Control method, system and equipment of heating furnace and medium

Abstract

The invention discloses a control method and system of a heating furnace, computer equipment and a medium. The method comprises the steps that state parameters of a current steel billet are obtained, wherein the state parameters comprise the steel grade, the width, the length, the weight, the charging temperature, the target discharging temperature and the finished product thickness; matching a plurality of samples of the corresponding steel grade in a pre-established excellent operation mode library based on the steel grade in the state parameters, wherein each sample comprises the state parameter and the operation parameter of the sample; and the similarity between the state parameters of the steel billet and the state parameters of each sample is calculated, and the operation parameters corresponding to the samples with the similarity not smaller than a threshold value serve as the actual operation parameters of the heating furnace. According to the scheme provided by the embodiment of the invention, accurate recommendation and dynamic optimization of the process parameters of the heating furnace are realized by constructing the excellent operation mode library and the decision-making mechanism.

Description

Technical Field

[0001] This invention relates to the field of iron and steel metallurgical control, specifically to a control method, system, equipment, and medium for a heating furnace. Background Technology

[0002] In hot strip mill production lines, the heating furnace, as a core piece of equipment, plays a crucial role in heating steel billets to the target tapping temperature, accounting for over 65% of the total energy consumption of the entire production line. Currently, the industry faces a prominent problem of generally low energy efficiency, with actual operating thermal efficiency mostly maintained in the 20%-30% range, and only a few optimized process parameters reaching close to 40%, resulting in significant energy waste. This inefficiency severely restricts the green and low-carbon transformation of the steel industry.

[0003] Furthermore, the internal thermodynamic processes of the heating furnace are highly complex, involving the coupling of multiple physical fields such as heat conduction, radiation, convection, and combustion reactions. These processes are also affected by differences in operator experience, leading to unstable billet heating temperature control, high energy consumption, and significant burn-off. Existing research often employs mechanistic modeling or simple self-learning control, which is insufficient to adapt to the actual production conditions of multiple steel grades, specifications, and varying operating conditions.

[0004] Therefore, there is an urgent need for an intelligent recommendation solution that can balance temperature hit rate, energy consumption optimization, and process stability. Summary of the Invention

[0005] In view of this, in order to overcome at least one aspect of the above problems, embodiments of the present invention provide a method for controlling a heating furnace, comprising the following steps: Obtain the current state parameters of the steel billet, wherein the state parameters include steel grade, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness; Based on the steel grade in the state parameters, multiple samples of the corresponding steel grade are matched in a pre-established library of superior operating modes, wherein each sample includes the sample's state parameters and operating parameters. Calculate the similarity between the state parameters of the billet and the state parameters of each sample, and take the operating parameters corresponding to the samples with similarity not less than a threshold as the actual operating parameters of the heating furnace.

[0006] In some embodiments, the method further includes: In response to the absence of samples with a similarity greater than the threshold, the state parameters of the billet are input into the furnace steel temperature prediction model to predict the furnace temperature. The particle swarm optimization algorithm is used to find multiple predicted operating parameters in the operating parameter space, and each predicted operating parameter is input into the unit consumption prediction model to obtain multiple predicted values ​​of gas unit consumption. The comprehensive operating condition evaluation index is obtained by calculating each predicted gas consumption value and the furnace outlet temperature. The operating parameter corresponding to the largest comprehensive operating condition evaluation index shall be used as the actual operating parameter of the heating furnace.

[0007] In some embodiments, calculating a comprehensive operating condition evaluation index by combining each predicted gas consumption value with the furnace outlet temperature further includes: The first score is calculated based on the following formula:

[0008] Wherein, y1 is the furnace exit temperature, and y2 is the target furnace exit temperature; The second score is calculated based on the following formula:

[0009] in, i 1. Predicted unit gas consumption; based on The comprehensive working condition evaluation index was obtained.

[0010] In some embodiments, the method further includes: The newly optimized and superior operation mode data obtained through optimization will be temporarily stored in the temporary data storage database. The furnace steel temperature prediction model and the unit consumption prediction model are backtested and verified at preset time intervals. If the backtest verification passes, the operation model in the temporary database is added to the excellent database.

