A multivariable anti-coupling cooperative control system for heating furnace

Abstract

The present invention relates to the technical field of burners, and specifically discloses a multivariable anti-coupling collaborative control system for a heating furnace, which is used to solve the problems caused by the coupling of heavy oil preheating and atomizing nozzle pressure in a heating furnace burner of a steel plant using heavy oil as fuel. The multivariable anti-coupling collaborative control system comprises a central control module, a heavy oil preheating control system, an atomizing nozzle control system, a coupling effect identification module, a data monitoring module and a feedback adjustment module, and proposes a coupling effect identification formula in the coupling effect identification module. The present invention effectively manages the coupling effect by dynamically adjusting the heavy oil preheating temperature and the nozzle pressure, ensures that the fuel is fully atomized, improves combustion efficiency, and reduces equipment wear and maintenance costs.

Description

Technical Field

[0001] The present invention relates to the technical field of burners, and more particularly to a multivariable anti-coupling cooperative control system for a heating furnace. Background Art

[0002] In the heating furnace of the steel plant, the burner using heavy oil as fuel needs to ensure that the fuel is fully atomized before entering the combustion chamber to achieve ideal combustion efficiency and stability. Due to its high viscosity, heavy oil needs to be heated to a suitable temperature through the fuel preheating system to reduce its viscosity and facilitate atomization through the nozzle system. However, there is a complex coupling relationship between the fuel preheating temperature and the nozzle pressure, and this coupling effect directly affects the overall performance of the burner. By heating the heavy oil to a specific temperature, its viscosity is reduced, so that it can be atomized into fine droplets when passing through the nozzle. The nozzle system relies on a specific pressure to atomize the preheated heavy oil into fine droplets, and the atomization effect directly affects the sufficiency and stability of combustion. The preheating temperature of the fuel and the pressure of the nozzle need to be precisely matched. Too high a preheating temperature requires reducing the nozzle pressure to avoid excessive atomization of the fuel, resulting in unstable combustion; too low a preheating temperature requires increasing the nozzle pressure to enhance the atomization effect. However, the adjustment of the preheating temperature and the nozzle pressure often affects each other. If you are not careful, it may trigger a series of coupling effects, resulting in system instability and affecting the operating efficiency and safety of the heating furnace. If the fuel preheating temperature is too low and the fuel viscosity is too high, it will be difficult to achieve sufficient atomization even if the nozzle pressure is increased, which will lead to incomplete combustion, increased fuel consumption, and more pollutant emissions. When the fuel preheating temperature is too high and the nozzle pressure is not reduced accordingly, the fuel will be over-atomized, resulting in unstable flames in the combustion chamber and the risk of local high temperatures or flame extinction. When the fuel flow changes rapidly, the response speeds of the fuel preheating system and the nozzle system will be out of sync, resulting in a mismatch between the fuel temperature and the nozzle pressure, further aggravating the coupling effect and affecting combustion stability. Since the coupling effect is not managed in a timely and effective manner, the burner equipment will be subjected to uneven pressure and temperature distribution, resulting in increased wear, shortened equipment life, and increased costs. Summary of the invention

[0003] In order to overcome the above-mentioned defects of the prior art, the present invention provides an anti-coupling collaborative control system for a heating furnace, which effectively manages the coupling effect by dynamically adjusting the heavy oil preheating temperature and the nozzle pressure, ensures sufficient atomization of the fuel, improves combustion efficiency, and reduces equipment wear and maintenance costs.

[0004] During a production cycle of a steel plant, an urgent order is received, and the load of the heating furnace needs to be increased quickly. The heavy oil flow rate of the burner needs to be greatly increased. In order to maintain the ideal combustion efficiency, the heavy oil preheating temperature and the atomizing nozzle pressure also need to be adjusted accordingly. However, with the rapid increase of the heavy oil flow rate, the response speed of the heavy oil preheating system lags, resulting in the heavy oil preheating temperature temporarily unable to keep up with the increase rate of the flow rate. In this case, the nozzle pressure needs to be adjusted quickly to avoid unstable combustion in the burner. Due to the untimely preheating response, the heavy oil viscosity is high, and the atomization effect is poor, which will lead to incomplete combustion, increase pollutant emissions and heavy oil consumption. If the nozzle pressure is not adjusted in time, excessive pressure will lead to excessive atomization, resulting in unstable flames or even flameout. Too low pressure will make it impossible to fully atomize the high-viscosity heavy oil. In the rapid changes of flow, temperature and pressure, the coupling effect cannot be managed in time and effectively, which will lead to system instability, aggravate the mechanical wear of the atomizing nozzle due to internal friction and pressure load, shorten its service life, and increase maintenance costs. In order to cope with the above application scenarios and achieve the above purposes, the present invention provides the following technical solutions:

