Furnace apparatus flow control system and method

Abstract

The application discloses a furnace equipment flow control system and method, and relates to the field of thermal process control.The method comprises the following steps: S1, collecting multi-source sensor data in real time, performing time synchronization, filtering and normalization processing, and obtaining a standardized real-time data stream; S2, calculating characteristic parameters based on the standardized real-time data stream, wherein the characteristic parameters comprise a dynamic thermal flow equalization factor, a process entropy and an equivalent purging efficiency; and S3, generating a gas flow control instruction based on the characteristic parameters through a preset intelligent decision rule.Through intelligent data fusion and self-learning optimization, accurate flow control and temperature control precision improvement are realized.Through intelligent purging mode switching, the purging efficiency and energy-saving effect are optimized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of thermal process control, specifically a furnace equipment flow control system and method. BACKGROUND

[0002] Furnace equipment is widely used in high-temperature environments, and accurate flow control is crucial for improving production efficiency, reducing energy consumption, and ensuring product quality. Existing flow control methods mostly rely on simple temperature, flow, and pressure feedback mechanisms, which can achieve basic temperature control but cannot handle multiple variables in complex process environments, resulting in insufficient control accuracy and response speed.

[0003] Current technology mainly relies on single physical feedback adjustment, ignoring comprehensive regulation of multiple parameters and multiple process stages. Especially in the face of complex situations such as thermal flow fluctuations and gas flow changes, traditional methods often cannot achieve timely and accurate flow regulation. In addition, existing intelligent control systems lack dynamic feedback response and fail to adjust in time according to different process stages, resulting in significant control lag.

[0004] Therefore, there is an urgent need for a flow control method that combines multi-source sensor data, dynamic process parameters, and intelligent decision-making algorithms to achieve efficient and accurate gas flow regulation. This method should have the ability to dynamically adjust and optimize processes to improve the overall performance and intelligence level of furnace equipment. SUMMARY

[0005] Based on the shortcomings of the existing technology described above, the purpose of the present application is to provide a furnace equipment flow control system and method to solve the above technical problems.

[0006] To achieve the above purpose, the present application provides the following technical solution: a furnace equipment flow control method, comprising:

[0007] S1: Real-time acquisition of multi-source sensor data, time synchronization, filtering, and normalization processing to obtain standardized real-time data stream;

[0008] S2: Calculate feature parameters based on the standardized real-time data stream, including dynamic thermal flow equalization factor, process entropy, and equivalent purge efficiency;

[0009] S3: Based on the feature parameters, generate gas flow control instructions through a pre-set intelligent decision-making rule.

[0010] The present application further provides that S1 includes:

[0011] Real-time acquisition of raw data from multi-source sensors, including: furnace gas inlet temperature, heating zone temperature, heater power, gas flow, furnace pressure, furnace inlet pressure, and oxygen concentration;

[0012] The original data is subjected to time synchronization processing, which is marked based on a unified system timestamp;

[0013] The time-synchronized data is subjected to filtering processing, and high-frequency noise is eliminated by using a sliding average method or a low-pass filter;

[0014] The filtered data is subjected to normalization processing, and data of different dimensions are normalized to a predetermined standard range;

[0015] All processed data is summarized to generate a standardized real-time data stream.

[0016] The application further provides that the S2 comprises a dynamic heat flow balance factor generation unit, a process entropy generation unit and an equivalent purge efficiency generation unit;

[0017] The dynamic heat flow balance factor generation unit comprises:

[0018] The standardized real-time data stream output by S1 is received, and the furnace inlet gas temperature, the heater power and the gas flow are extracted;

[0019] According to the pre-stored property-temperature relationship mapping table, the real-time collected furnace inlet gas temperature is taken as a query key for matching or interpolation calculation to obtain the corresponding gas physical properties at the current temperature;

[0020] Based on the gas flow, the furnace inlet gas temperature, the gas physical properties and the preset target furnace temperature, the theoretical heat load is calculated;

[0021] By real-time energy matching method, the heater power is taken as the actual energy input, and the deviation calculation is performed with the theoretical heat load to generate a dynamic heat flow balance factor.

[0022] The application further provides that the process entropy generation unit comprises:

[0023] The standardized real-time data stream output by S1 is received, and the heating zone temperature and the gas flow in a preset historical time period are extracted;

[0024] A plurality of time scale windows of different spans are set, and the heating zone temperature fluctuation characteristic quantity and the gas flow fluctuation characteristic quantity in each time scale window are calculated respectively;

[0025] The system determines the current process stage according to the current process state, and the process stage comprises a temperature control stage and a purge stage, and when in the temperature control stage, further comprises a heating stage, a holding stage and a cooling stage;

[0026] If the current is in the temperature control stage, based on the pre-stored temperature control stage-weight configuration mapping relationship, the fusion weight is assigned to the heating area temperature fluctuation feature quantity and the gas flow fluctuation feature quantity under each time scale window;

[0027] Based on the fusion weight, the heating area temperature fluctuation feature quantity and the gas flow fluctuation feature quantity under each time scale window are weighted and fused to generate the process process entropy.

