A water tank parameter prediction method and device, electronic equipment and storage medium

Abstract

This invention provides a method, apparatus, electronic device, and storage medium for predicting water tank parameters. The method acquires current condition data of the steel to be cooled, as well as historical condition data from past cooling processes. It calculates the similarity between the current and historical condition data, filters the historical condition data based on a comparison of the similarity with a preset similarity threshold, and predicts the water tank parameters under the current condition data by combining the filtering results with the current condition data. The steel to be cooled is then cooled according to the water tank parameters under the current condition data. This invention improves the accuracy and efficiency of predicting water tank parameters under current condition data by filtering historical condition data and combining the filtering results with the current condition data.

Description

Technical Field

[0001] This invention relates to the field of steel rolling technology, and in particular to a method, apparatus, electronic device and storage medium for predicting water tank parameters. Background Technology

[0002] The controlled cooling process of steel is a crucial step that affects its final properties. During cooling, precise control of the water volume and pressure in the tank is necessary to ensure the steel cools to the target temperature.

[0003] Currently, in the controlled cooling process of steel, the water volume and pressure in the water tank are usually predicted using empirical formulas or complex physical simulation methods. However, these methods rely solely on the current conditions of the steel to be cooled for calculation or simulation, resulting in low prediction accuracy and efficiency. Therefore, it is necessary to improve the current methods for predicting water tank parameters. Summary of the Invention

[0004] In view of the shortcomings of the prior art described above, this application provides a method, apparatus, electronic device and storage medium for predicting water tank parameters to solve the above technical problems.

[0005] According to one aspect of the embodiments of this application, a method for predicting water tank parameters is provided, comprising: acquiring current condition data of steel to be cooled, and historical condition data of historical cooling processes; calculating the similarity between the current condition data and the historical condition data, and filtering the historical condition data based on the comparison result of the similarity with a preset similarity threshold; combining the filtering result and the current condition data to predict the water tank parameters under the current condition data, and cooling the steel to be cooled according to the water tank parameters under the current condition data.

[0006] In one embodiment of this application, if there are multiple similarities, the process of filtering the historical condition data based on the comparison results of the similarities with a preset similarity threshold includes: selecting a similarity as a target similarity; if the target similarity is greater than the preset similarity threshold, then using the historical condition data of the calculated target similarity as target condition data; if the target similarity is less than or equal to the preset similarity threshold, then continuing to select another similarity as a target similarity and comparing it with the preset similarity threshold, until all the similarities are compared with the preset similarity threshold to obtain all target condition data; and using all target condition data as the filtering result.

[0007] In one embodiment of this application, the process of predicting the water tank parameters under the current condition data by combining the screening results and the current condition data includes: obtaining the water tank parameters under the target condition data; inputting the target condition data, the water tank parameters under the target condition data, and the current condition data into a water tank parameter prediction model to obtain the water tank parameters under the current condition data; and training the water tank parameter prediction model based on the historical condition data and the water tank parameters under the historical condition data to obtain a preset water tank parameter prediction model.

[0008] In one embodiment of this application, if the preset water tank parameter prediction model includes a preset steel feature construction model and a preset water tank parameter regression model, then the process of training the preset water tank parameter prediction model based on the historical condition data and the water tank parameters under the historical condition data to obtain the water tank parameter prediction model includes: selecting a first preset number of data points from the historical condition data as sample condition data, and selecting a second preset number of data points from the historical condition data as test condition data, wherein the test condition data is different from the sample condition data; and training the preset steel feature construction model and the preset water tank parameter regression model using the sample condition data. The trained feature construction model and the trained parametric regression model are obtained; the test condition data is input into the trained feature construction model to obtain the test vector of the steel to be cooled; the test vector is input into the trained parametric regression model to obtain the test result; the error value between the test result and the water tank parameters under the test condition data is calculated; based on the comparison result of the error value and the preset error threshold, it is determined whether to stop training the preset steel feature construction model and the preset water tank parameter regression model, and the feature construction model obtained when training stops is combined with the parametric regression model obtained when training stops to obtain the water tank parameter prediction model.