[0005] The anti-coupling collaborative control system of the heating furnace includes a central control module, a heavy oil preheating control system, an atomizing nozzle control system, a coupling effect identification module, a data monitoring module and a feedback adjustment module. The heavy oil preheating control system is connected to the atomizing nozzle control system, the data monitoring module is connected to the preheating control system, the atomizing nozzle control system is connected to the data monitoring module, the data monitoring module is connected to the coupling effect identification module, the coupling effect identification module is connected to the central control module, the central control module is connected to the heavy oil preheating control system and the atomizing nozzle control system, the heavy oil preheating control system and the atomizing nozzle control system are connected to the feedback adjustment module, and the feedback adjustment module Connected to the central control module and the data monitoring module, the coupling effect identification module takes the difference between the integral variable and the instantaneous variable of the nozzle pressure to the heavy oil preheating temperature as the variable of the piecewise function, and integrates the piecewise function to capture the instantaneous coupling effect, and then obtains the second-order derivative of the heavy oil preheating temperature to the nozzle pressure to capture the nonlinear change of the coupling effect. Then, the coupling effect caused by the change of the heavy oil flow rate is captured by integrating the sine function of the product of the integral variable and the first-order derivative of the heavy oil flow rate to the heavy oil preheating temperature in the interval from 0 to the heavy oil flow rate. The three coupling effects are weighted and summed to obtain the coupling effect index. The identification formula of the coupling effect index is:

[0006] ;

[0007] Where: For the current moment, is the integration variable, is a piecewise function, hour, , hour, , is the nozzle pressure at the current moment, is the heavy oil preheating temperature at the current moment, is the heavy oil flow rate, , , is the weight coefficient, which is adjusted according to historical data and actual working conditions.

[0008] Item is used to capture the nozzle pressure Preheating temperature of heavy oil By calculating the derivative of the rate of change of pressure with respect to temperature, the system is able to identify how the temperature responds when the nozzle pressure changes rapidly and the instantaneous effect of this response on the system. For the derivative When approaching a critical value (usually zero), the changes at that moment are magnified. This is useful for capturing coupling effects that occur instantaneously, such as when changes in temperature are very sensitive to changes in pressure.

[0009] Used to capture heavy oil preheat temperature Nozzle pressure The nonlinear coupling effect caused by the second derivative of reflects the nonlinear relationship between temperature and pressure, especially in high-order changes. For example, when a pressure change causes an acceleration or deceleration of the temperature response, this item can identify and quantify this nonlinear effect. By calculating the second-order derivative of temperature with respect to pressure, the system is able to identify nonlinear coupling effects, i.e., how the acceleration (or deceleration) of the temperature response when pressure changes affects the stability of the overall system.

[0010] Used to capture heavy oil flows Preheating temperature of heavy oil This term reflects the effect of flow rate changes on temperature, especially when the fuel flow rate changes rapidly, how the relationship between flow rate and temperature affects combustion stability. Sine function It is used to capture the periodic or oscillatory coupling effects caused by flow changes. By integrating over the interval of , we can obtain the overall impact of flow changes on the system coupling effect.

[0011] As a further solution of the present invention, the central control module predicts the future coupling effect index through the identification value of the coupling effect index based on the model predictive control algorithm, dynamically adjusts the nozzle pressure and heavy oil preheating temperature by combining the coupling effect index prediction value and the comprehensive reward function value, and optimizes the control strategy of heavy oil preheating and nozzle pressure in real time. The formula for obtaining the current optimal control strategy is:

[0012] ;

[0013] Where: For the future optimal control strategy, including and The adjustment amount, To predict the time step, set it according to actual needs. To minimize the objective function, is the penalty coefficient, which is set according to actual needs. For the future time step The predicted value of the coupling effect index is is the comprehensive reward function value.

[0014] As a further solution of the present invention, in the central control module, the specific implementation process of predicting the future coupling effect index through the coupling effect index at the current moment based on the model prediction algorithm is as follows:

[0015] Step 11, data acquisition and preprocessing: the central control module obtains the nozzle pressure, heavy oil preheating temperature, and heavy oil flow rate in real time through the data monitoring module, obtains the coupling effect index at the current moment from the coupling effect identification module, and filters and time-aligns the collected data;

[0016] Step 12, data storage and analysis: store historical data within a set time range, analyze the historical data, identify the correlation and change pattern between various variables based on the time series analysis method, and obtain the trend and periodicity in the historical data;

[0017] Step 13, prediction by model predictive control algorithm: based on the data of current and historical moments, a long short-term memory network prediction model of the coupling effect index is established, the model is trained and optimized, and the nozzle pressure, heavy oil preheating temperature, and heavy oil flow rate at the current moment are used as inputs, and the coupling effect index after the future time step is predicted through the prediction model;

[0018] Step 14, dynamic adjustment and feedback: by comparing the predicted coupling effect index with the actual coupling effect index, the model prediction accuracy is evaluated, and the model parameters and structure are continuously optimized through the feedback mechanism;

[0019] Step 15, model calibration and optimization: Regularly calibrate the model prediction algorithm, adjust model parameters, retrain the machine learning model, and switch to a more appropriate model structure.