[0011] In some embodiments, the method further includes constructing a library of superior operating modes, wherein the step of constructing the library of superior operating modes includes: Collect historical production data, each of which includes slab condition parameters, furnace operating parameters, and process evaluation parameters; The heating efficiency of each historical production data is determined based on the furnace operating parameters; the temperature control accuracy of each historical production data is determined based on the slab state parameters and process evaluation parameters; and the comprehensive operating condition evaluation index is determined based on the process evaluation parameters. Based on the heating efficiency, temperature control accuracy, and comprehensive operating condition evaluation indicators, excellent historical production data are selected, and a library of excellent operating modes is constructed based on the excellent historical production data.

[0012] In some embodiments, the method further includes: Determine the quantity of excellent historical production data for the same steel grade; If the number of excellent historical production data for the same steel grade does not exceed a threshold, each excellent historical production data point will be treated as a sample. If the number of excellent historical production data for the same steel grade exceeds a threshold, multiple cluster centers are obtained by using the mean clustering algorithm. The other excellent historical production data are then assigned to the corresponding cluster centers as subsets, and each cluster center is treated as a sample.

[0013] In some embodiments, calculating the similarity between the state parameters of the billet and the state parameters of each sample, and using the operating parameters corresponding to the samples with a similarity not less than a threshold as the actual operating parameters of the heating furnace, further includes: If the sample is a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center is not less than a threshold, then the operating parameters corresponding to the cluster center are directly used as the actual operating parameters of the heating furnace. In response to the sample being a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center being less than a threshold, the similarity between each excellent historical production data in the subset and the state parameters of the billet is calculated, and the operating parameters corresponding to the excellent historical production data in the subset with a similarity not less than the threshold are used as the actual operating parameters of the heating furnace.

[0014] Based on the same inventive concept, according to another aspect of the present invention, embodiments of the present invention also provide a control system for a heating furnace, comprising: The acquisition module is configured to acquire the current billet status parameters, wherein the status parameters include steel grade, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness. The matching module is configured to match multiple samples of the corresponding steel type in a pre-established library of superior operating modes based on the steel type in the state parameters, wherein each sample includes the sample's state parameters and operating parameters. The calculation module is configured to calculate the similarity between the state parameters of the billet and the state parameters of each sample, and to use the operating parameters corresponding to the samples with similarity not less than a threshold as the actual operating parameters of the heating furnace.

[0015] Based on the same inventive concept, according to another aspect of the present invention, embodiments of the present invention also provide a computer device, comprising: At least one processor; and The memory stores a computer program that can run on the processor, which, when executing the program, performs the steps of any of the furnace control methods described above.

[0016] Based on the same inventive concept, according to another aspect of the present invention, embodiments of the present invention also provide a computer-readable storage medium storing a computer program that, when executed by a processor, performs the steps of any of the control methods for a heating furnace as described above.

[0017] The present invention has one of the following beneficial technical effects: the solution proposed in the embodiments of the present invention achieves accurate recommendation and dynamic optimization of heating furnace process parameters by constructing an excellent operation mode library and decision-making mechanism. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic flowchart of a control method for a heating furnace provided in an embodiment of the present invention; Figure 2 Statistics on the temperature deviation and different temperature deviation ranges between the predicted furnace exit temperature and the target furnace exit steel temperature provided for embodiments of the present invention; Figure 3 A comparison of the percentage of different intervals between the predicted unit consumption value and the actual value provided for embodiments of the present invention; Figure 4 The present invention provides a comparison of heating temperatures for different work groups and sections in an embodiment of the present invention, wherein (a) is the average heating temperature and (b) is the median heating temperature; Figure 5 The present invention provides a comparison of heating time for different work groups and sections in an embodiment of the present invention, wherein (a) is the average heating time and (b) is the median heating time.

