Method for constructing heat transfer coefficient model of cooling system after rolling of medium and heavy plates
By constructing a heat transfer coefficient model that considers the initial temperature field and dynamic flow response in the thickness direction, and optimizing the physical property parameters, the accuracy and adaptability problems of the post-rolling cooling system for medium and heavy plates in the existing technology are solved, realizing high-precision control and flexible production of thick steel plates, and improving product quality and production efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING SCI&TECH UNIV DESIGN RES YUAN CO
- Filing Date
- 2025-12-10
- Publication Date
- 2026-04-24
AI Technical Summary
Existing heat transfer coefficient models for post-rolling cooling systems of medium and heavy plates suffer from problems such as neglecting the temperature gradient in the thickness direction in the initial temperature field, not relating the heat transfer coefficient calculation to changes in water volume and pressure, not considering the influence of chemical composition in the physical property parameters, low computational efficiency, and weak adaptability. These issues result in insufficient accuracy in hitting the red-hot temperature of thick steel plates, failing to meet the needs of the high-end equipment manufacturing industry.
By constructing an initial temperature field that considers the temperature gradient in the thickness direction, establishing a heat transfer coefficient correlation model for dynamic flow response, optimizing physical property parameters, introducing a carbon content correction term, and employing implicit difference calculation and batch self-learning functions, the model achieves accurate prediction and rapid adaptation.
It improved the prediction accuracy and adaptability of thick steel plates, reduced production fluctuations, shortened the commissioning cycle, and improved product quality and market competitiveness.
Smart Images
Figure CN121920185A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of medium and heavy plate rolling and cooling control technology, specifically to a method for constructing a heat transfer coefficient model for a medium and heavy plate post-rolling cooling system. Background Technology
[0002] Post-rolling cooling of medium and heavy plates is a crucial step in the production process. It refers to the controlled cooling treatment of steel plates after rolling through a specific cooling system (such as laminar flow cooling, ultra-fast cooling, etc.) to regulate the temperature change process of the steel plates, thereby affecting the microstructure transformation, mechanical properties (such as strength and toughness), residual stress, and surface quality. It is an important process to ensure that the quality of medium and heavy plate products meets the needs of downstream industries (such as coal mining machinery, shipbuilding, bridges, pressure vessels, high-strength building steel, etc.).
[0003] Existing heat transfer coefficient models for post-rolling cooling systems of medium and heavy plates have several shortcomings: the initial temperature field ignores the temperature gradient along the thickness direction, resulting in a red-hot temperature hit rate of less than 85% for steel plates thicker than 40mm, which cannot meet the strength fluctuation requirements of high-end equipment manufacturing; the heat transfer coefficient calculation does not consider changes in water volume and pressure, making it difficult to adapt to ultra-fast cooling equipment with high-low pressure switching capabilities, requiring relearning after parameter adjustments; the physical property parameters do not consider the influence of chemical composition, leading to significant accuracy fluctuations in billet re-rolling scenarios; explicit differential calculations are inefficient and prone to divergence due to CFL limitations; the feedforward adaptive capability is weak, resulting in a delayed response to sudden changes in parameters such as starting cooling temperature and water temperature; historical data has poor universality, requiring data to be accumulated from scratch after new production lines or equipment modifications, leading to long debugging cycles. Based on these problems, a method for constructing a heat transfer coefficient model for post-rolling cooling systems of medium and heavy plates is proposed. Summary of the Invention
[0004] To address the aforementioned issues, this invention provides a method for constructing a heat transfer coefficient model for a medium-thick plate post-rolling cooling system. By constructing an initial temperature field that considers the temperature gradient in the thickness direction and establishing a heat transfer coefficient correlation model with dynamic flow response, the heat transfer coefficient model can accurately predict the performance of thick steel plates, overcoming the limitations of traditional models in terms of accuracy, adaptability, and debugging cycle.
[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A method for constructing a heat transfer coefficient model for a medium-thick plate post-rolling cooling system, comprising the following steps:
[0006] Step 1, Constructing the initial temperature field: Based on the surface temperature and thickness information of the steel plate after final rolling, and combined with the temperature self-organization characteristics during the air cooling process, an initial temperature field reflecting the temperature distribution in the thickness direction is generated.