[0009] In one embodiment of this application, the process of determining whether to stop training the preset steel feature construction model and the preset water tank parameter regression model based on the comparison result of the error value and the preset error threshold includes: if the error value is less than the preset error threshold, then stop training the preset steel feature construction model and the preset water tank parameter regression model; if the error value is greater than or equal to the preset error threshold, then continue training the preset steel feature construction model and the preset water tank parameter regression model until the error value between the test result obtained after continued training and the water tank parameters under the test conditions is less than the preset error threshold, then stop training the preset steel feature construction model and the preset water tank parameter regression model.

[0010] In one embodiment of this application, if the tank parameter prediction model includes a steel feature construction model and a tank parameter regression model, the process of inputting the target condition data, the tank parameters under the target condition data, and the current condition data into the tank parameter prediction model to obtain the tank parameters under the current condition data includes: inputting the target condition data, the tank parameters under the target condition data, and the current condition data into the steel feature construction model to obtain the feature representation vector of the steel to be cooled; and inputting the feature representation vector into the tank parameter regression model to obtain the tank parameters under the current condition data.

[0011] In one embodiment of this application, if the steel feature construction model includes a preset feature representation vector construction model, the process of inputting the target condition data, the water tank parameters under the target condition data, and the current condition data into the steel feature construction model to obtain the feature representation vector of the steel to be cooled includes: representing the target condition data in vector form to obtain a target condition vector; representing the current condition data in vector form to obtain a current condition vector; representing the water tank parameters under the target condition data in vector form to obtain a water tank parameter vector; calculating the difference between the target condition vector and the current condition vector to obtain a difference vector; combining the difference vector with the water tank parameter vector to obtain a combined vector; concatenating the combined vector with the current condition vector to obtain a concatenated vector; and inputting the concatenated vector into the preset feature representation vector construction model to obtain the feature representation vector.

[0012] According to one aspect of the embodiments of this application, a water tank parameter prediction device is provided, comprising: a data acquisition module, configured to acquire current condition data of the steel to be cooled, and historical condition data of a historical cooling process; a data filtering module, configured to calculate the similarity between the current condition data and the historical condition data, and filter the historical condition data based on the comparison result of the similarity with a preset similarity threshold; and a parameter prediction module, configured to combine the filtering result and the current condition data to predict the water tank parameters under the current condition data, and cool the steel to be cooled according to the water tank parameters under the current condition data.

[0013] According to one aspect of the embodiments of this application, an electronic device is provided, the electronic device comprising: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device enables the water tank parameter prediction method as described above.

[0014] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a computer processor, causes the computer to perform the water tank parameter prediction method described above.

[0015] The beneficial effects of this invention are as follows: This invention obtains the current condition data of the steel to be cooled, as well as the historical condition data of the historical cooling process, calculates the similarity between the current condition data and the historical condition data, and filters the historical condition data based on the comparison results of the similarity with a preset similarity threshold. Combining the filtering results with the current condition data, the water tank parameters under the current condition data are predicted, and the steel to be cooled is cooled according to the water tank parameters under the current condition data. The above process improves the prediction accuracy and efficiency of the water tank parameters under the current condition data by filtering the historical condition data and combining the filtering results with the current condition data.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. In the drawings:

[0018] Figure 1 This is a schematic diagram illustrating an exemplary system architecture as shown in an exemplary embodiment of this application;

[0019] Figure 2 This is a flowchart illustrating a water tank parameter prediction method in an exemplary embodiment of this application;

[0020] Figure 3 This is a schematic diagram illustrating a splicing vector in an exemplary embodiment of this application;

[0021] Figure 4 This is a flowchart illustrating the processing of spliced ​​vectors, as shown in an exemplary embodiment of this application;

[0022] Figure 5 A block diagram is shown that is suitable for implementing a water tank parameter prediction device according to embodiments of this application;

[0023] Figure 6 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown. Detailed Implementation

[0024] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0025] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0026] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily need to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0027] In this application, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0028] The technical solutions of this application involve technologies related to welding of steel structural components, which are specifically illustrated through the following embodiments:

[0029] Figure 1 This is a schematic diagram illustrating an exemplary system architecture as shown in an exemplary embodiment of this application.