[0020] Figure 6 A comparison of comprehensive heating operating conditions and temperature deviations for different work shifts provided in the embodiments of the present invention; Figure 7 A schematic diagram of the heating furnace control and communication architecture provided for an embodiment of the present invention; Figure 8 A schematic diagram of the control system of the heating furnace provided for an embodiment of the present invention; Figure 9 A schematic diagram of the structure of a computer device provided for an embodiment of the present invention; Figure 10 A schematic diagram of the structure of a computer-readable storage medium provided for an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to specific examples and the accompanying drawings.

[0022] It should be noted that all uses of "first" and "second" in the embodiments of the present invention are for the purpose of distinguishing two entities or parameters with the same name but different names. It is clear that "first" and "second" are only for the convenience of expression and should not be construed as limiting the embodiments of the present invention. Subsequent embodiments will not explain this in detail.

[0023] According to one aspect of the present invention, embodiments of the present invention provide a method for controlling a heating furnace, such as... Figure 1 As shown, it may include the following steps: S1, Obtain the current state parameters of the steel billet, wherein the state parameters include steel grade, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness; S2, based on the steel grade in the state parameters, match multiple samples of the corresponding steel grade in a pre-established excellent operation mode library, wherein each sample includes the sample's state parameters and operation parameters; S3, calculate the similarity between the state parameters of the billet and the state parameters of each sample, and take the operating parameters corresponding to the samples with similarity not less than the threshold as the actual operating parameters of the heating furnace.

[0024] The solution proposed in this invention achieves accurate recommendation and dynamic optimization of heating furnace process parameters by constructing a superior operation mode library and decision-making mechanism.

[0025] In some embodiments, the method further includes: In response to the absence of samples with a similarity greater than the threshold, the state parameters of the billet are input into the furnace steel temperature prediction model to predict the furnace temperature. The particle swarm optimization algorithm is used to find multiple predicted operating parameters in the operating parameter space, and each predicted operating parameter is input into the unit consumption prediction model to obtain multiple predicted values ​​of gas unit consumption. The comprehensive operating condition evaluation index is obtained by calculating each predicted gas consumption value and the furnace outlet temperature. The operating parameter corresponding to the largest comprehensive operating condition evaluation index shall be used as the actual operating parameter of the heating furnace.

[0026] Specifically, the excellent operation mode library pre-stores multiple excellent operation samples. Each excellent sample includes the sample's state parameters (steel grade, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness) and operation parameters (heating time for each stage, such as heat recovery, preheating, heating, and soaking, and furnace temperature setpoints for each stage). Thus, when the current billet's state parameters (steel grade, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness) are obtained, the corresponding steel grade module in the excellent operation mode library is located based on the steel grade. The similarity is calculated by calculating the state parameters, and then the corresponding sample is selected based on the similarity threshold (e.g., 0.95).

[0027] If there are samples with a similarity greater than the threshold, they can be output directly, and the operating parameters recorded in the samples can be used to control the heating furnace. If no samples with a similarity greater than the threshold are found, such as when encountering a new steel grade or a special specification slab, the pre-trained furnace temperature prediction model and unit gas consumption prediction model are invoked. These two models use state parameters and operating parameters as inputs to predict the furnace temperature and unit gas consumption, respectively.

[0028] Then, the Particle Swarm Optimization (PSO) algorithm is used to optimize the operating parameters. The operating parameters obtained by the PSO algorithm and the input state parameters are passed through a pre-trained furnace steel temperature prediction model and a gas consumption prediction model to output predicted values ​​for furnace temperature and gas consumption. Based on these predicted values, the comprehensive operating condition evaluation index J is further calculated. In this way, the PSO optimization algorithm iteratively searches within the operating parameter space, continuously calling the prediction model to evaluate the J value under different combinations of operating parameters, ultimately finding the optimal operating parameters that maximize J for recommendation.

[0029] In some embodiments, calculating a comprehensive operating condition evaluation index by combining each predicted gas consumption value with the furnace outlet temperature further includes: The first score is calculated based on the following formula:

[0030] Wherein, y1 is the furnace exit temperature, and y2 is the target furnace exit temperature; The second score is calculated based on the following formula:

[0031] in, i 1. Predicted unit gas consumption; based on The comprehensive working condition evaluation index was obtained.