[0007] Step 2, Dynamically calculate the heat transfer coefficient: Based on the water supply pressure, flow rate and nozzle structure parameters of the cooling system, by analyzing the flow characteristics of the cooling medium, the ratio of the flow rate to the reference flow rate is introduced as a dynamic correction term to establish a correlation model of the heat transfer coefficient with the change of cooling medium parameters.
[0008] Step 3, optimize physical property parameters: take the carbon content of the steel plate as the key variable, and construct thermal conductivity calculation model and segmented specific heat capacity calculation model that take into account the influence of carbon content, respectively, to adapt to steel with different chemical compositions;
[0009] Step 4, Model self-learning training: Develop a batch self-learning function module to preprocess and incrementally learn historical production data to achieve autonomous optimization of model parameters;
[0010] Step 5, Model Validation and Iteration: Verify the model accuracy based on the key indicator of the red return temperature measured on-site, and iteratively adjust the model parameters until the preset accuracy requirements are met.
[0011] Furthermore, in step 1, the method of combining the temperature self-organization characteristics during the air cooling process is as follows: the law that the internal temperature field of steel plates with different initial temperature distributions tends to be consistent when air-cooled to the same surface temperature is determined by simulation calculation, and the iterative calculation process of the initial temperature field is simplified based on this law.
[0012] Furthermore, in step 2, the analysis of the flow characteristics of the cooling medium includes: determining the total system flow rate based on the law of conservation of mass and the pressure balance equation in fluid mechanics. Nozzle cross-sectional area With nozzle outlet flow rate The relationship between the total system flow rate and the nozzle outlet velocity satisfies: Furthermore, under constant flow control, the nozzle outlet velocity is determined only by the flow rate and the nozzle cross-sectional area.
[0013] Furthermore, in step 2, the specific formula for establishing the correlation model of heat transfer coefficient as a function of cooling medium parameters is as follows:
[0014]
[0015] Where h is the heat transfer coefficient, α is the thermal diffusivity, and Q is the actual flow rate. As the baseline flow rate, The surface temperature of the steel plate. The temperature of the cooling medium. is the specific heat capacity coefficient of the cooling medium.
[0016] Furthermore, in step 3, the formula for calculating the thermal conductivity considering the influence of carbon content is as follows:
[0017]
[0018] in, Let be the thermal conductivity of node i. Let be the temperature of node i, and C be the carbon content of the steel plate.
[0019] Furthermore, in step 3, the segmented specific heat capacity calculation model is a four-segment model, specifically including:
[0020] When Temp < 650℃ ;
[0021] When 650℃<Temp≤768℃ ;
[0022] When 768℃<Temp≤900℃ ;
[0023] When Temp > 900℃ ;
[0024] in, is the specific heat capacity, and Temp is the temperature of the steel plate.
[0025] Furthermore, in step 4, the working process of the batch self-learning function module includes: after initializing the parameters, connecting to the database to obtain steel plate production data blocks, performing preprocessing on the data such as missing value handling and outlier removal, using an incremental learning algorithm to train the model, recording learning logs and evaluation results, and outputting optimized model parameters when the number of learned data blocks reaches a preset value N.
[0026] Furthermore, in step 2, the theoretical basis for introducing the ratio of flow rate to baseline flow rate as a dynamic correction term includes: based on the intensified laminar flow cooling criterion equation:
[0027]
[0028] Where Nu is the Nusselt number, Re is the Reynolds number, and Pr is the Prandtl number, the heat transfer coefficient is derived to be proportional to the 0.342nd power of the flow velocity through dimensionless parameter conversion, and then converted to be proportional to the 0.342nd power of the flow rate.
[0029] Furthermore, in step 3, when constructing the calculation model for thermal conductivity and specific heat capacity that takes into account the influence of carbon content, regression analysis needs to be performed based on experimental data of typical steel grades. The typical steel grades include rimmed steel, killed steel, low carbon steel, and medium carbon steel. The experimental data cover the measured values of thermal conductivity and specific heat capacity in the temperature range of 400℃ to 1100℃.
[0030] Furthermore, in step 5, the preset accuracy requirements include: for steel plates with a thickness ≤ 40mm, the red-hot temperature hit rate is not less than 96%; for steel plates with a thickness > 40mm, the red-hot temperature hit rate is not less than 93%.