[0028] The application further provides that the equivalent purge efficiency generation unit comprises:

[0029] The system determines the current process stage according to the current process state;

[0030] If the current is in the purge stage, the normalized real-time data stream output by S1 is received, and the oxygen concentration and the gas flow are extracted;

[0031] The oxygen concentration value at the starting moment of the purge stage is taken as the initial oxygen concentration, and a preset target oxygen concentration value is called;

[0032] Based on the oxygen concentration and the gas flow, the ratio of the oxygen concentration change amount to the gas consumption amount per unit time is calculated to obtain the oxygen concentration change rate.

[0033] The application further provides that the oxygen concentration change rate is tracked in real time, and the decay trend is calculated by linear regression method;

[0034] The decay trend is compared with a preset decay threshold value, and when the decay trend reaches or is lower than the decay threshold value, it is determined that the purge process enters the low-efficiency stage, and the purge duration is recorded;

[0035] Based on the oxygen concentration change rate and the purge duration, an equivalent purge efficiency is generated by an efficiency calculation model, and a benefit decay flag is also generated.

[0036] The application further provides that the S3 comprises a parameter optimization and control unit and an intelligent purge termination unit;

[0037] The parameter optimization and control unit comprises:

[0038] The system determines the current process stage according to the current process state;

[0039] When the system is in the temperature control stage, the dynamic heat flow balance factor output by S2 is received;

[0040] The dynamic heat flow balance factor is compared with a preset upper limit and lower limit of the heat flow target range;

[0041] If the dynamic heat flow balance factor is continuously higher than the upper limit, an instruction for incrementing the gas flow setting value is generated and executed;

[0042] If the dynamic heat flow balance factor continues to be lower than the lower limit, an instruction for decreasing the gas flow setting value is generated and executed.

[0043] The application is further configured to, since the second process cycle, receive the process entropy output by S2, compare the current cycle process entropy with the last cycle process entropy;

[0044] If the current cycle process entropy is greater than the last cycle process entropy, the PID control parameters of the temperature controller are optimized by using the parameter perturbation method to reduce the process entropy, and a parameter optimization instruction is generated and issued to the temperature controller.

[0045] The application is further configured that the intelligent purging termination unit comprises:

[0046] When the system is in the purging phase, the equivalent purging efficiency output by S2 and the change trend of the equivalent purging efficiency are received.

[0047] The change trend is compared with the preset efficiency decay threshold in real time;

[0048] When it is monitored that the change trend continues to be lower than the efficiency decay threshold, a purging mode switching instruction is generated and executed to adjust the gas flow and control the flow actuator to switch from the high-flow purging mode to the low-flow maintenance mode.

[0049] The application also provides a furnace equipment flow control system, which comprises:

[0050] The data acquisition and preprocessing module: real-time acquisition of multi-source sensor data, time synchronization, filtering and normalization processing, and obtaining of standardized real-time data stream;

[0051] The high-dimensional feature parameter calculation module: calculation of feature parameters based on the standardized real-time data stream, the feature parameters including a dynamic heat flow balance factor, a process entropy and an equivalent purging efficiency;

[0052] The intelligent decision-making and setting value generation module: generation of a gas flow control instruction based on the feature parameters through a preset intelligent decision-making rule.

[0053] The application provides a furnace equipment flow control system and method, which comprises the following steps: S1: real-time acquisition of multi-source sensor data, time synchronization, filtering and normalization processing, and obtaining of standardized real-time data stream; S2: calculation of feature parameters based on the standardized real-time data stream, the feature parameters including a dynamic heat flow balance factor, a process entropy and an equivalent purging efficiency; S3: generation of a gas flow control instruction based on the feature parameters through a preset intelligent decision-making rule, and the beneficial effects include:

[0054] (1) Multi-dimensional intelligent data fusion and real-time processing: The present application can comprehensively reflect the process state of the furnace equipment by real-time collection and standardization of multi-source sensor data combined with high-dimensional feature parameters. Compared with the prior art, the present application provides more accurate flow control and regulation through data fusion and dynamic optimization algorithm, so that the system can realize real-time self-adaptive adjustment in different process stages, significantly improving the control precision and reaction speed.