[0032] Specifically, in order to scientifically and quantitatively evaluate the merits of a set of operating parameters, this invention defines a comprehensive operating condition evaluation index (...). This indicator mainly examines two process parameters: the deviation between the furnace outlet temperature (y1) and the target furnace outlet temperature (y2), and the gas consumption per unit area. i 1) The specific calculation formula is as follows:

[0033]

[0034]

[0035] In some embodiments, the method further includes: The newly optimized and superior operation mode data obtained through optimization will be temporarily stored in the temporary data storage database. The furnace steel temperature prediction model and the unit consumption prediction model are backtested and verified at preset time intervals. If the backtest verification passes, the operation model in the temporary database is added to the excellent database.

[0036] Specifically, after the optimization decision is initiated and the optimal operating parameters are obtained, the new superior operating mode data obtained from the optimization is temporarily stored in the data temporary storage database. After the two models are verified through backtesting, they are then stored in the superior database.

[0037] In this embodiment, backtesting can be performed once every preset period, such as half a year. Backtesting is performed using historical data that was not used in training (50 data points are randomly selected from each month, for a total of 300 data points). This includes benchmark verification of the prediction model, testing of the operation parameter recommendation system, and virtual production simulation in a digital twin environment.

[0038] The benchmark validation of the prediction models involves inputting the state parameters and corresponding actual operation parameters from the independent test set into the pre-trained furnace steel temperature prediction model and unit consumption prediction model, respectively, to obtain the predicted values ​​corresponding to the current actual operation, and calculating the relative error between the predicted and actual values. If the error of the furnace steel temperature prediction model is less than 5% and the error of the unit consumption prediction model is less than 10%, the generalization ability and stability of the two basic prediction models on unseen data are confirmed, establishing a reliable evaluation benchmark for subsequent system evaluation. In this embodiment, the average relative error of the furnace steel temperature prediction model is only 0.58%, and the average relative error of the unit consumption prediction model is 6.83%.

[0039] The operating parameter recommendation system test involves inputting the same test set—this time using only its state parameters—line by line into the heating furnace operating parameter recommendation system. Based on its built-in excellent operating mode library and intelligent optimization algorithm, the system generates an optimal set of recommended operating parameters for each set of operating conditions, achieving a recommendation success rate of over 90%. For example... Figure 2As shown, in this embodiment, the system's recommendation success rate reaches 94.33%, and 90.11% of the recommended solutions have a furnace temperature deviation within 50°C.

[0040] In a digital twin environment, virtual production simulation involves recombining the system-recommended operating parameters from the previous step with the original state parameters and then inputting them again into the previously validated prediction model for furnace steel temperature and unit consumption. By comparing the prediction results using the "recommended parameters" with those using the actual parameters from the first step, the overall performance of the recommendation system in ensuring process compliance and improving energy efficiency can be quantitatively evaluated in the virtual space. Figure 3 As shown, in this embodiment, the average unit energy consumption reduction under the recommended parameters reaches 14.67%, and the energy consumption distribution is significantly concentrated in the low-consumption range.

[0041] In some embodiments, the method further includes constructing a library of superior operating modes, wherein the step of constructing the library of superior operating modes includes: Collect historical production data, each of which includes slab condition parameters, furnace operating parameters, and process evaluation parameters; The heating efficiency of each historical production data is determined based on the furnace operating parameters; the temperature control accuracy of each historical production data is determined based on the slab state parameters and process evaluation parameters; and the comprehensive operating condition evaluation index is determined based on the process evaluation parameters. Based on the heating efficiency, temperature control accuracy, and comprehensive operating condition evaluation indicators, excellent historical production data are selected, and a library of excellent operating modes is constructed based on the excellent historical production data.

[0042] Specifically, historical production data was first collected from the heating furnace production control system (such as the L2 system) and process database. This included over 100,000 data entries covering the entire furnace cycle from slab entry to exit. The data was then categorized into three types: Slab condition parameters: including steel grade, slab width, length, thickness, weight, furnace entry temperature, target furnace exit temperature, and finished product thickness, etc. Heating furnace operating parameters: including heating shift information, heating time of each section (heat recovery, preheating, heating, soaking, etc.), furnace temperature setpoint of each section, etc. Process evaluation parameters include actual furnace temperature and gas consumption per unit area.