[0031] The above approach has the following beneficial effects:
[0032] 1. This scheme effectively improves the model's prediction accuracy for thick steel plates by constructing an initial temperature field that considers the temperature gradient in the thickness direction and combining it with the temperature self-organization characteristics of the air cooling process. It solves the problem of insufficient control accuracy for thick plates caused by the neglect of temperature gradient in traditional models, and makes the temperature prediction of the cooling process more in line with the actual situation.
[0033] 2. This scheme uses a dynamic calculation method for the heat transfer coefficient, introducing the ratio of flow rate to reference flow rate as a dynamic correction term. It establishes a correlation model between the heat transfer coefficient and cooling medium parameters, allowing the model to adapt to different operating conditions such as water supply pressure and flow rate. It is especially suitable for ultra-fast cooling equipment with high and low pressure switching function, and can maintain good accuracy without frequent relearning, thus enhancing the adaptability and flexibility of the model.
[0034] 3. In this scheme, the carbon content of steel plate is used as a key variable to optimize the physical property parameters. The thermal conductivity and segmented specific heat capacity calculation models that take into account the influence of carbon content are constructed respectively, so that the model can be adapted to steel with different chemical compositions, improving its versatility in flexible rolling scenarios and reducing the impact of fluctuations in the chemical composition of raw materials on the accuracy of the model.
[0035] 4. This solution, through model verification and iteration, continuously optimizes parameters based on key indicators of the red-hot temperature measured on-site, ensuring that the model accuracy meets actual production needs. It provides a more accurate and reliable control basis for the cooling process after rolling of medium and heavy plates, which helps to improve the mechanical properties, microstructure uniformity and residual stress distribution of steel plates, thereby enhancing product quality and market competitiveness. Attached Figure Description
[0036] Figure 1 This is a schematic diagram of the method steps in an embodiment of the method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium and thick plates according to the present invention;
[0037] Figure 2 This is a schematic diagram of three temperature distributions of a 40mm steel plate, illustrating an embodiment of the method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to the present invention.
[0038] Figure 3 This is a schematic diagram of the temperature field of a 40mm steel plate cooled to 800℃ on the surface under three different conditions, which is an embodiment of the construction method of the heat transfer coefficient model of the post-rolling cooling system of the medium-thick plate of the present invention.
[0039] Figure 4This is a schematic diagram of the SUPIC model and the thermal conductivity curves of typical steel grades, representing an embodiment of the method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium and heavy plates in this invention.
[0040] Figure 5 This is a schematic diagram of the SUPIC model and the specific heat value curve of a typical steel grade, representing an embodiment of the construction method of the heat transfer coefficient model of the post-rolling cooling system for thick plates in this invention. Detailed Implementation
[0041] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0042] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.
[0043] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0044] The following detailed description illustrates the specific implementation method:
[0045] Example:
[0046] In the post-rolling cooling process of medium and heavy plates, the accuracy of the heat transfer coefficient model directly determines the stability of the steel plate's performance. Traditional models have many limitations: First, the initial temperature field ignores the temperature gradient along the thickness direction, resulting in a red-hot temperature hit rate of less than 85% for thick steel plates (>40mm), which cannot meet the requirements of high-end equipment manufacturing for strength fluctuations ≤30MPa; Second, the heat transfer coefficient calculation does not consider changes in water volume and pressure, making it difficult to adapt to ultra-fast cooling equipment with high and low pressure switching functions such as SUPIC-DQ+ACC, requiring relearning every time parameters are adjusted; Third, the physical property parameters do not consider the influence of chemical composition, resulting in significant accuracy fluctuations in billet re-rolling scenarios (such as pipeline steel X65 being re-rolled into Q355B); Fourth, explicit differential calculation is limited by CFL conditions, leading to low efficiency and easy divergence in thick plate calculations; Fifth, the feedforward adaptive capability is weak, resulting in a delayed response to sudden changes in parameters such as starting cooling temperature and water temperature; Sixth, the historical data has poor universality, requiring data to be accumulated from scratch after new production lines or equipment modifications, with a debugging cycle of more than 3 months.
[0047] To address the aforementioned issues, the inventors proposed a method for constructing a heat transfer coefficient model that integrates mechanistic modeling and data-driven approaches. This method optimizes the initial temperature field by considering the temperature gradient along the thickness direction, establishes a heat transfer coefficient correlation model based on dynamic flow response, embeds carbon content to correct physical property parameters, employs implicit differential algorithms to improve computational stability, introduces a random forest algorithm for adaptive adjustment, and develops a batch self-learning function to shorten the debugging cycle. Ultimately, this method increases the accuracy of the red-hot temperature hit rate for thick steel plates to over 93%, and reduces the debugging cycle of new production lines to one month, meeting the production requirements for high precision, flexibility, and rapid deployment.