[0020] As a further solution of the present invention, in the central control module, the comprehensive reward function is a weighted sum of the reward function values ​​of fuel utilization and pollutant emission optimization, the reward function values ​​of combustion stability and safety, and the reward function values ​​of rapid response to high load demand. The formula of the comprehensive reward function is:

[0021] ;

[0022] Where: is the reward function value for optimizing the fuel utilization rate and pollutant emissions at the current moment, which is equal to the ratio of the effective heavy oil utilization rate to the target heavy oil utilization rate minus the pollutant emissions. The combustion stability and safety reward function value is equal to the inverse of the sum of the temperature fluctuation rate, pressure fluctuation rate and safety risk index in the combustion chamber. The safety risk index is obtained by identifying and quantifying risk factors. The reward function value for quickly responding to high-load demands is equal to the inverse of the sum of the response time and the response time overshoot. The response time overshoot is the ratio of the difference between the maximum response time and the target response time to the target response time. , , is the weight adjustment coefficient, which is adjusted according to the actual working conditions.

[0023] As a further solution of the present invention, in the central control module, the process of obtaining the security risk index includes:

[0024] Step 21, data collection: real-time collection of temperature data of the combustion chamber and key components, pressure change data of the combustion chamber, pipeline and nozzle, intensity, shape and color data of the flame, flow data of heavy oil and air, and alarm log and fault record of the system;

[0025] Step 22, risk factor identification and quantification: identify the overheating risk, overpressure risk, flame instability risk, and flow abnormality risk in the data, and quantify the risk value;

[0026] Step 23, risk weight allocation: After identifying and quantifying each risk factor, risk weights are allocated based on historical safety events and accident analysis, combined with expert knowledge and experience, and according to the current operating environment;

[0027] Step 24, safety risk index calculation: considering the quantitative value of each risk factor and its weight, calculate the safety risk index.

[0028] As a further solution of the present invention, in the central control module, the overheating risk in step 22 is that the temperature exceeds the preset safety threshold, the overpressure risk is that the system pressure exceeds the designed pressure, the flame instability risk is that the flame goes out or swings, and the flow abnormality risk is that the heavy oil or air flow deviates from the normal working range.

[0029] As a further solution of the present invention, the heavy oil preheating control system includes a heater, a heat exchanger, a temperature controller, a circulation pump, a safety valve, a filter, an oil pipe and a flow control valve. The heavy oil passes through the filter to filter out impurities and enters the heat exchanger through the pipeline. In the heat exchanger, the heavy oil exchanges heat with the heat medium and is heated. The circulation pump drives the heat medium to circulate in the heat exchanger. The temperature controller automatically adjusts the output power of the heater according to the control instructions of the central control module. The safety valve is used to release excess pressure. The flow control valve controls the heavy oil flow through the control instructions of the central control module.

[0030] As a further solution of the present invention, an atomizing nozzle control system includes a nozzle, a nozzle pressure regulating device, a flow controller, an air supply system, a nozzle cleaning device and a feedback controller, and the nozzle is connected to the nozzle pressure regulating device, the flow controller, the air supply system, the nozzle cleaning device and the feedback controller.

[0031] As a further solution of the present invention, the implementation process of the system for heavy oil preheating and atomizing nozzle pressure anti-coupling control is as follows:

[0032] Step 31, data collection and preprocessing: The data monitoring module collects the heavy oil preheating temperature, nozzle pressure and heavy oil flow in real time, filters and performs noise reduction on the collected data, and the preprocessed data is sent to the coupling effect identification module;

[0033] Step 32, coupling effect identification: the coupling effect identification module uses the piecewise function and the difference between the instantaneous rate of change of the nozzle pressure to the heavy oil preheating temperature to integrate, so as to capture the instantaneous coupling effect, and identifies the nonlinear change in the coupling effect by calculating the second-order derivative of the heavy oil preheating temperature to the nozzle pressure, and captures the coupling effect caused by the flow change by integrating the sine function of the integral variable and the first-order derivative of the heavy oil flow to the heavy oil preheating temperature in the interval from 0 to the heavy oil flow, and weights and sums the coupling effect results obtained above according to the weight coefficient to obtain the coupling effect index, and feeds the index back to the central control module;

[0034] Step 33, model prediction and control strategy optimization: The central control module uses the long short-term memory network model based on the model predictive control algorithm, uses the current and historical data to predict the coupling effect index at future moments, combines the coupling effect index prediction value and the comprehensive reward function value, calculates the current control strategy reward value, and solves the objective function , determine the optimal control strategy at the future moment, including the adjustment amount of nozzle pressure and heavy oil preheating temperature;

[0035] Step 34, control strategy execution and feedback adjustment: The central control module issues instructions to adjust the corresponding parameters in the heavy oil preheating system and the atomizing nozzle control system according to the optimal control strategy. The temperature controller automatically adjusts the power output of the heater, and the flow control valve adjusts the heavy oil flow to ensure that the preheating temperature reaches the optimal state; at the same time, the nozzle pressure is adjusted to match the optimized preheating temperature. The data monitoring module continuously monitors the response of the system, including the new nozzle pressure, heavy oil preheating temperature and flow, and feeds back to the central control module in real time. The central control module compares the new real-time data with the predicted value to evaluate the control effect. If there is a deviation, the system adjusts the control parameters again through the feedback mechanism to optimize the control strategy;

[0036] Step 35, model calibration and optimization: Regularly evaluate the accuracy of model predictions, analyze error distribution, identify room for model improvement, adjust the parameters or structure of the long short-term memory network model based on actual operating results, retrain the model when necessary to ensure that it adapts to the actual operating conditions of the system, and update the control strategy algorithm after model calibration.