[0030] Reference Figure 1As shown, the system architecture may include a data storage device 101 and a computer device 102. The computer device 102 may be at least one of a desktop graphics processing unit (GPU) computer, a GPU computing cluster, or a neural network computer. Those skilled in the art can use the computer device 102 to acquire current condition data of the steel to be cooled, as well as historical condition data from previous cooling processes. They can calculate the similarity between the current and historical condition data, and based on a comparison of the similarity with a preset similarity threshold, filter the historical condition data. Combining the filtering results with the current condition data, they can predict the water tank parameters under the current condition data and cool the steel according to these parameters. The data storage device 101 stores the current condition data of the steel to be cooled and the historical condition data from previous cooling processes. In this embodiment, the data storage device 101 uses a read-only memory (ROM) or random access memory (RAM) to store the current condition data of the steel to be cooled and the historical condition data from previous cooling processes, and provides this data to the computer device 102 for processing.

[0031] Indicatively, after acquiring the current condition data of the steel to be cooled from the data storage device 101, as well as the historical condition data of the historical cooling process, the computer device 102 calculates the similarity between the current condition data and the historical condition data. Based on the comparison result of the similarity with a preset similarity threshold, the historical condition data is filtered. Combining the filtering result with the current condition data, the water tank parameters under the current condition data are predicted, and the steel to be cooled is cooled according to the water tank parameters under the current condition data. The above process, by filtering the historical condition data and combining the filtering result with the current condition data to predict the water tank parameters under the current condition data, improves the prediction accuracy and prediction efficiency of the water tank parameters under the current condition data.

[0032] It should be noted that the water tank parameter prediction method provided in this application embodiment is generally executed by computer device 102, and correspondingly, the water tank parameter prediction device is generally installed in computer device 102.

[0033] The implementation details of the technical solutions in the embodiments of this application are described in detail below:

[0034] Figure 2 This is a flowchart illustrating a water tank parameter prediction method in an exemplary embodiment of this application. The water tank parameter prediction method can be executed by a computational processing device, which may be... Figure 1 The computer device 102 shown is illustrated. (Refer to...) Figure 2As shown, the method for predicting water tank parameters includes at least steps S210 to S230, which are described in detail below:

[0035] In step S210, the current condition data of the steel to be cooled and the historical condition data of the historical cooling process are obtained.

[0036] In one embodiment of this application, current condition data includes chemical composition, dimensions, current temperature, target temperature after cooling, etc. Historical condition data is a collection of condition data of the steel before cooling during historical cooling processes.

[0037] In step S220, the similarity between the current condition data and the historical condition data is calculated, and the historical condition data is filtered based on the comparison result between the similarity and the preset similarity threshold.

[0038] In this embodiment, before calculating the similarity, it is necessary to normalize the historical condition data, remove outliers, and normalize the current condition data. The calculation method for normalizing the historical condition data is the same as that for normalizing the current condition data.

[0039] In this embodiment, the similarity measurement method between current conditional data and historical conditional data can be Euclidean distance, cosine similarity, etc., and no limitation is made here.

[0040] In this embodiment, the preset similarity threshold can be set according to the actual situation, and no specific limitation is made here.

[0041] In step S230, the water tank parameters under the current conditions are predicted by combining the screening results and the current condition data, and the steel to be cooled is cooled according to the water tank parameters under the current conditions data.

[0042] In this embodiment, by acquiring the current condition data of the steel to be cooled and the historical condition data of the historical cooling process, the similarity between the current condition data and the historical condition data is calculated. Based on the comparison result of the similarity with the preset similarity threshold, the historical condition data is filtered. Combining the filtering result with the current condition data, the water tank parameters under the current condition data are predicted, and the steel to be cooled is cooled according to the water tank parameters under the current condition data. The above process improves the prediction accuracy and efficiency of the water tank parameters under the current condition data by filtering the historical condition data and combining the filtering result with the current condition data to predict the water tank parameters under the current condition data.

[0043] In one embodiment of this application, if there are multiple similarities, the process of filtering historical conditional data based on the comparison results of the similarities and preset similarity thresholds includes:

[0044] Select a similarity as the target similarity. If the target similarity is greater than the preset similarity threshold, the historical conditional data of the calculated target similarity will be used as the target conditional data.