[0048] This embodiment uses the method applied to the production of 40mm thick Q355B steel plates in a steel plant as an example. Traditional models, due to neglecting the temperature gradient along the thickness direction, often have a prediction deviation of over 50℃ for the red-hot temperature. Therefore, specifically, as shown in the attached... Figure 1 As shown: A method for constructing a heat transfer coefficient model for a medium-thick plate post-rolling cooling system, comprising the following steps:
[0049] Step 1, Constructing the initial temperature field: Based on the surface temperature and thickness information of the steel plate after final rolling, and combined with the temperature self-organization characteristics during the air cooling process, the method of combining the temperature self-organization characteristics during the air cooling process is as follows: through simulation calculation, the law that the internal temperature field of steel plates with different initial temperature distributions tends to be consistent when air-cooled to the same surface temperature is determined. Based on this law, the iterative calculation process of the initial temperature field is simplified to generate an initial temperature field that reflects the temperature distribution in the thickness direction.
[0050] First, the surface temperature of 800℃ was obtained by surface temperature measurement after final rolling. Based on the self-organization characteristics of air cooling temperature, the air cooling process was simulated and calculated for three initial temperature distributions (average 1000℃ with no gradient, average 1000℃ with a center-to-surface temperature difference of 185℃, and average 1100℃ with no gradient). (See attached diagram) Figure 2 and Figure 3 As shown, the results indicate that when the surface temperature drops to 800℃, the internal temperature fields in the three cases tend to be consistent (the core temperature is 820±5℃). Based on this, the iterative calculation is simplified: the initial core temperature is set to the surface temperature + 200℃, and the temperature field is iteratively updated with a step size of 0.5s until the surface temperature matches the measured value. After applying this method, the accuracy of the red-hot temperature prediction for 40mm steel plates increased from 80% to 94%, solving the problem of the inability to use the automatic model for thick plates. By utilizing the temperature self-organization characteristics of steel plates during air cooling—steel plates with different initial temperature distributions will have their internal temperature fields tend to be consistent through heat conduction when air-cooled to the same surface temperature—an accurate initial field can be quickly constructed without complex full-process temperature tracing. Compared to the uniform temperature field assumed by traditional models, its prediction error for the core temperature of thick plates is reduced by 40%.
[0051] Because traditional models, with their fixed heat transfer coefficient, exhibit a 30°C temperature deviation when switching water pressures, step 2 of this scheme optimizes the dynamic calculation of the heat transfer coefficient. Based on the cooling system's water supply pressure, flow rate, and nozzle structure parameters, the flow characteristics of the cooling medium are analyzed. This analysis includes determining the total system flow rate based on the law of conservation of mass and pressure balance equations in fluid mechanics. Nozzle cross-sectional area With nozzle outlet flow rate The relationship between the total system flow rate and the nozzle outlet velocity satisfies: Furthermore, under constant flow control, the nozzle outlet velocity is determined only by the flow rate and the nozzle cross-sectional area.
[0052] The ratio of flow rate to baseline flow rate is introduced as a dynamic correction term. The theoretical basis for this is based on the intensified laminar flow cooling criterion equation:
[0053]
[0054] Where Nu is the Nusselt number, Re is the Reynolds number, and Pr is the Prandtl number, the heat transfer coefficient is derived to be proportional to the 0.342nd power of the flow velocity through dimensionless parameter transformation, and further converted to be proportional to the 0.342nd power of the flow rate. Based on the intensified laminar flow cooling criterion equation, the heat transfer coefficient is derived to be proportional to the 0.342nd power of the flow rate. A correlation model of the heat transfer coefficient with the cooling medium parameters is established. The specific formula for establishing the correlation model of the heat transfer coefficient with the cooling medium parameters is as follows:
[0055]
[0056] Where h is the heat transfer coefficient, α is the thermal diffusivity, and Q is the actual flow rate. As the baseline flow rate, The surface temperature of the steel plate. The temperature of the cooling medium. This represents the specific heat capacity coefficient of the cooling medium. The reference flow rate is... Set as design flow In actual production, when the flow rate in the low-pressure section is adjusted to... The model automatically calculates correction coefficients and updates heat transfer coefficients in real time. After application, when the rolling mill switches between medium pressure (200 m³ / h) and low pressure (100 m³ / h), the longitudinal temperature difference decreases from 25℃ to 18℃, and the red-hot temperature hit rate remains stable at over 96%. Compared to traditional models that need to relearn for different flow rates, the new model requires no manual intervention and adapts to the needs of frequent adjustments to cooling parameters in flexible production.