[0037] In order to cope with the above-mentioned application scenarios and solve the above-mentioned technical problems, the technical effects of the present invention are as follows: the present invention analyzes the complex relationship between the fuel preheating temperature and the nozzle pressure in real time through the coupling effect identification module. The system can timely capture and identify the coupling effects caused by instantaneous coupling effects, nonlinear changes and fuel flow changes, and then calculate the coupling effect index. The central control module dynamically adjusts the nozzle pressure and heavy oil preheating temperature according to the index in combination with historical data and actual operating conditions to ensure the ideal atomization effect of the fuel and avoid the occurrence of problems such as incomplete combustion and unstable flame. Through the coordinated control of the system, not only the combustion efficiency of the heating furnace is improved and pollutant emissions are reduced, but also the equipment wear and life shortening caused by the coupling effect are effectively prevented, and the equipment maintenance cost is reduced, thereby ensuring the efficient and stable operation of the steel plant production. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 is a system block diagram of the present invention;

[0039] Figure 2 is a system flow chart of the present invention;

[0040] Figure 3 This is a system block diagram of a heavy oil preheating control system according to the present invention;

[0041] Figure 4 The present invention is a system block diagram of the atomizing nozzle control system. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the present invention to clearly and completely describe the technical solutions in the present invention. Obviously, the described content is only a part of the present invention, not all of it. Based on the content of the present invention, all other technical solutions obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0043] like Figure 1 As shown, a multivariable anti-coupling collaborative control system for a heating furnace proposed by the present invention includes a central control module, a heavy oil preheating control system, an atomizing nozzle control system, a coupling effect identification module, a data monitoring module and a feedback adjustment module, the heavy oil preheating control system is connected to the atomizing nozzle control system, the data monitoring module is connected to the coupling effect identification module, the coupling effect identification module is connected to the central control module, the central control module is connected to the heavy oil preheating control system and the atomizing nozzle control system, the heavy oil preheating control system and the atomizing nozzle control system are connected to the feedback adjustment module The feedback adjustment module is connected to the central control module and the data monitoring module. The coupling effect identification module uses the difference between the integral variable and the instantaneous variable of the nozzle pressure to the heavy oil preheating temperature as the variable of the piecewise function, and integrates the piecewise function to capture the instantaneous coupling effect, and then obtains the second-order derivative of the heavy oil preheating temperature to the nozzle pressure to capture the nonlinear change of the coupling effect. Then, the sine function of the product of the integral variable and the first-order derivative of the heavy oil flow rate to the heavy oil preheating temperature is integrated in the interval from 0 to the heavy oil flow rate to capture the coupling effect caused by the change of the heavy oil flow rate. The three coupling effects are weighted and summed to obtain the coupling effect index. The identification formula of the coupling effect index is:

[0044] ;

[0045] Where: For the current moment, is the integration variable, is a piecewise function, hour, , hour, , is the nozzle pressure at the current moment, is the heavy oil preheating temperature at the current moment, is the heavy oil flow rate, , , is the weight coefficient, which is adjusted according to historical data and actual working conditions.

[0046] The present invention analyzes the complex relationship between the fuel preheating temperature and the nozzle pressure in real time through the coupling effect identification module. The system can timely capture and identify the coupling effects caused by instantaneous coupling effects, nonlinear changes and fuel flow changes, and then calculate the coupling effect index. The central control module dynamically adjusts the nozzle pressure and heavy oil preheating temperature according to the index in combination with historical data and actual operating conditions to ensure the ideal atomization effect of the fuel and avoid the occurrence of problems such as incomplete combustion and unstable flame. Through the coordinated control of the system, not only the combustion efficiency of the heating furnace is improved and pollutant emissions are reduced, but also the equipment wear and life shortening caused by the coupling effect are effectively prevented, and the equipment maintenance cost is reduced, thereby ensuring the efficient and stable operation of the steel plant production.

[0047] It should be noted that the central control module predicts the future coupling effect index through the identification value of the coupling effect index based on the model predictive control algorithm, and dynamically adjusts the nozzle pressure and heavy oil preheating temperature by combining the coupling effect index prediction value and the comprehensive reward function value, and optimizes the control strategy of heavy oil preheating and nozzle pressure in real time. The formula for obtaining the current optimal control strategy is:

[0048] ;

[0049] Where: For the future optimal control strategy, including and The adjustment amount, To predict the time step, set it according to actual needs. To minimize the objective function, is the penalty coefficient, which is set according to actual needs. For the future time step The predicted value of the coupling effect index is is the comprehensive reward function value.