[0055] (2) Self-learning optimization and precise regulation capability: The present application adopts a self-optimization mechanism based on process process entropy, which automatically adjusts the PID control parameters of the temperature controller by comparing the entropy value changes in different process cycles. This innovative method can continuously optimize the control strategy according to the changes in the process state, effectively reducing oscillation and improving temperature control accuracy. This self-learning optimization capability greatly improves the intelligence level of the system, reduces manual intervention, and ensures efficient operation of the furnace equipment in different process stages.

[0056] (3) Intelligent purging process and benefit maximization: In the purging stage, the present application introduces equivalent purging efficiency and purging mode switching mechanism, which accurately judges the benefit of the purging stage by real-time monitoring of oxygen concentration change trend and comparison with the preset decay threshold. When the purging efficiency is continuously lower than the threshold, the system will automatically switch to low flow maintenance mode, avoiding energy waste. This optimization strategy not only improves the efficiency of the purging process, but also effectively reduces energy consumption, achieving energy saving and benefit optimization. BRIEF DESCRIPTION OF DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor. In the drawings:

[0058] Figure 1 A flow chart of a furnace equipment flow control method is shown for an exemplary embodiment of the present application;

[0059] Figure 2 A structural schematic diagram of a furnace equipment flow control system is shown for an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0060] The present application will be described in more detail by the following embodiments with reference to the drawings, and other advantages and effects of the present application will be apparent to those skilled in the art from this disclosure. The present application can be implemented or applied in other different specific embodiments, and various modifications or changes can be made based on different views and applications without departing from the spirit of the present application. It should be understood that the preferred embodiments are only for illustrating the present application, but not for limiting the scope of protection of the present application.

[0061] It should be noted that the drawings provided in the following embodiments only schematically illustrate the basic concept of the present application, and only the components related to the present application are shown in the drawings, but not drawn according to the number, shape and size of the components in actual implementation, and the shape, number and ratio of each component in actual implementation can be arbitrarily changed, and the layout pattern of the components can be more complex.

[0062] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application, however, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details, and in other embodiments, the known structures and devices are shown in the form of block diagrams rather than in the form of details, to avoid making the embodiments of the present application difficult to understand.

[0063] Embodiment one: a furnace device flow control method, as shown in Figure 1 , comprising:

[0064] S1: collecting multi-source sensor data in real time, performing time synchronization, filtering and normalization processing to obtain standardized real-time data stream;

[0065] S2: calculating feature parameters based on the standardized real-time data stream, the feature parameters including dynamic thermal flow equalization factor, process entropy and equivalent purge efficiency;

[0066] S3: generating gas flow control instructions based on the feature parameters through a pre-set intelligent decision rule.

[0067] The present application is further provided that the S1 comprises:

[0068] collecting raw data of multi-source sensors in real time, the raw data including: furnace gas inlet temperature, heating zone temperature, heater power, gas flow, furnace pressure, furnace inlet pressure, oxygen concentration;

[0069] performing time synchronization processing on the raw data, the time synchronization being marked based on a unified system timestamp;

[0070] performing filtering processing on the time-synchronized data, eliminating high-frequency noise by using a moving average method or a low-pass filter;

[0071] The filtered data is normalized to normalize data of different dimensions to a predetermined standard range;

[0072] All processed data is summarized to generate standardized real-time data stream. Specifically, first, the system collects various types of raw data in the furnace equipment in real time through multiple sensors. The raw data includes the following contents: 1, furnace gas inlet temperature: used to reflect the temperature of the gas at the furnace mouth, usually obtained by the thermocouple sensor installed at the furnace mouth; 2, heating zone temperature: used to monitor the temperature of the heating zone, usually collected in real time by temperature sensors such as thermocouples or infrared sensors; 3, heater power: provided by the power output sensor of the temperature control system, indicating the current working power of the heater; 4, gas flow: the gas flow feedbacks the flow data in real time through the mass flow meter; 5, furnace pressure: detected by the furnace pressure sensor, reflecting the pressure condition inside the furnace; 6, furnace inlet gas pressure: data collected by the furnace inlet gas pressure sensor, used to monitor the inlet gas pressure at the furnace mouth; 7, oxygen concentration: the oxygen probe detects the oxygen concentration in the furnace, commonly used to control the stability of the atmosphere. All raw data collected by the sensors are marked with a unified system timestamp to ensure that the data from different sensors can be processed synchronously within the same time period. The purpose of time synchronization processing is to eliminate the data timing problems caused by different response times of different sensors, and to ensure the accuracy and timeliness of data processing. Time synchronization is generally marked and synchronized by a centralized control system combined with high-precision clock signals. The data after time synchronization may contain some high-frequency noise, causing unstable data. Therefore, filtering processing is needed to improve data quality. Common filtering methods include: moving average method: by setting a certain window period, the average value of data in each window is calculated to smooth the data and remove sudden noise fluctuations. Low-pass filter: using low-pass filter algorithm, it can filter out noise signals higher than the set threshold frequency, retain the low-frequency components of the data, and ensure the smoothness of the data. These two methods are used together in the processing process, the moving average method handles rapid changes in interference, and the low-pass filter handles high-frequency noise. The filtered data is standardized. Each data source is scaled to a unified standard range, which is 0 to 1 by default, through linear conversion. This step is completed by min-max normalization or Z-score standardization method, so that data of different dimensions can be uniformly processed. Finally, all processed data is summarized into a standardized real-time data stream as input for subsequent calculation modules, providing a basis for flow control and other intelligent decision-making.