[0057] Step 3, Optimize physical property parameters: Using the carbon content of the steel plate as a key variable, for example, when the carbon content is 0.18% and 0.22%, the traditional model, because it does not consider the influence of carbon content, results in a 20°C fluctuation in the specific heat capacity calculation. Therefore, a thermal conductivity calculation model that considers the influence of carbon content and a segmented specific heat capacity calculation model are constructed to adapt to steels with different chemical compositions (such as...). Figure 4 As shown); where:
[0058] The formula for calculating thermal conductivity considering the effect of carbon content is as follows:
[0059]
[0060] in, Let be the thermal conductivity of node i. Let be the temperature of node i, and C be the carbon content of the steel plate; when the carbon content increases from 0.18% to 0.22%, the thermal conductivity at 800℃ decreases from 32W / (m·K) to 30.5W / (m·K), with a deviation from the measured data of ≤3%;
[0061] like Figure 5As shown, for specific heat capacity, a four-segment calculation model is adopted, specifically including:
[0062] When Temp < 650℃ ;
[0063] When 650℃<Temp≤768℃ ;
[0064] When 768℃<Temp≤900℃ ;
[0065] When Temp > 900℃ ;
[0066] in, , where is the specific heat capacity, Temp is the steel plate temperature, and C is the carbon content; a carbon content correction term is introduced in the 650-900℃ phase transformation range, and the calculated specific heat capacity of 0.22% carbon steel plate is reduced by 5% compared with the traditional model, which is closer to the actual phase transformation heat absorption characteristics;
[0067] When constructing a calculation model for thermal conductivity and specific heat capacity that considers the influence of carbon content, regression analysis is required based on experimental data from typical steel grades, including rimmed steel, killed steel, low-carbon steel, and medium-carbon steel. The experimental data cover measured values of thermal conductivity and specific heat capacity in the temperature range of 400℃ to 1100℃. After optimization, the standard deviation of the red-hot temperature of Q355B steel plates produced from the two billets decreased from 18℃ to 8℃, solving the accuracy problem caused by chemical composition fluctuations in flexible rolling. Compared with traditional methods that rely on historical data matching, the new model has improved generalization ability and shortened the learning cycle by 60%.
[0068] Step 4, Model Self-Learning Training: Develop a batch self-learning function module to preprocess and incrementally learn historical production data, enabling autonomous optimization of model parameters. In the debugging of new projects, the batch self-learning function is used to import 3,000 sets of historical data from similar production lines, and the model parameters are quickly optimized through incremental learning, achieving stable accuracy in just 35 days.
[0069] Step 5, Model Validation and Iteration: Verify the model accuracy based on the key indicators of the red-hot temperature measured on-site, and iteratively adjust the model parameters until the preset accuracy requirements are met. The preset accuracy requirements include: for steel plates with a thickness ≤ 40mm, the red-hot temperature hit rate is not less than 96%; for steel plates with a thickness > 40mm, the red-hot temperature hit rate is not less than 93%.
[0070] This solution achieves three major breakthroughs through multi-dimensional optimization: First, it improves the control precision of thick steel plates, increasing the red-hot temperature hit rate of steel plates over 40mm from below 85% to over 93%, meeting the needs of high-end equipment manufacturing; second, it enhances the adaptability of flexible production, enabling it to cope with scenarios such as billet re-rolling and parameter mutations, reducing precision fluctuations by 50%; and third, it improves engineering efficiency, reducing the new production line commissioning cycle from 3 months to 1 month, and achieving a historical data reuse rate of 70%. Through the deep integration of mechanisms and data, it provides core technical support for the intelligent upgrading of post-rolling cooling of medium and heavy plates.
[0071] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.