[0043] Next, the raw data is cleaned, including deleting outliers that clearly violate the process procedures (such as furnace temperature exceeding process requirements, heating time being negative, heating time significantly exceeding process requirements, etc.) and deleting furnace data with missing values, based on the process operation procedures for different steel grades.

[0044] Then, exemplary work teams were identified by analyzing operational data from different teams. To ensure accuracy and fairness, only cases where both the furnace feeding and unloading operations were performed by the same team were statistically analyzed, effectively eliminating interference from shift handover. For example... Figures 4 to 6 As shown, based on a comprehensive evaluation of three dimensions—heating efficiency (shortest average total heating time), temperature control accuracy (low deviation rate between furnace outlet temperature and target temperature, such as 54.92% for Class D), and overall operating condition qualification rate (such as 69% for Class D)—Class D can be determined to have the best overall performance in terms of efficiency, accuracy, and stability. Therefore, the operating data of Class D was selected as the benchmark for constructing a library of excellent operating modes.

[0045] Finally, the historical data of the best operating team (team D in this example) was traversed after cleaning, and the comprehensive score J was calculated for each furnace cycle. A threshold of J ≥ 0.85 was set, which was then judged as "excellent operating condition", and all data with J < 0.85 were deleted. In the end, a total of 34,504 sets of excellent operating parameters were selected in the example, forming the initial set of excellent operating modes.

[0046] If all the above verifications pass, the operation model in the temporary database will be added to the excellent database. If the verifications fail, the model will be retrained and optimized.

[0047] In some embodiments, the input parameters of the furnace steel temperature prediction model are slab state parameters, and the output parameter is the furnace temperature; the input parameters of the unit consumption prediction model are heating furnace operating parameters, and the output parameter is gas consumption per unit. When training both models, the input parameters are first normalized using a max-min normalization method, scaling all feature parameters to the [0, 1] interval to eliminate the influence of dimensions and accelerate model training. The max-min normalization formula is as follows:

[0048] in, These are the original feature values, that is, the original values ​​of a certain feature in the dataset; The minimum value of this feature in the dataset is used to determine the lower bound of the scaling range; The maximum value of this feature in the dataset is used to determine the upper bound of the scaling range; These are the normalized eigenvalues.

[0049] The processed data is then randomly divided into training, testing, and validation datasets in a 7:2:1 ratio. The model is then optimized using the SGD algorithm to improve training and validation performance. 2 (Formula 5) are all greater than 0.9.

[0050]

[0051] in,n The number of samples; The true value of the sample; The predicted value for the sample; This is the average of the true values.

[0052] In some embodiments, the method further includes: Determine the quantity of excellent historical production data for the same steel grade; If the number of excellent historical production data for the same steel grade does not exceed a threshold, each excellent historical production data point will be treated as a sample. If the number of excellent historical production data for the same steel grade exceeds a threshold, multiple cluster centers are obtained by using the mean clustering algorithm. The other excellent historical production data are then assigned to the corresponding cluster centers as subsets, and each cluster center is treated as a sample.

[0053] Specifically, to facilitate efficient retrieval of the superior operation patterns, a structured approach can be adopted. First, all superior operation patterns are initially categorized according to steel type to ensure clear process relevance for subsequent analysis. Based on this, the amount of operation pattern data under each steel type is evaluated. If the number of operation patterns under a certain steel type is no more than 10, they are all directly included in the superior pattern library for that steel type as typical operation examples. If the number of data exceeds 10, a fuzzy C-means clustering algorithm is used for in-depth analysis. Through system evaluation, the number of clusters is set within the range of 2 to 8, automatically identifying representative cluster centers. Each operation pattern is then assigned as a subset to its corresponding cluster center, forming a structured pattern set centered on typical operation characteristics. Through this hierarchical classification and clustering approach, a clearly structured and highly representative "superior operation pattern library" is ultimately constructed.