[0073] The application further provides that the S2 includes a dynamic heat flow equalization factor generation unit, a process entropy generation unit, and an equivalent purge efficiency generation unit.

[0074] The dynamic heat flow equalization factor generation unit comprises:

[0075] Receiving the normalized real-time data stream output by S1, extracting the furnace mouth gas inlet temperature, heater power and gas flow rate;

[0076] According to the pre-stored physical property-temperature mapping table, the real-time collected furnace mouth gas inlet temperature is taken as a query key for matching or interpolation calculation to obtain the corresponding gas physical property at the current temperature;

[0077] Based on the gas flow rate, the furnace mouth gas inlet temperature, the gas physical property and the preset target hearth temperature, the theoretical heat load is calculated;

[0078] Through real-time energy benchmarking, the heater power is taken as the actual energy input to perform deviation calculation with the theoretical heat load to generate the dynamic heat flow equalization factor. Specifically, the furnace mouth gas inlet temperature, the heater power and the gas flow rate are extracted from the normalized real-time data stream output by S1, which reflect the current working state of the furnace in real time. Then, the furnace mouth gas inlet temperature is used to search the pre-stored physical property-temperature mapping table to obtain the corresponding gas physical property at the current temperature, which includes the specific heat capacity and the density. If there is no corresponding temperature point in the mapping table, the gas physical property is calculated through an interpolation algorithm such as linear interpolation. The theoretical heat load is calculated in combination with the gas flow rate, the furnace mouth gas inlet temperature, the gas physical property and the preset target hearth temperature. The theoretical heat load represents the heat that needs to be brought into the hearth by the gas flow per unit time to maintain the target hearth temperature. The specific calculation process is as follows: according to the gas flow rate and the gas physical property, the heat carried away by the gas is calculated, and the calculation principle is that the theoretical heat load is equal to the product of the gas mass flow rate, the gas constant-pressure specific heat capacity and the difference between the target hearth temperature and the furnace mouth gas inlet temperature. Finally, the dynamic heat flow equalization factor is generated through real-time energy benchmarking. The real-time energy benchmarking compares the actual output power of the heater as the actual energy input with the calculated theoretical heat load to obtain the ratio, which is the dynamic heat flow equalization factor. The dynamic heat flow equalization factor quantifies the dynamic balance state of the heat flow in the furnace. If the value is close to 1, it means that the heating power in the hearth matches the actual demand. If it is greater than 1, it means that there is heat surplus. If it is less than 1, it means that the heat energy input needs to be increased.

[0079] The present application further provides that the process entropy generation unit comprises:

[0080] Receiving the normalized real-time data stream output by S1, extracting the heating zone temperature and the gas flow rate in a preset historical time period;

[0081] A plurality of time scale windows with different spans are set, and a heating zone temperature fluctuation feature and a gas flow fluctuation feature in each time scale window are calculated respectively;

[0082] The system determines a current process phase according to a current process state, the process phase including a temperature control phase and a purge phase, and when in the temperature control phase, further including a temperature rising phase, a temperature maintaining phase and a temperature falling phase;

[0083] If the current process is in the temperature control phase, a pre-stored temperature control phase-weight configuration mapping relationship is used to assign a fusion weight to the heating zone temperature fluctuation feature and the gas flow fluctuation feature in each time scale window;