[0045] In this embodiment, when the target similarity is greater than a preset similarity threshold, the historical condition data corresponding to the target similarity is used as the target condition data, so as to filter out the historical condition data with high similarity to the current condition data and improve the accuracy of predicting the water tank parameters under the current condition data.

[0046] If the target similarity is less than or equal to the preset similarity threshold, another similarity is selected as the target similarity and compared with the preset similarity threshold until all similarities are compared with the preset similarity threshold, and all target condition data are obtained.

[0047] In this embodiment, if the similarity of the next selection is less than or equal to a preset similarity threshold, the historical condition data corresponding to that similarity will not be selected, so as to avoid introducing noise into the filtering results and reducing the accuracy of predicting the tank parameters under the current condition data.

[0048] Use all data related to the target criteria as the filtering results.

[0049] In this embodiment, the target condition data can be stored in the form of an array or a matrix, etc., and there is no limitation on this.

[0050] In this embodiment, the process of filtering historical condition data can be carried out by introducing a similarity searcher with a fixed similarity threshold. The similarity searcher dynamically selects the historical condition data that is closest to the current condition data, which overcomes the defect of uneven coverage of similar condition data in related technologies. This ensures that the water tank parameter prediction model can make full use of the effective information in the historical condition data, while avoiding the introduction of invalid information in the historical condition data to interfere with the prediction results.

[0051] In one embodiment of this application, the process of predicting the water tank parameters under the current conditions, by combining the screening results and the current condition data, includes:

[0052] Obtain the water tank parameters under the target conditions.

[0053] In this embodiment, the water tank parameters under the target conditions are the water tank parameters used in the historical cooling process, including water volume and water pressure.

[0054] The target condition data, the water tank parameters under the target condition data, and the current condition data are all input into the water tank parameter prediction model to obtain the water tank parameters under the current condition data.

[0055] In this embodiment, the water tank parameter prediction model is obtained by training a preset water tank parameter prediction model based on historical condition data and water tank parameters under historical condition data.

[0056] In this embodiment, the water tank parameters under the current conditions are predicted by the water tank parameter prediction model. This not only improves the efficiency of predicting the water tank parameters under the current conditions, but also improves the accuracy of predicting the water tank parameters under the current conditions compared with the empirical formula calculation method or complex physical simulation method.

[0057] In one embodiment of this application, if the preset water tank parameter prediction model includes a preset steel feature construction model and a preset water tank parameter regression model, then the process of training the preset water tank parameter prediction model based on historical condition data and water tank parameters under historical condition data to obtain the water tank parameter prediction model includes:

[0058] Select a first preset number of data points from the historical condition data as sample condition data, and select a second preset number of data points from the historical condition data as test condition data.

[0059] In this embodiment, the test condition data is different from the sample condition data. The first preset quantity and the second preset quantity can be determined according to the actual situation, and no specific limitation is made here.

[0060] The preset steel feature construction model and preset water tank parameter regression model are trained using sample conditional data to obtain the trained feature construction model and trained parameter regression model.

[0061] In this embodiment, training the preset steel feature construction model and the preset water tank parameter regression model is the process of adjusting the parameters in the preset steel feature construction model and the preset water tank parameter regression model, thereby optimizing the preset steel feature construction model and the preset water tank parameter regression model.

[0062] The test condition data is input into the trained feature construction model to obtain the test vector of the steel to be cooled; and the test vector is input into the trained parametric regression model to obtain the test results.

[0063] In this embodiment, the trained feature construction model and the trained parameter regression model are tested using test condition data, thereby verifying the training effect of the preset steel feature construction model and the preset water tank parameter regression model through the test results.

[0064] Calculate the error between the test results and the tank parameters under the test conditions.

[0065] In this embodiment, the larger the error value between the test result and the water tank parameters under the test conditions, the greater the gap between the test effect and the water tank parameters under the test conditions. Conversely, the smaller the error value between the test result and the water tank parameters under the test conditions, the smaller the gap between the test effect and the water tank parameters under the test conditions. This indicates a better training effect on the preset steel feature construction model and the preset water tank parameter regression model.