[0054] In some embodiments, calculating the similarity between the state parameters of the billet and the state parameters of each sample, and using the operating parameters corresponding to the samples with a similarity not less than a threshold as the actual operating parameters of the heating furnace, further includes: If the sample is a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center is not less than a threshold, then the operating parameters corresponding to the cluster center are directly used as the actual operating parameters of the heating furnace. In response to the sample being a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center being less than a threshold, the similarity between each excellent historical production data in the subset and the state parameters of the billet is calculated, and the operating parameters corresponding to the excellent historical production data in the subset with a similarity not less than the threshold are used as the actual operating parameters of the heating furnace.

[0055] Specifically, when calculating similarity, if the sample located based on the steel grade is the cluster center, then the Euclidean distance between the current billet's state parameters and the cluster center is calculated to obtain the similarity:

[0056] in, , These represent the input state parameters and the cluster center state parameters, respectively.

[0057] If the similarity of cluster centers is not less than the threshold, the operation parameters corresponding to that center are directly recommended for output. If there are no cluster centers with similarity not less than the threshold, the cluster center with the highest similarity is selected, and the similarity of each subset with the input state parameters is further compared. If the similarity of a subset is not less than the threshold, the operation parameters corresponding to that subset are recommended for output.

[0058] In some embodiments, such as Figure 7 The heating furnace control and communication architecture shown is characterized by a three-tiered division of labor: real-time on-site control, edge intelligent preprocessing, and cloud-based optimization decision-making. This ensures high data security while maintaining high scalability and low latency.

[0059] Based on the same inventive concept, according to another aspect of the present invention, embodiments of the present invention also provide a control system 400 for a heating furnace, such as... Figure 8 As shown, it includes: The acquisition module 401 is configured to acquire the current billet status parameters, wherein the status parameters include steel type, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness. The matching module 402 is configured to match multiple samples of the corresponding steel type in a pre-established library of superior operating modes based on the steel type in the state parameters, wherein each sample includes the state parameters and operating parameters of the sample. The calculation module 403 is configured to calculate the similarity between the state parameters of the billet and the state parameters of each sample, and to use the operating parameters corresponding to the samples with similarity not less than a threshold as the actual operating parameters of the heating furnace.

[0060] In some embodiments, a decision module is also included, configured as follows: In response to the absence of samples with a similarity greater than the threshold, the state parameters of the billet are input into the furnace steel temperature prediction model to predict the furnace temperature. The particle swarm optimization algorithm is used to find multiple predicted operating parameters in the operating parameter space, and each predicted operating parameter is input into the unit consumption prediction model to obtain multiple predicted values ​​of gas unit consumption. The comprehensive operating condition evaluation index is obtained by calculating each predicted gas consumption value and the furnace outlet temperature. The operating parameter corresponding to the largest comprehensive operating condition evaluation index shall be used as the actual operating parameter of the heating furnace.

[0061] In some embodiments, calculating a comprehensive operating condition evaluation index by combining each predicted gas consumption value with the furnace outlet temperature further includes: The first score is calculated based on the following formula:

[0062] Wherein, y1 is the furnace exit temperature, and y2 is the target furnace exit temperature; The second score is calculated based on the following formula:

[0063] in, i 1. Predicted unit gas consumption; based on The comprehensive working condition evaluation index was obtained.

[0064] In some embodiments, a backtesting module is also included, configured as follows: The newly optimized and superior operation mode data obtained through optimization will be temporarily stored in the temporary data storage database. The furnace steel temperature prediction model and the unit consumption prediction model are backtested and verified at preset time intervals. If the backtest verification passes, the operation model in the temporary database is added to the excellent database.

[0065] In some embodiments, a building module is further configured to build a library of superior operating modes, wherein the step of building the library of superior operating modes includes: Collect historical production data, each of which includes slab condition parameters, furnace operating parameters, and process evaluation parameters; The heating efficiency of each historical production data is determined based on the furnace operating parameters; the temperature control accuracy of each historical production data is determined based on the slab state parameters and process evaluation parameters; and the comprehensive operating condition evaluation index is determined based on the process evaluation parameters. Based on the heating efficiency, temperature control accuracy, and comprehensive operating condition evaluation indicators, excellent historical production data are selected, and a library of excellent operating modes is constructed based on the excellent historical production data.