[0084] Based on the fusion weight, the heating zone temperature fluctuation feature quantity and the gas flow fluctuation feature quantity under each time scale window are weighted and fused to generate the process entropy. Specifically, the process entropy generation unit first extracts the heating zone temperature and gas flow data from the standardized real-time data stream output by S1. These data reflect the state of the furnace equipment at each process stage, especially the temperature change of the heating zone and the fluctuation of the gas flow. The heating zone temperature is used to represent the temperature of the heating zone, reflecting the energy distribution during the heating process of the equipment. The gas flow represents the flow rate of the gas in the furnace, reflecting the influence of gas circulation on the heat and process in the furnace. In order to fully understand the fluctuation characteristics of the process in the furnace, the system needs to set multiple time scale windows with different spans. These windows can be adjusted flexibly according to actual needs; each window represents a different time period, aiming to capture the characteristics of temperature and flow fluctuations at different time scales. The time scale window can be seconds, minutes, or hours, depending on the fluctuation frequency and period to be analyzed, and the default setting is minutes. For each time scale window, the system needs to calculate the following two fluctuation feature quantities: 1. Heating zone temperature fluctuation feature quantity: The heating zone temperature fluctuation feature quantity represents the degree of change and volatility of the heating zone temperature within a certain time window, which is represented by calculating the standard deviation of the temperature within the window. 2. Gas flow fluctuation feature quantity: The gas flow fluctuation feature quantity represents the fluctuation of the gas flow within a certain time window. Similar to the heating zone temperature fluctuation feature quantity, it is reflected by calculating the standard deviation of the flow within the window. These fluctuation feature quantities help measure the stability of heat and gas flow in the process and the degree of their influence on the system. Based on the calculation of the fluctuation feature quantities, the system determines the process stage of the furnace equipment according to the current process state. Different stages have different effects on the stability of the process and the entropy value, so weight configuration needs to be performed according to the process stage. The process stage mainly includes: 1. Temperature control stage: refers to the stage of temperature control in the furnace, which is further divided into: heating stage: the equipment is heating or in the process of heating; holding stage: the stage of maintaining a constant temperature state of the equipment; cooling stage: the equipment is cooling. 2. Purging stage: refers to the stage of purging or cleaning the equipment. If it is currently in the temperature control stage, the heating zone temperature fluctuation feature quantity and the gas flow fluctuation feature quantity under different time scale windows are weighted and distributed according to the pre-stored process stage-weight configuration mapping relationship. Specifically, in the design at the beginning, when designing the pre-set process stage-weight configuration mapping relationship, higher weights are assigned to short time scale windows in the heating or cooling stage; higher weights are assigned to long time scale windows in the holding stage.Finally, based on the assigned fusion weight, the temperature and flow fluctuation features in each time scale window are weighted and fused to calculate the process entropy; the process entropy comprehensively considers the influence of temperature and flow fluctuation on the process entropy, and quantifies the fluctuation and disorder degree of the process; the greater the entropy value, the higher the fluctuation and uncertainty of the system; the smaller the entropy value, the more stable and orderly the process.

[0085] The application further provides that the equivalent purging efficiency generation unit comprises:

[0086] The system determines the current process phase according to the current process state;

[0087] If the current process is in the purging phase, the normalized real-time data stream output by S1 is received, and the oxygen concentration and gas flow are extracted;

[0088] The oxygen concentration value at the starting moment of the purging phase is taken as the initial oxygen concentration, and a preset target oxygen concentration value is called;

[0089] Based on the oxygen concentration and gas flow, the ratio of the oxygen concentration change amount to the gas consumption amount per unit time is calculated to obtain the oxygen concentration change rate. Specifically, when the system determines that the current process is in the purging phase, the system first extracts the key real-time data from the normalized real-time data stream output by S1, which includes the oxygen concentration and the gas flow. Specifically, the oxygen concentration is measured and fed back by the oxygen concentration sensor, reflecting the content of oxygen in the furnace atmosphere; the gas flow is monitored by the mass flow meter, indicating the flow rate of the gas in the purging phase. The oxygen concentration value at the starting moment of the purging phase is taken as the initial oxygen concentration, and the initial oxygen concentration represents the oxygen content in the furnace at the start of the purging phase, which is the atmosphere state in the furnace at the start of the purging phase. This initial oxygen concentration value will be used as a reference value for subsequent oxygen concentration change tracking. At the same time, the system obtains the ideal oxygen concentration from the preset target oxygen concentration value, which is based on the process requirements and design standards. Based on the oxygen concentration change and the cumulative gas consumption from the start of the purging to the current moment, the oxygen concentration change rate is calculated. In a preferred embodiment, the oxygen concentration change rate is calculated by dividing the difference between the initial oxygen concentration and the current oxygen concentration by the cumulative gas consumption volume from the start of the purging to the current moment, and the ratio of them is the oxygen concentration change rate. The oxygen concentration change rate is a measure of the ratio of the oxygen concentration change amount to the gas consumption amount, which represents the oxygen concentration that can be reduced by consuming a unit volume of purging gas.