[0066] In this embodiment, the error between the test result and the water tank parameters under the test conditions can be calculated using a loss function, the expression of which is shown below:

[0067]

[0068] Where loss represents the error between the test result and the tank parameters under the test conditions, y represents the test result, and y represents the tank parameters under the test conditions.

[0069] Based on the comparison results between the error value and the preset error threshold, it is determined whether to stop training the preset steel feature construction model and the preset water tank parameter regression model. The feature construction model obtained when training stops is combined with the parameter regression model obtained when training stops to obtain the water tank parameter prediction model.

[0070] In this embodiment, if the error value is less than a preset error threshold, training of the preset steel feature construction model and the preset water tank parameter regression model is stopped; if the error value is greater than or equal to the preset error threshold, training of the preset steel feature construction model and the preset water tank parameter regression model continues until the error value between the test result obtained after continued training and the water tank parameters under the test conditions is less than the preset error threshold, at which point training of the preset steel feature construction model and the preset water tank parameter regression model is stopped.

[0071] In one embodiment of this application, the process of determining whether to stop training the preset steel feature construction model and the preset water tank parameter regression model based on the comparison result of the error value and the preset error threshold includes:

[0072] If the error value is less than the preset error threshold, then training of the preset steel feature construction model and the preset water tank parameter regression model will be stopped.

[0073] In this embodiment, the preset error threshold is set according to the actual situation, and no specific limitation is made here.

[0074] If the error value is greater than or equal to the preset error threshold, the training of the preset steel feature construction model and the preset water tank parameter regression model will continue until the error value between the test result obtained after training and the water tank parameters under the test conditions is less than the preset error threshold, at which point the training of the preset steel feature construction model and the preset water tank parameter regression model will stop.

[0075] In this embodiment, when the error value is greater than or equal to the preset error threshold, the preset steel feature construction model and the preset water tank parameter regression model are iteratively trained to continuously optimize the parameters of the preset steel feature construction model and the preset water tank parameter regression model until the error value between the test result obtained after further training and the water tank parameters under the test conditions meets the error requirements, thereby enabling the preset steel feature construction model and the preset water tank parameter regression model to achieve the expected training effect.

[0076] In one embodiment of this application, if the water tank parameter prediction model includes a steel feature construction model and a water tank parameter regression model, then the process of inputting the target condition data, the water tank parameters under the target condition data, and the current condition data into the water tank parameter prediction model to obtain the water tank parameters under the current condition data includes:

[0077] The target condition data, the water tank parameters under the target condition data, and the current condition data are input together into the steel feature model to obtain the feature representation vector of the steel to be cooled.

[0078] In this embodiment, the steel feature construction model is used to construct a feature representation vector of the steel to be cooled based on the target condition data, the water tank parameters under the target condition data, and the current condition data.

[0079] In this embodiment, the steel feature construction model can process different amounts of target condition data, overcoming the limitations of fully connected neural networks and convolutional neural networks, which can only process a fixed amount of input information.

[0080] Input the feature representation vector into the water tank parameter regression model to obtain the water tank parameters under the current conditions.

[0081] In this embodiment, the water tank parameter regression model can be a Convolutional Neural Network (CNN) model, a Support Vector Machine (SVM) model, a fully connected neural network model, etc.

[0082] In one embodiment of this application, if the steel feature construction model includes a preset feature representation vector construction model, then the process of inputting the target condition data, the water tank parameters under the target condition data, and the current condition data into the steel feature construction model to obtain the feature representation vector of the steel to be cooled includes:

[0083] The target condition data is represented in vector form to obtain the target condition vector. The current condition data is represented in vector form to obtain the current condition vector. The water tank parameters under the target condition data are represented in vector form to obtain the water tank parameter vector.

[0084] In this embodiment, forming a target condition vector facilitates the representation of target condition data features, forming a current condition vector facilitates the representation of current condition data features, and forming a water tank parameter vector facilitates the representation of water tank parameter features under target condition data.

[0085] Calculate the difference between the target condition vector and the current condition vector to obtain the difference vector.

[0086] In this embodiment, if there are multiple target condition vectors, there are also multiple difference vectors, and the number of difference vectors is the same as the number of target condition vectors.