[0050] The central control module predicts the coupling effect index based on the model predictive control algorithm, and combined with the comprehensive reward function value, the system can dynamically adjust the nozzle pressure and heavy oil preheating temperature, and optimize the control strategy in real time. This method effectively responds to the coupling effect caused by the rapid change of fuel flow, ensures the stability and combustion efficiency of the burner, reduces fuel consumption and equipment wear, and thus improves the efficiency and reliability of the overall operation of the heating furnace.

[0051] It should be noted that in the central control module, the specific implementation process of predicting the future coupling effect index through the coupling effect index at the current moment based on the model prediction algorithm is as follows:

[0052] Step 11, data acquisition and preprocessing: the central control module obtains the nozzle pressure, heavy oil preheating temperature, and heavy oil flow rate in real time through the data monitoring module, obtains the coupling effect index at the current moment from the coupling effect identification module, and filters and time-aligns the collected data;

[0053] Step 12, data storage and analysis: store historical data within a set time range, analyze the historical data, identify the correlation and change pattern between various variables based on the time series analysis method, and obtain the trend and periodicity in the historical data;

[0054] Step 13, prediction by model predictive control algorithm: based on the data of current and historical moments, a long short-term memory network prediction model of the coupling effect index is established, the model is trained and optimized, and the nozzle pressure, heavy oil preheating temperature, and heavy oil flow rate at the current moment are used as inputs, and the coupling effect index after the future time step is predicted through the prediction model;

[0055] Step 14, dynamic adjustment and feedback: by comparing the predicted coupling effect index with the actual coupling effect index, the model prediction accuracy is evaluated, and the model parameters and structure are continuously optimized through the feedback mechanism;

[0056] Step 15, model calibration and optimization: Regularly calibrate the model prediction algorithm, adjust model parameters, retrain the machine learning model, and switch to a more appropriate model structure.

[0057] Through the implementation process of the central control module based on the model prediction algorithm, the system can accurately predict the future coupling effect index and dynamically adjust the nozzle pressure and heavy oil preheating temperature according to the prediction results. Through real-time data acquisition and preprocessing, historical data analysis, establishment and optimization of prediction models, as well as dynamic feedback and calibration, the system can effectively manage the coupling effect during heavy oil preheating and atomization. This process ensures that the heating furnace can operate stably under the condition of rapid changes in fuel flow, optimizes combustion efficiency, reduces fuel consumption and equipment wear, and improves the overall operating efficiency and reliability of the system.

[0058] It should be noted that in the central control module, the comprehensive reward function is the weighted sum of the reward function values ​​for fuel utilization and pollutant emission optimization, the reward function values ​​for combustion stability and safety, and the reward function values ​​for rapid response to high load demand. The formula for the comprehensive reward function is:

[0059] ;

[0060] Where: is the reward function value for optimizing the fuel utilization rate and pollutant emissions at the current moment, which is equal to the ratio of the effective heavy oil utilization rate to the target heavy oil utilization rate minus the pollutant emissions. The combustion stability and safety reward function value is equal to the inverse of the sum of the temperature fluctuation rate, pressure fluctuation rate and safety risk index in the combustion chamber. The safety risk index is obtained by identifying and quantifying risk factors. The reward function value for quickly responding to high-load demands is equal to the inverse of the sum of the response time and the response time overshoot. The response time overshoot is the ratio of the difference between the maximum response time and the target response time to the target response time. , , is the weight adjustment coefficient, which is adjusted according to the actual working conditions.

[0061] Through the comprehensive reward function in the central control module, the system can quickly respond to high-load demands while optimizing fuel utilization, reducing pollutant emissions, and ensuring combustion stability and safety. The comprehensive reward function weights and integrates key factors such as fuel utilization, pollutant emissions, combustion stability, safety, and response speed, and dynamically adjusts the weight coefficient according to actual operating conditions to achieve balance and optimization of different control objectives. In this way, the system can not only improve combustion efficiency and reduce environmental pollution, but also quickly adjust operating parameters under high load conditions to ensure the stability and safety of the heating furnace, and ultimately improve the overall operating efficiency and reliability of the system.

[0062] It should be noted that in the central control module, the process of obtaining the security risk index includes:

[0063] Step 21, data collection: real-time collection of temperature data of the combustion chamber and key components, pressure change data of the combustion chamber, pipeline and nozzle, intensity, shape and color data of the flame, flow data of heavy oil and air, and alarm log and fault record of the system;

[0064] Step 22, risk factor identification and quantification: identify the overheating risk, overpressure risk, flame instability risk, and flow abnormality risk in the data, and quantify the risk value;

[0065] Step 23, risk weight allocation: After identifying and quantifying each risk factor, risk weights are allocated based on historical safety events and accident analysis, combined with expert knowledge and experience, and according to the current operating environment;

[0066] Step 24, safety risk index calculation: considering the quantitative value of each risk factor and its weight, calculate the safety risk index.

[0067] Through the acquisition process of the safety risk index in the central control module, the system can monitor and evaluate the potential safety risks in the combustion process in real time. Through data collection and real-time analysis, the system identifies and quantifies risk factors such as overheating, overpressure, flame instability and flow abnormality, and finally calculates the safety risk index by combining historical data and expert experience to assign weights. This index provides a reliable safety assessment basis for the central control module, enabling the system to adjust operating parameters in time before instability or abnormal conditions occur during the combustion process, prevent possible safety accidents, ensure the safe and stable operation of the heating furnace under high-load production conditions, and reduce equipment failures and accident rates.