[0090] The application further provides that the oxygen concentration change rate is tracked in real time, and the decay trend is calculated by linear regression;

[0091] The decay trend is compared with a preset decay threshold, and when the decay trend reaches or is lower than the decay threshold, it is determined that the purging process enters the inefficient stage, and the purging duration is recorded;

[0092] Based on the oxygen concentration change rate and the purging duration, an equivalent purging efficiency is generated through an efficiency calculation model, and a benefit decay flag is also generated. Specifically, the system tracks the change trend of the oxygen concentration change rate in real time, and within a preset time range, a linear regression method is used to fit the data sequence of the oxygen concentration change rate over time. The slope of the straight line obtained by fitting is the decay trend. Linear regression describes the change of the oxygen concentration change rate over time through a mathematical model, and then calculates a decay trend. The decay trend quantifies the rate of decline of the oxygen concentration change rate. The calculated decay trend is compared with a preset decay threshold, and when the absolute value of the decay trend reaches or is lower than the threshold, the system determines that the purging process enters the inefficient stage, and records the purging duration since the start of purging. Subsequently, based on the oxygen concentration change rate and the purging duration, an equivalent purging efficiency is generated through an efficiency calculation model. The equivalent purging efficiency is a comprehensive efficiency value considering the oxygen concentration change and gas flow consumption, representing the overall benefit of the purging process. The higher the value, the better the gas utilization efficiency of the purging. When the system calculates the equivalent purging efficiency, a benefit decay flag is further generated. The benefit decay flag is used to indicate whether the efficiency is declining during the purging process, and is a Boolean value indicating whether the current is in the inefficient purging stage.

[0093] The S3 further includes a parameter optimization and control unit and an intelligent purging termination unit.

[0094] The parameter optimization and control unit includes:

[0095] The system determines the current process stage according to the current process state.

[0096] When the system is in the temperature control stage, the dynamic heat flow balance factor output by S2 is received.

[0097] The dynamic heat flow balance factor is compared with the preset upper and lower limits of the heat flow target range.

[0098] If the dynamic heat flow balance factor is continuously higher than the upper limit, an instruction for incrementing the gas flow set value is generated and executed.

[0099] If the dynamic heat flow balance factor continues to be lower than the lower limit, instructions for decreasing the gas flow setting value are generated and executed. Specifically, the system first determines the current process stage according to real-time data such as the furnace mouth temperature, the heater power, and the gas flow, and confirms whether the current process stage is a temperature control stage. When the current process stage is the temperature control stage, the dynamic heat flow balance factor received from the S2 layer is compared with the preset upper limit and lower limit of the heat flow target range, wherein the heat flow target range includes the upper limit and lower limit of the dynamic heat flow balance factor, and is used to ensure that the temperature in the furnace is maintained in the ideal range. If the dynamic heat flow balance factor continues to be higher than the upper limit, it indicates that the heating power in the furnace is excessive, and the system generates and executes instructions for increasing the gas flow setting value. By increasing the gas flow, the cooling effect on the furnace is enhanced, so that the factor falls back to the target range. If the dynamic heat flow balance factor continues to be lower than the lower limit, it indicates that the heat carried away by the gas flow is too much, and the system generates and executes instructions for decreasing the gas flow setting value. By reducing the gas flow, the cooling effect is weakened, which helps to increase the temperature of the furnace, so that the dynamic heat flow balance factor returns to the target range. The adjustment process is usually a short cycle, which ensures real-time response to temperature changes. The short cycle control is based on the feedback of real-time data, and the system continuously fine-tunes the gas flow to respond to changes in the temperature of the furnace. During the execution process, the system continuously monitors the temperature control effect. If the adjusted gas flow does not make the temperature in the furnace return to the target range, the system will re-evaluate and make new adjustments.

[0100] The application is further provided that, from the second process cycle, the process entropy output by S2 is received, and the current cycle process entropy is compared with the last cycle process entropy.