[0087] The difference vector is combined with the water tank parameter vector to obtain a combined vector, and then the combined vector is concatenated with the current condition vector to obtain a concatenated vector.

[0088] In this embodiment, if there are multiple difference vectors, there are also multiple combination vectors. All combination vectors are concatenated with the current condition vector to obtain a concatenated vector.

[0089] The concatenated vector is input into the preset feature representation vector to construct the model, and the feature representation vector is obtained.

[0090] In this embodiment, the preset feature representation vector construction model can be a recurrent neural network model (e.g., a recurrent neural network (RNN) model, a long short-term memory network (LSTM) model), or a Transformer Encoder network architecture model. No specific limitation is made here.

[0091] Figure 3 This is a schematic diagram illustrating the splicing of vectors in an exemplary embodiment of this application, as shown below. Figure 3 As shown, the concatenated vector includes: k j There are target condition vectors, k j A vector of water tank parameters, kj The objective condition vector and k j There is a one-to-one correspondence between the water tank parameter vectors and the current condition vector. Since there is no corresponding water tank parameter for the current condition data, the zero vector is used to replace the water tank parameter vector corresponding to the current condition vector. The target condition vector and the current condition vector have the same vector length d. x The water tank parameter vector corresponding to the target condition vector has the same vector length d as the water tank parameter vector corresponding to the current condition vector. y .

[0092] In this embodiment, the current condition vector is arranged in k. j Before each target condition vector, the tank parameter vector corresponding to the current condition vector is arranged before the tank parameter vector corresponding to the target condition vector.

[0093] In this embodiment, the expression for the concatenated vector is as follows:

[0094]

[0095] Among them, input j This represents the concatenation of vectors, x j This represents the current condition vector, where j represents the index of the current condition vector, and 0 represents the water tank parameter vector corresponding to the current condition vector. Indicates the kth j A target condition vector. X represents the historical knowledge base of the target condition vector. Indicates the kth j The water tank parameter vector corresponding to each target condition vector. Y represents the historical knowledge base of the water tank parameter vector.

[0096] In this embodiment, it can be seen from formula (2) that the concatenated vector has a feature dimension of (d) x +d y )*(k j +1), where the sequence length of the concatenated vector is (k j +1), and at each time step, the feature dimension is (d x +d y ).

[0097] Figure 4 This is a flowchart illustrating the processing of spliced ​​vectors as shown in an exemplary embodiment of this application, such as... Figure 4 As shown, the process of processing the spliced ​​vector includes: (1) inputting the spliced ​​vector into the preset feature representation vector to construct the model and obtain the feature representation vector; (2) inputting the feature representation vector into the water tank parameter regression model to obtain the water tank parameters under the current conditions.

[0098] In this embodiment, taking the TransformerEncoder network architecture model as an example, the expression of the feature representation vector is as follows:

[0099] s j =TransformerEncoder(input) j Equation (3)

[0100] Among them, s j The feature vector is represented by the input vector. j This represents the concatenation of vectors.

[0101] In this embodiment, the steel feature construction model can flexibly handle input sequences of different lengths (i.e., the total number of current condition data and target condition data). It not only integrates the tank parameters under the target condition data, but also considers the differences between the current condition data and the target condition data, as well as the influence of the current condition data with nonlinear relationships on the tank parameters. Therefore, by using the current condition data as the feature vector of the first time step, the robustness of the steel feature construction model in handling the current condition data with nonlinear relationships and in dealing with changes in the current condition data is improved.

[0102] In this embodiment, when the number of target condition data obtained after screening is small, by retaining and utilizing the current condition data of the steel to be cooled, and combining the feature construction mechanism set in the steel feature construction model and the parameter prediction mechanism set in the water tank parameter regression model, the uncertainty in the initial prediction stage of the water tank parameter prediction model is effectively alleviated, and the adaptability and accuracy of the water tank parameter prediction model are enhanced.

[0103] In this embodiment, incremental learning is performed using target condition data and water tank parameters under each target condition data. This not only improves the accuracy of the water tank parameter prediction model, but also effectively reduces the adverse effects of single condition data and abnormal water tank parameters under that condition data or noisy data in the target condition data on the prediction results, thus ensuring the stability and reliability of the prediction results.