[0068] It should be noted that in the central control module, the overheating risk in step 22 is that the temperature exceeds the preset safety threshold, the overpressure risk is that the system pressure exceeds the designed pressure, the flame instability risk is that the flame goes out or swings, and the flow abnormality risk is that the heavy oil or air flow deviates from the normal working range.

[0069] It should be noted that if Figure 3 As shown, the heavy oil preheating control system includes a heater, a heat exchanger, a temperature controller, a circulation pump, a safety valve, a filter, an oil pipe and a flow control valve. The heavy oil passes through the filter to filter out impurities and enters the heat exchanger through the pipeline. In the heat exchanger, the heavy oil exchanges heat with the heat medium and is heated. The circulation pump drives the heat medium to circulate in the heat exchanger. The temperature controller automatically adjusts the output power of the heater according to the control command of the central control module. The safety valve is used to release excess pressure. The flow control valve controls the heavy oil flow through the control command of the central control module.

[0070] It should be noted that if Figure 4 As shown, the atomizing nozzle control system includes a nozzle, a nozzle pressure regulating device, a flow controller, an air supply system, a nozzle cleaning device and a feedback controller, and the nozzle is connected to the nozzle pressure regulating device, the flow controller, the air supply system, the nozzle cleaning device and the feedback controller.

[0071] It should be noted that if Figure 2 As shown, the implementation process of the system for heavy oil preheating and atomizing nozzle pressure anti-coupling control is as follows:

[0072] Step 31, data collection and preprocessing: The data monitoring module collects the heavy oil preheating temperature, nozzle pressure and heavy oil flow in real time, filters and performs noise reduction on the collected data, and the preprocessed data is sent to the coupling effect identification module;

[0073] Step 32, coupling effect identification: the coupling effect identification module uses the piecewise function and the difference between the instantaneous rate of change of the nozzle pressure to the heavy oil preheating temperature to integrate, so as to capture the instantaneous coupling effect, and identifies the nonlinear change in the coupling effect by calculating the second-order derivative of the heavy oil preheating temperature to the nozzle pressure, and captures the coupling effect caused by the flow change by integrating the sine function of the integral variable and the first-order derivative of the heavy oil flow to the heavy oil preheating temperature in the interval from 0 to the heavy oil flow, and weights and sums the coupling effect results obtained above according to the weight coefficient to obtain the coupling effect index, and feeds the index back to the central control module;

[0074] Step 33, model prediction and control strategy optimization: The central control module uses the long short-term memory network model based on the model predictive control algorithm, uses the current and historical data to predict the coupling effect index at future moments, combines the coupling effect index prediction value and the comprehensive reward function value, calculates the current control strategy reward value, and solves the objective function , determine the optimal control strategy at the future moment, including the adjustment amount of nozzle pressure and heavy oil preheating temperature;

[0075] Step 34, control strategy execution and feedback adjustment: The central control module issues instructions to adjust the corresponding parameters in the heavy oil preheating system and the atomizing nozzle control system according to the optimal control strategy. The temperature controller automatically adjusts the power output of the heater, and the flow control valve adjusts the heavy oil flow to ensure that the preheating temperature reaches the optimal state; at the same time, the nozzle pressure is adjusted to match the optimized preheating temperature. The data monitoring module continuously monitors the response of the system, including the new nozzle pressure, heavy oil preheating temperature and flow, and feeds back to the central control module in real time. The central control module compares the new real-time data with the predicted value to evaluate the control effect. If there is a deviation, the system adjusts the control parameters again through the feedback mechanism to optimize the control strategy;

[0076] Step 35, model calibration and optimization: Regularly evaluate the accuracy of model predictions, analyze error distribution, identify room for model improvement, adjust the parameters or structure of the long short-term memory network model based on actual operating results, retrain the model when necessary to ensure that it adapts to the actual operating conditions of the system, and update the control strategy algorithm after model calibration.

[0077] Through this implementation process, the system can effectively manage the complex coupling effect between the heavy oil preheating temperature and the nozzle pressure. The process of steps 31 to 35 ensures that the system can accurately collect and process key data in a high-load production environment, identify coupling effects in real time, and predict future coupling trends. The central control module dynamically adjusts the heavy oil preheating temperature and nozzle pressure through model prediction and control strategy optimization to maximize combustion efficiency and enhance system stability. At the same time, through feedback mechanisms and model calibration, the system can continuously optimize control strategies, adapt to changes in actual working conditions, and reduce the occurrence of problems such as incomplete combustion and unstable flames, thereby improving the overall operating efficiency of the heating furnace, reducing equipment wear and maintenance costs, and ensuring the efficiency and safety of the production process.

