combustion process control method according to claim 1, characterized in that: The state vector X(k) of the coupled Kalman filter is: ； In the formula, k represents time, F(k) is the fuel supply at time k, A(k) is the combustion air flow at time k, and T(k) is the combustion chamber temperature at time k.

3. The TO furnace combustion process control method according to claim 1, characterized in that: The specific formula for the state equation of the coupled Kalman filter is as follows: ； In the formula, F( k 1) represents the fuel supply at time k-1, and A(k-1) represents the combustion air flow rate at time k-1. opt (k 1) represents the optimal combustion air quantity at time k-1, and T(k-1) represents the combustion chamber temperature at time k-1. For process noise in fuel supply data, The process noise for combustion airflow data, For combustion chamber temperature data, process noise, This refers to the nominal calorific value of the fuel. For the theoretical air-fuel ratio, α β is the thermal efficiency conversion coefficient, and β is the excess air correction term coefficient; The observation equation of the coupled Kalman filter Z ( k The specific formula is: ； In the formula, H is the observation matrix, which is a 3×a matrix, where a is the state vector. The dimension, Fs ensor (k) represents the sensor measurement of fuel supply data, A sensor (k) represents the sensor measurement value of the combustion air flow data, T sensor (k) represents the sensor measurement value of the combustion chamber temperature data. Let be the state vector of the coupled Kalman filter. To observe noise.

4. The TO furnace combustion process control method according to claim 1, characterized in that: The coupled Kalman filter achieves joint estimation of state variables through an extended Kalman filter framework, and its specific steps are as follows: Based on the state estimate from the previous moment Covariance The prior estimate at the current time is calculated using the state equation. with prior covariance ; The sensor measurements of fuel supply, combustion air flow, and combustion chamber temperature at the current moment are obtained using the Kalman gain K(k) and the prior estimate. with prior covariance After making corrections, the posterior estimate is obtained. and posterior covariance Wherein, the posterior estimate is filtered data of the combustion chamber temperature data, fuel supply data, and combustion air flow data.

5. The TO furnace combustion process control method according to claim 4, characterized in that: The calculation process for the Kalman gain K(k) is as follows: The formula for calculating the Kalman gain K(k) is based on the covariance propagation of the state equation and the noise statistics of the observation equation, specifically: ； In the formula, Let H(k) be the prior covariance, and H(k) be the Jacobian matrix of the observation equation. R ( k ) represents the observation noise covariance.

6. The TO furnace combustion process control method according to claim 1, characterized in that: Specifically, S2.2 is as follows: S221: Detect the transient state of combustion in the TO furnace; S222: Dynamically synchronize the operating status data of the TO furnace.

7. The TO furnace combustion process control method according to claim 6, characterized in that: In S221, the transient state start time and duration of combustion chamber temperature data, exhaust gas VOCs concentration data, and exhaust gas flow data are identified in real time through short-time change rate monitoring, absolute change amount monitoring, and dynamic threshold determination.

8. The TO furnace combustion process control method according to claim 7, characterized in that: The monitoring data selected for transient states include: combustion chamber temperature, exhaust gas VOCs concentration, and exhaust gas flow rate; monitoring indicators include short-term change rate and absolute change amount; The short-time rate of change Slope i The formula for calculating (k) is: ； In the formula, This refers to the value of monitoring data i at time k, where i=1 indicates the monitoring data is the combustion chamber temperature, i=2 indicates the monitoring data is the VOCs concentration in the exhaust gas, and i=3 indicates the monitoring data is the exhaust gas flow rate. n Δ is the size of the sliding window. t The sampling interval; The absolute change Δ i ( k The formula for calculating ) is: ； The dynamic threshold is adaptively determined based on the historical quantiles of the monitoring data; specifically, it calculates the 95th quantile change rate for each monitoring data point based on the steady-state operating data of the past 24 hours. absolute change of the 95th percentile The dynamic threshold is defined as: ； ； In the formula, Threshold Slope,i The short-term rate of change dynamic threshold, The dynamic threshold for absolute change. a and b This is a conservative coefficient; The transient state is determined based on the dynamic threshold. Specifically, when the short-time rate of change Slope i (k), the absolute change Δ i ( k If either of the two indicators meets the following conditions, it is considered a transient state; ； ； After the transient state is identified, the start time of the transient state is automatically recorded. t start The system continues to monitor the indicators until they fall back to within the steady-state threshold, at which point the transient ends. t end This yields the transient time period.

9. The TO furnace combustion process control method according to claim 6, characterized in that: The specific steps for dynamically synchronizing the TO furnace operating status data are as follows: When the combustion state of the TO furnace is in a steady state, the low-frequency operating state data is linearly interpolated based on the highest sampling frequency to generate a time series aligned with the highest frequency parameter. When the combustion state of the TO furnace is in a transient state, the operating state data involving this transient state is interpolated using a high-frequency interpolation method, and the synchronization rhythm is dynamically adjusted based on causal time delay. The high-frequency interpolation method is as follows: the sampling frequency of the monitoring data during the transient period is increased from the original frequency to 10Hz through cubic spline interpolation; the dynamic adjustment of the synchronization rhythm based on causal delay is as follows: the physical coupling delay between the running status data is analyzed, and the time axis of the delayed running status data is fine-tuned during synchronization.

Images