Claims
1. A method for constructing a heat transfer coefficient model for a medium-thick plate post-rolling cooling system, characterized in that, Includes the following steps: Step 1, Construct the initial temperature field: Based on the surface temperature and thickness information of the steel plate after final rolling, and combined with the temperature self-organization characteristics during the air cooling process, an initial temperature field reflecting the temperature distribution in the thickness direction is generated. Step 2, Dynamically calculate the heat transfer coefficient: Based on the water supply pressure, flow rate and nozzle structure parameters of the cooling system, by analyzing the flow characteristics of the cooling medium, the ratio of the flow rate to the reference flow rate is introduced as a dynamic correction term to establish a correlation model of the heat transfer coefficient with the change of cooling medium parameters. Step 3, optimize physical property parameters: take the carbon content of the steel plate as the key variable, and construct thermal conductivity calculation model and segmented specific heat capacity calculation model that take into account the influence of carbon content, respectively, to adapt to steel with different chemical compositions; Step 4, Model self-learning training: Develop a batch self-learning function module to preprocess and incrementally learn historical production data to achieve autonomous optimization of model parameters; Step 5, Model Validation and Iteration: Verify the model accuracy based on the key indicators of the red return temperature measured on-site, and iteratively adjust the model parameters until the preset accuracy requirements are met.
2. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 1, characterized in that, In step 1, the method of combining the temperature self-organization characteristics during the air cooling process is as follows: the law that the internal temperature field of steel plates with different initial temperature distributions tends to be consistent when air-cooled to the same surface temperature is determined by simulation calculation, and the iterative calculation process of the initial temperature field is simplified based on this law.
3. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 2, characterized in that, Step 2 involves analyzing the flow characteristics of the cooling medium, including determining the total system flow rate based on the law of conservation of mass and the pressure balance equation in fluid mechanics. Nozzle cross-sectional area With nozzle outlet flow rate The relationship between the total system flow rate and the nozzle outlet velocity satisfies: Furthermore, under constant flow control, the nozzle outlet velocity is determined only by the flow rate and the nozzle cross-sectional area.
4. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 3, characterized in that, In step 2, the specific formula for establishing the correlation model of heat transfer coefficient with the variation of cooling medium parameters is as follows: Where h is the heat transfer coefficient, α is the thermal diffusivity, and Q is the actual flow rate. As the baseline flow rate, The surface temperature of the steel plate. The temperature of the cooling medium. is the specific heat capacity coefficient of the cooling medium.
5. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 4, characterized in that, In step 3, the formula for calculating thermal conductivity considering the effect of carbon content is as follows: in, Let be the thermal conductivity of node i. Let be the temperature of node i, and C be the carbon content of the steel plate.
6. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 5, characterized in that, In step 3, the segmented specific heat capacity calculation model is a four-segment model, specifically including: When Temp < 650℃ ; When 650℃ < Temp ≤ 768℃ ; When 768℃<Temp≤900℃ ; When Temp > 900℃ ; in, , where is the specific heat capacity, Temp is the temperature of the steel plate, and C is the carbon content.
7. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 6, characterized in that, In step 4, the working process of the batch self-learning function module includes: after initializing the parameters, connecting to the database to obtain steel plate production data blocks, performing preprocessing on the data such as missing value handling and outlier removal, using an incremental learning algorithm to train the model, recording learning logs and evaluation results, and outputting optimized model parameters when the number of learned data blocks reaches a preset value N.
8. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 7, characterized in that, In step 2, the theoretical basis for introducing the ratio of flow rate to baseline flow rate as a dynamic correction term includes: based on the intensified laminar flow cooling criterion equation: Where Nu is the Nusselt number, Re is the Reynolds number, and Pr is the Prandtl number, the heat transfer coefficient is derived to be proportional to the 0.342nd power of the flow velocity through dimensionless parameter conversion, and then converted to be proportional to the 0.342nd power of the flow rate.
9. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 8, characterized in that, In step 3, when constructing the calculation model for thermal conductivity and specific heat capacity that takes into account the influence of carbon content, regression analysis needs to be performed based on experimental data of typical steel grades. The typical steel grades include rimmed steel, killed steel, low carbon steel and medium carbon steel. The experimental data cover the measured values of thermal conductivity and specific heat capacity in the temperature range of 400℃ to 1100℃.
10. The method for constructing the heat transfer coefficient model of the post-rolling cooling system for medium-thick plates according to claim 9, characterized in that, In step 5, the preset accuracy requirements include: for steel plates with a thickness ≤ 40mm, the red-hot temperature hit rate is not less than 96%; for steel plates with a thickness > 40mm, the red-hot temperature hit rate is not less than 93%.