[0066] In some embodiments, it also includes: Determine the quantity of excellent historical production data for the same steel grade; If the number of excellent historical production data for the same steel grade does not exceed a threshold, each excellent historical production data point will be treated as a sample. If the number of excellent historical production data for the same steel grade exceeds a threshold, multiple cluster centers are obtained by using the mean clustering algorithm. The other excellent historical production data are then assigned to the corresponding cluster centers as subsets, and each cluster center is treated as a sample.

[0067] In some embodiments, calculating the similarity between the state parameters of the billet and the state parameters of each sample, and using the operating parameters corresponding to the samples with a similarity not less than a threshold as the actual operating parameters of the heating furnace, further includes: If the sample is a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center is not less than a threshold, then the operating parameters corresponding to the cluster center are directly used as the actual operating parameters of the heating furnace. In response to the sample being a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center being less than a threshold, the similarity between each excellent historical production data in the subset and the state parameters of the billet is calculated, and the operating parameters corresponding to the excellent historical production data in the subset with a similarity not less than the threshold are used as the actual operating parameters of the heating furnace.

[0068] Based on the same inventive concept, according to another aspect of the present invention, such as Figure 9 As shown, an embodiment of the present invention also provides a computer device 501, comprising: At least one processor 520; and The memory 510 stores a computer program 511 that can run on a processor. When the processor 520 executes the program, it performs the steps of any of the above-described furnace control methods.

[0069] Based on the same inventive concept, according to another aspect of the present invention, such as Figure 10 As shown, embodiments of the present invention also provide a computer-readable storage medium 601, which stores a computer program 610. When the computer program 610 is executed by a processor, it performs the steps of any of the above-described furnace control methods.

[0070] Finally, it should be noted that those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods.

[0071] Furthermore, it should be understood that the computer-readable storage medium (e.g., memory) described herein may be volatile memory or non-volatile memory, or may include both volatile memory and non-volatile memory.

[0072] Those skilled in the art will also understand that the various exemplary logic blocks, modules, circuits, and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the functionality of various illustrative components, modules, circuits, and steps has been generally described. Whether this functionality is implemented as software or as hardware depends on the specific application and the design constraints imposed on the system as a whole. Those skilled in the art can implement the functionality in various ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the embodiments disclosed herein.

[0073] The above are exemplary embodiments disclosed in this invention. However, it should be noted that various changes and modifications can be made without departing from the scope of the embodiments of this invention as defined by the claims. The functions, steps, and / or actions of the methods according to the disclosed embodiments described herein do not need to be performed in any particular order. Furthermore, although the elements disclosed in the embodiments of this invention may be described or claimed individually, they may be understood as multiple unless explicitly limited to a singular number.

[0074] It should be understood that, as used herein, the singular form “a” is intended to include the plural form as well, unless the context clearly supports an exception. It should also be understood that, as used herein, “and / or” refers to any and all possible combinations of one or more of the associated listed items.

[0075] The embodiment numbers disclosed in the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0076] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0077] Those skilled in the art should understand that the discussion of any of the above embodiments is merely exemplary and is not intended to imply that the scope of the invention (including the claims) is limited to these examples. Within the framework of the invention, technical features of the above embodiments or different embodiments can be combined, and many other variations of different aspects of the invention exist, which are not provided in the details for the sake of brevity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. A method for controlling a heating furnace, the heating furnace being used to heat steel billets, characterized in that, Includes the following steps: Obtain the current state parameters of the steel billet, wherein the state parameters include steel grade, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness; Based on the steel grade in the state parameters, multiple samples of the corresponding steel grade are matched in a pre-established library of superior operating modes, wherein each sample includes the sample's state parameters and operating parameters. Calculate the similarity between the state parameters of the billet and the state parameters of each sample, and take the operating parameters corresponding to the samples with similarity not less than a threshold as the actual operating parameters of the heating furnace.