[0078] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0079] Finally: The above description is only the technical solution 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 multivariable anti-coupling cooperative control system for a heating furnace, comprising a central control module, a heavy oil preheating control system and an atomizing nozzle control system, characterized in that: It also includes a coupling effect identification module, a data monitoring module and a feedback adjustment module. The heavy oil preheating control system is connected to the atomizing nozzle control system. The heavy oil preheating control system and the atomizing nozzle control system are connected to the data monitoring module. The data monitoring module is connected to the coupling effect identification module. The coupling effect identification module is connected to the central control module. The central control module is connected to the heavy oil preheating control system and the atomizing nozzle control system. The heavy oil preheating control system and the atomizing nozzle control system are connected to the feedback adjustment module. The feedback adjustment module is connected to the central control module and the data monitoring module. The coupling effect identification module uses the difference between the integral variable and the instantaneous variable of the nozzle pressure to the heavy oil preheating temperature as the variable of the piecewise function, and integrates the piecewise function to capture the instantaneous coupling effect, and then obtains the second-order derivative of the heavy oil preheating temperature to the nozzle pressure to capture the nonlinear change of the coupling effect. Then, the sine function of the product of the integral variable and the first-order derivative of the heavy oil flow to the heavy oil preheating temperature is integrated in the interval from 0 to the heavy oil flow to capture the coupling effect caused by the change of the heavy oil flow. The three coupling effects are weighted summed to obtain the coupling effect index. The identification formula of the coupling effect index is: Where: t is the current time, x is the integral variable, δ(x) is the piecewise function, when x = 0, δ(x) = ∞, when x ≠ 0, δ(x) = 0, P n (t) is the nozzle pressure at the current moment, T f (t) is the heavy oil preheating temperature at the current moment, Q f (t) is the heavy oil flow rate, α1, α2, and α3 are weight coefficients, which are adjusted according to historical data and actual operating conditions.

2. A multivariable anti-coupling cooperative control system for a heating furnace according to claim 1, characterized in that: The central control module predicts the future coupling effect index through the identification value of the coupling effect index based on the model predictive control algorithm, dynamically adjusts the nozzle pressure and heavy oil preheating temperature by combining the coupling effect index prediction value and the comprehensive reward function value, and optimizes the control strategy of heavy oil preheating and nozzle pressure in real time. The formula for obtaining the current optimal control strategy is: A(t+k)=argmin[I CEI (t+k)+λ·R total (t+k)] Where A(t) is the optimal control strategy for the future, k is the prediction time step, which is set according to actual needs, arg min[·] is the parameter that minimizes the objective function, λ is the penalty coefficient, which is set according to actual needs, and I CEI (t+k) is the predicted value of the coupling effect index at the future time step k, R total (t) is the comprehensive reward function value.

3. A multivariable anti-coupling cooperative control system for a heating furnace according to claim 2, characterized in that: In the central control module, the specific implementation process of predicting the future coupling effect index through the current coupling effect index based on the model prediction algorithm is as follows: Step 11, data acquisition and preprocessing: the central control module obtains the nozzle pressure, heavy oil preheating temperature, and heavy oil flow rate in real time through the data monitoring module, obtains the coupling effect index at the current moment from the coupling effect identification module, and filters and time-aligns the collected data; Step 12, data storage and analysis: store historical data within a set time range, analyze the historical data, identify the correlation and change pattern between various variables based on the time series analysis method, and obtain the trend and periodicity in the historical data; Step 13, prediction by model predictive control algorithm: based on the data of current and historical moments, a long short-term memory network prediction model of the coupling effect index is established, the model is trained and optimized, and the nozzle pressure, heavy oil preheating temperature, and heavy oil flow rate at the current moment are used as inputs, and the coupling effect index after the future time step is predicted through the prediction model; Step 14, dynamic adjustment and feedback: by comparing the predicted coupling effect index with the actual coupling effect index, the model prediction accuracy is evaluated, and the model parameters and structure are continuously optimized through the feedback mechanism; Step 15, model calibration and optimization: Regularly calibrate the model prediction algorithm, adjust model parameters, retrain the machine learning model, and switch to a more appropriate model structure.

4. The multivariable anti-coupling cooperative control system for a heating furnace according to claim 2, characterized in that: In the central control module, the comprehensive reward function is the weighted sum of the reward function values ​​for fuel utilization and pollutant emission optimization, the reward function values ​​for combustion stability and safety, and the reward function values ​​for rapid response to high load demand. The formula for the comprehensive reward function is: R total (t)=β1·R eff (t)+β2·R sta (t)+β3·R res (t) Where: R eff (t) is the reward function value for optimizing fuel utilization and pollutant emissions at the current moment, which is equal to the ratio of effective heavy oil utilization to target heavy oil utilization minus pollutant emissions, R sta (t) The combustion stability and safety reward function value is equal to the inverse of the sum of the temperature fluctuation rate, pressure fluctuation rate and safety risk index in the combustion chamber, where the safety risk index is obtained by identifying and quantifying risk factors, R res (t) is the reward function value for quick response to high load demand, which is equal to the inverse of the sum of response time and response time overshoot. The response time overshoot is the ratio of the difference between the maximum response time and the target response time to the target response time. β1, β2, and β3 are weight adjustment coefficients, which are adjusted according to actual working conditions.