[0101] If the process entropy of the current period is greater than the process entropy of the last period, the PID control parameters of the temperature controller are optimized by using the parameter perturbation method with the target of reducing the process entropy, and a parameter optimization instruction is generated and sent to the temperature controller. Specifically, from the second process period, at the end of each period, the system receives the process entropy of the current period output by the S2 layer, and compares it with the process entropy of the last period stored in the database. If the process entropy of the current period is less than or equal to the entropy value of the last period, it indicates that the process stability is flat or improved, and the system maintains the current PID parameters unchanged and directly enters the next period. If the process entropy of the current period is greater than the entropy value of the last period, it indicates that the process fluctuation increases, and the system starts the parameter optimization process. This process is based on the parameter perturbation method, and the specific steps are as follows: the system first backs up the current PID parameter set, including the proportional gain, integral time and differential time, then applies a preset and small positive perturbation to one of the parameters, for example, increases the proportional gain value by 5%, generates a new candidate parameter set, sends the candidate parameter set to the temperature controller, and applies it to the next complete process period. After the period ends, the process entropy generated under the new parameters is obtained, and the optimized process entropy is compared with the process entropy before optimization. If the optimized process entropy is less than the process entropy before optimization, it indicates that the perturbation direction is correct and the optimization is effective, and the system retains this parameter set as the new benchmark and performs the next perturbation in this direction. If the optimized process entropy is greater than the process entropy before optimization, it indicates that the perturbation direction is incorrect, the system restores the original parameters from the backup, and applies a negative perturbation to the parameter, for example, reduces the proportional gain value by 5%, and repeats the previous steps until the process entropy no longer decreases significantly or the maximum number of iterations is reached. At this time, it is considered that the current scheme is the optimal parameter. Finally, the optimized PID control parameter instruction is sent to the temperature controller, and the temperature controller adjusts the heating power of the heating zone according to the new PID parameters to achieve more accurate temperature control.

[0102] The application further provides that the intelligent purging termination unit comprises:

[0103] When the system is in the purging phase, the equivalent purging efficiency and the change trend of the equivalent purging efficiency output by S2 are received;

[0104] The change trend is compared with the preset efficiency decay threshold in real time;

[0105] When the change trend is continuously lower than the efficiency decay threshold, a purge mode switching instruction is generated and executed, the gas flow is adjusted, and the flow executor is controlled to switch from the high-flow purge mode to the low-flow maintenance mode. Specifically, when the system is in the purging stage, the equivalent purge efficiency and its change trend received by S2 output are monitored and compared with the preset efficiency decay threshold in real time, wherein the specific value of the efficiency decay threshold is determined by engineering experience or historical data and is set in advance according to the change range of the gas flow and the oxygen concentration; when the change trend is continuously lower than the efficiency decay threshold, it indicates that the purging process has entered the low-efficiency stage, and at this time, the purification effect improved by unit gas consumption is minimal, the system generates and executes a purge mode switching instruction, and controls the flow executor to switch the gas flow from the high-flow purge mode to the low-flow maintenance mode, thereby significantly saving gas consumption while ensuring the quality of the furnace atmosphere.

[0106] Embodiment two: please refer to Figure 2 The exemplary furnace device flow control system includes:

[0107] The data acquisition and preprocessing module: real-time acquisition of multi-source sensor data, time synchronization, filtering and normalization processing, and obtaining of standardized real-time data stream;

[0108] The high-dimensional feature parameter calculation module: calculating feature parameters based on the standardized real-time data stream, the feature parameters including dynamic heat flow equalization factor, process entropy and equivalent purge efficiency;

[0109] The intelligent decision-making and set value generation module: generating gas flow control instructions based on feature parameters through preset intelligent decision-making rules.

[0110] It should be noted that the furnace device flow control system provided by the above embodiment and the furnace device flow control method provided by the above embodiment belong to the same concept, wherein the specific operation execution manner of each module and unit has been described in detail in the method embodiment, which will not be repeated here. The furnace device flow control system provided by the above embodiment can complete the above-described all or part of functions by different functional modules according to the needs in actual application, i.e., the internal structure of the system is divided into different functional modules to complete the above-described all or part of functions, and this is not limited herein.

[0111] 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 can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A furnace apparatus flow control method characterized by, The method comprises the following steps: S1: collecting multi-source sensor data in real time, performing time synchronization, filtering and normalization processing to obtain standardized real-time data stream; S2: calculating feature parameters based on the standardized real-time data stream, the feature parameters comprising a dynamic heat flow balance factor, a process entropy and an equivalent purging efficiency, and S2 comprising a dynamic heat flow balance factor generation unit, a process entropy generation unit and an equivalent purging efficiency generation unit; The dynamic heat flow balance factor generation unit comprises: receiving the standardized real-time data stream output by S1, extracting the furnace gas inlet temperature, the heater power and the gas flow; using the real-time collected furnace gas inlet temperature as a query key to match or interpolate and calculate the corresponding gas physical properties at the current temperature according to the pre-stored physical property-temperature mapping table; calculating the theoretical heat load based on the gas flow, the furnace gas inlet temperature, the gas physical properties and the pre-set target furnace temperature; and generating the dynamic heat flow balance factor by performing real-time energy deviation calculation, taking the heater power as the actual energy input and comparing it with the theoretical heat load. S3: generating a gas flow control instruction based on the feature parameters through a pre-set intelligent decision rule, and S3 comprising a parameter optimization and control unit and an intelligent purging termination unit; The parameter optimization and control unit comprises: judging the current process phase according to the current process state; receiving the dynamic heat flow balance factor output by S2 when the system is in the temperature control phase; comparing the dynamic heat flow balance factor with the pre-set upper and lower limits of the heat flow target range; generating and executing an instruction for increasing the gas flow set value if the dynamic heat flow balance factor continuously exceeds the upper limit; and generating and executing an instruction for decreasing the gas flow set value if the dynamic heat flow balance factor continuously falls below the lower limit. The intelligent purging termination unit comprises: receiving the equivalent purging efficiency and the change trend of the equivalent purging efficiency output by S2 when the system is in the purging phase; comparing the change trend with the pre-set efficiency decay threshold in real time; and generating and executing a purging mode switching instruction to adjust the gas flow and control the flow actuator to switch from the high-flow purging mode to the low-flow maintenance mode when it is monitored that the change trend continuously falls below the efficiency decay threshold.