2. The method as described in claim 1, characterized in that, Also includes: In response to the absence of samples with a similarity greater than the threshold, the state parameters of the billet are input into the furnace steel temperature prediction model to predict the furnace temperature. The particle swarm optimization algorithm is used to find multiple predicted operating parameters in the operating parameter space, and each predicted operating parameter is input into the unit consumption prediction model to obtain multiple predicted values ​​of gas unit consumption. The comprehensive operating condition evaluation index is obtained by calculating each predicted gas consumption value and the furnace outlet temperature. The operating parameter corresponding to the largest comprehensive operating condition evaluation index shall be used as the actual operating parameter of the heating furnace.

3. The method as described in claim 2, characterized in that, The comprehensive operating condition evaluation index is calculated by combining each predicted gas consumption value with the furnace outlet temperature, and further includes: The first score is calculated based on the following formula: Wherein, y1 is the furnace exit temperature, and y2 is the target furnace exit temperature; The second score is calculated based on the following formula: in, i 1. Predicted unit gas consumption; based on The comprehensive working condition evaluation index was obtained.

4. The method as described in claim 2, characterized in that, Also includes: The newly optimized and superior operation mode data obtained through optimization will be temporarily stored in the temporary data storage database. The furnace steel temperature prediction model and the unit consumption prediction model are backtested and verified at preset time intervals. If the backtest verification passes, the operation model in the temporary database is added to the excellent database.

5. The method as described in claim 1, characterized in that, It also includes building a library of superior operating modes, wherein the steps for building the library of superior operating modes include: Collect historical production data, each of which includes slab condition parameters, furnace operating parameters, and process evaluation parameters; The heating efficiency of each historical production data is determined based on the furnace operating parameters; the temperature control accuracy of each historical production data is determined based on the slab state parameters and process evaluation parameters; and the comprehensive operating condition evaluation index is determined based on the process evaluation parameters. Based on the heating efficiency, temperature control accuracy, and comprehensive operating condition evaluation indicators, excellent historical production data are selected, and a library of excellent operating modes is constructed based on the excellent historical production data.

6. The method as described in claim 5, characterized in that, Also includes: Determine the quantity of excellent historical production data for the same steel grade; If the number of excellent historical production data for the same steel grade does not exceed a threshold, each excellent historical production data point will be treated as a sample. If the number of excellent historical production data for the same steel grade exceeds a threshold, multiple cluster centers are obtained by using the mean clustering algorithm. The other excellent historical production data are then assigned to the corresponding cluster centers as subsets, and each cluster center is treated as a sample.

7. The method as described in claim 6, characterized in that, Calculating the similarity between the state parameters of the steel billet and the state parameters of each sample, and using the operating parameters corresponding to the samples with similarity not less than a threshold as the actual operating parameters of the heating furnace, further includes: If the sample is a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center is not less than a threshold, then the operating parameters corresponding to the cluster center are directly used as the actual operating parameters of the heating furnace. In response to the sample being a cluster center and the similarity between the state parameters of the billet and the state parameters of the cluster center being less than a threshold, the similarity between each excellent historical production data in the subset and the state parameters of the billet is calculated, and the operating parameters corresponding to the excellent historical production data in the subset with a similarity not less than the threshold are used as the actual operating parameters of the heating furnace.

8. A control system for a heating furnace, characterized in that, include The acquisition module is configured to acquire the current billet status parameters, wherein the status parameters include steel grade, width, length, weight, furnace inlet temperature, target furnace outlet temperature, and finished product thickness. The matching module is configured to match multiple samples of the corresponding steel type in a pre-established library of superior operating modes based on the steel type in the state parameters, wherein each sample includes the sample's state parameters and operating parameters. The calculation module is configured to calculate the similarity between the state parameters of the billet and the state parameters of each sample, and to use the operating parameters corresponding to the samples with similarity not less than a threshold as the actual operating parameters of the heating furnace.

9. A computer device, comprising: At least one processor; as well as A memory storing a computer program executable on the processor, characterized in that the processor executes the program by performing the steps of the method as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it performs the steps of the method as described in any one of claims 1-7.

Images