5. The multivariable anti-coupling cooperative control system for a heating furnace according to claim 4, characterized in that: In the central control module, the process of obtaining the security risk index includes: Step 21, data collection: real-time collection of temperature data of the combustion chamber and key components, pressure change data of the combustion chamber, pipeline and nozzle, intensity, shape and color data of the flame, flow data of heavy oil and air, and alarm log and fault record of the system; Step 22, risk factor identification and quantification: identify the overheating risk, overpressure risk, flame instability risk, and flow abnormality risk in the data, and quantify the risk value; Step 23, risk weight allocation: After identifying and quantifying each risk factor, risk weights are allocated based on historical safety events and accident analysis, combined with expert knowledge and experience, and according to the current operating environment; Step 24, safety risk index calculation: considering the quantitative value of each risk factor and its weight, calculate the safety risk index.

6. A multivariable anti-coupling cooperative control system for a heating furnace according to claim 5, characterized in that: In the central control module, the overheating risk in step 22 is that the temperature exceeds the preset safety threshold, the overpressure risk is that the system pressure exceeds the designed pressure, the flame instability risk is that the flame goes out or swings, and the flow abnormality risk is that the heavy oil or air flow deviates from the normal working range.

7. A multivariable anti-coupling cooperative control system for a heating furnace according to claim 6, characterized in that: The heavy oil preheating control system includes a heater, a heat exchanger, a temperature controller, a circulation pump, a safety valve, a filter, an oil pipe and a flow control valve. The heavy oil passes through the filter to filter out impurities and enters the heat exchanger through the pipe. In the heat exchanger, the heavy oil exchanges heat with the heat medium. The circulation pump pushes the heat medium to circulate in the heat exchanger. The temperature controller automatically adjusts the output power of the heater according to the control instructions of the central control module. The safety valve is used to release excess pressure. The flow control valve controls the heavy oil flow through the control instructions of the central control module.

8. The multivariable anti-coupling cooperative control system for a heating furnace according to claim 7, characterized in that: The atomizing nozzle control system comprises a nozzle, a nozzle pressure regulating device, a flow controller, an air supply system, a nozzle cleaning device and a feedback controller, and the nozzle is connected with the nozzle pressure regulating device, the flow controller, the air supply system, the nozzle cleaning device and the feedback controller.

9. A multivariable anti-coupling cooperative control system for a heating furnace according to claim 8, characterized in that: The implementation process of the system for heavy oil preheating and atomizing nozzle pressure anti-coupling control is as follows: Step 31, data collection and preprocessing: The data monitoring module collects the heavy oil preheating temperature, nozzle pressure and heavy oil flow in real time, filters and performs noise reduction on the collected data, and the preprocessed data is sent to the coupling effect identification module; Step 32, coupling effect identification: the coupling effect identification module takes the difference between the integral variable and the instantaneous variable of the nozzle pressure to the heavy oil preheating temperature as the variable of the piecewise function, and integrates the piecewise function to capture the instantaneous coupling effect, and then obtains the second-order derivative of the heavy oil preheating temperature to the nozzle pressure to capture the nonlinear change of the coupling effect, and then integrates the sine function of the product of the integral variable and the first-order derivative of the heavy oil flow rate to the heavy oil preheating temperature in the interval from 0 to the heavy oil flow rate to capture the coupling effect caused by the change of the heavy oil flow rate, and performs weighted summation of these three coupling effects to obtain the coupling effect index, and feeds the index back to the central control module; Step 33, model prediction and control strategy optimization: The central control module uses the long short-term memory network model based on the model predictive control algorithm, uses the current and historical data to predict the coupling effect index at the future moment, combines the coupling effect index prediction value and the comprehensive reward function value, calculates the current control strategy reward value, and solves the objective function A(t+k)=arg min[I CEI (t+k)+λ·R total (t+k)], determine the optimal control strategy at the future time, including the adjustment amount of nozzle pressure and heavy oil preheating temperature; Step 34, control strategy execution and feedback adjustment: the central control module issues instructions to adjust the corresponding parameters in the heavy oil preheating system and the atomizing nozzle control system according to the optimal control strategy; the temperature controller automatically adjusts the power output of the heater, and the flow control valve adjusts the heavy oil flow to ensure that the preheating temperature reaches the optimal state; at the same time, the nozzle pressure is adjusted to match the optimized preheating temperature, and the data monitoring module continuously monitors the response of the system, including the new nozzle pressure, heavy oil preheating temperature and flow, and feeds back to the central control module in real time. The central control module compares the new real-time data with the predicted value to evaluate the control effect. If there is a deviation, the system adjusts the control parameters again through the feedback mechanism to optimize the control strategy; Step 35, model calibration and optimization: Regularly evaluate the accuracy of model predictions, analyze error distribution, identify room for model improvement, adjust the parameters or structure of the long short-term memory network model based on actual operating results, retrain the model when necessary to ensure that it adapts to the actual operating conditions of the system, and update the control strategy algorithm after model calibration.

Images