2. A furnace apparatus flow control method according to claim 1, wherein S1 comprises: collecting raw data of multi-source sensors in real time, the raw data comprising: furnace gas inlet temperature, heating zone temperature, heater power, gas flow, furnace pressure, furnace inlet gas pressure and oxygen concentration; performing time synchronization processing on the raw data, the time synchronization being based on a unified system timestamp for marking; performing filtering processing on the time-synchronized data to eliminate high-frequency noise by using a moving average method or a low-pass filter; performing normalization processing on the filtered data to normalize the data of different dimensions to a pre-determined standard range; summarizing all processed data to generate a standardized real-time data stream.

3. The method of claim 1, wherein, The process entropy generation unit comprises: receiving the standardized real-time data stream output by S1, and extracting the heating zone temperature and the gas flow in a pre-set historical time period; Set multiple time scale windows with different spans, and calculate the heating zone temperature fluctuation feature and the gas flow fluctuation feature in each time scale window respectively; The system determines the current process phase according to the current process state, and the process phase includes a temperature control phase and a purge phase. When in the temperature control phase, the temperature control phase further includes a temperature rising phase, a temperature maintaining phase and a temperature falling phase; If the current is in the temperature control phase, the heating zone temperature fluctuation feature and the gas flow fluctuation feature in each time scale window are assigned fusion weights based on the pre-stored temperature control phase-weight configuration mapping relationship; Based on the fusion weights, the heating zone temperature fluctuation feature and the gas flow fluctuation feature in each time scale window are weighted and fused to generate a process entropy.

4. The method of claim 1, wherein, The equivalent purge efficiency generation unit includes: The system determines the current process phase according to the current process state; If the current is in the purge phase, the normalized real-time data stream output by S1 is received, and the oxygen concentration and the gas flow are extracted; The oxygen concentration value at the start time of the purge phase is taken as the initial oxygen concentration, and the preset target oxygen concentration value is called; Based on the oxygen concentration and the gas flow, the ratio of the oxygen concentration change amount to the gas consumption amount per unit time is calculated to obtain the oxygen concentration change rate.

5. The furnace equipment flow control method according to claim 4, characterized in that, The oxygen concentration change rate is tracked in real time, and the decay trend is calculated by linear regression method; The decay trend is compared with the preset decay threshold value, and when the decay trend reaches or is lower than the decay threshold value, it is determined that the purge process enters the low efficiency stage, and the purge duration is recorded; Based on the oxygen concentration change rate and the purge duration, the equivalent purge efficiency is generated by the efficiency calculation model, and the benefit decay flag is also generated.

6. The furnace equipment flow control method according to claim 1, characterized in that, From the second process cycle, the process entropy output by S2 is received, and the current cycle process entropy is compared with the last cycle process entropy; If the current cycle process entropy is greater than the last cycle process entropy, the PID control parameters of the temperature controller are optimized by the parameter disturbance method to reduce the process entropy, and the parameter optimization instruction is generated and sent to the temperature controller.

7. A furnace device flow control system for implementing a furnace device flow control method according to any one of claims 1 to 6, characterized by It includes: Data acquisition and preprocessing module: real-time acquisition of multi-source sensor data, time synchronization, filtering and normalization processing to obtain normalized real-time data stream; High-dimensional feature parameter calculation module: calculating feature parameters based on the normalized real-time data stream, the feature parameters including dynamic heat flow balance factor, process entropy and equivalent purge efficiency; Intelligent decision and set value generation module: generating gas flow control instructions based on feature parameters through preset intelligent decision rules.

Images