[0104] This application integrates machine learning models, neural network models, similarity search, and model optimization techniques to construct a highly adaptive and accurate prediction system for water tank parameters, which significantly improves the automation level of the steel production process and the ability to control product quality.

[0105] The following describes an embodiment of the apparatus described in this application, which can be used to execute the water tank parameter prediction method described above in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the water tank parameter prediction method described above in this application.

[0106] Figure 5 This is a block diagram illustrating a water tank parameter prediction device according to an exemplary embodiment of this application. The device can be applied to... Figure 1 The implementation environment shown is specifically configured in computer device 102. This device can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applicable.

[0107] like Figure 5 As shown, the exemplary water tank parameter prediction device includes:

[0108] The data acquisition module 501 is used to acquire the current condition data of the steel to be cooled, as well as the historical condition data of the historical cooling process.

[0109] The data filtering module 502 is used to calculate the similarity between current condition data and historical condition data, and to filter historical condition data based on the comparison results of the similarity with the preset similarity threshold.

[0110] The parameter prediction module 503 is used to combine the screening results and the current condition data to predict the water tank parameters under the current condition data, and to cool the steel to be cooled according to the water tank parameters under the current condition data.

[0111] In one embodiment of this application, current condition data includes chemical composition, dimensions, current temperature, target temperature after cooling, etc. Historical condition data is a collection of condition data of the steel before cooling during historical cooling processes.

[0112] In this embodiment, the similarity measurement method between current conditional data and historical conditional data can be Euclidean distance, cosine similarity, etc., and no limitation is made here.

[0113] In this embodiment, the preset similarity threshold can be set according to the actual situation, and no specific limitation is made here.

[0114] In this embodiment, by acquiring the current condition data of the steel to be cooled and the historical condition data of the historical cooling process, the similarity between the current condition data and the historical condition data is calculated. Based on the comparison result of the similarity with the preset similarity threshold, the historical condition data is filtered. Combining the filtering result with the current condition data, the water tank parameters under the current condition data are predicted, and the steel to be cooled is cooled according to the water tank parameters under the current condition data. The above process improves the prediction accuracy and efficiency of the water tank parameters under the current condition data by filtering the historical condition data and combining the filtering result with the current condition data to predict the water tank parameters under the current condition data.

[0115] It should be noted that the water tank parameter prediction device and the water tank parameter prediction method provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the water tank parameter prediction device provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0116] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by one or more processors, cause the electronic device to implement the water tank parameter prediction method provided in the above embodiments.

[0117] Figure 6 A schematic diagram of a computer system suitable for implementing the embodiments of this application is shown. It should be noted that... Figure 6 The computer system 600 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0118] like Figure 6 As shown, the computer system 600 includes a Central Processing Unit (CPU) 601, which can perform various appropriate actions and processes, such as executing the methods described in the above embodiments, based on programs stored in Read-Only Memory (ROM) 602 or programs loaded from Storage Unit 608 into Random Access Memory (RAM) 603. The RAM 603 also stores various programs and data required for system operation. The CPU 601, ROM 602, and RAM 603 are interconnected via a bus 604. An Input / Output (I / O) interface 605 is also connected to the bus 604.

[0119] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0120] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611. When the computer program is executed by central processing unit (CPU) 601, it performs various functions defined in the system of this application.

[0121] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0122] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0123] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0124] Another aspect of this application provides a computer-readable storage medium storing computer-readable instructions that, when executed by a computer's processor, cause the computer to perform the water tank parameter prediction method provided in the various embodiments described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not assembled into the electronic device.

[0125] It should be noted that although several modules or units for the device used to perform actions have been mentioned in the detailed description above, this division is not mandatory. In fact, according to the embodiments of this application, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided and embodied by multiple modules or units.

[0126] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to the embodiments of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, touch terminal, or network device, etc.) to execute the method according to the embodiments of this application.

[0127] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0128] It should be understood that the above content is only a preferred exemplary embodiment of this application and is not intended to limit the implementation of this application. Those skilled in the art can easily make corresponding modifications or alterations based on the main concept and spirit of this application. Therefore, the scope of protection of this application should be determined by the scope of protection claimed in the claims.

Claims

1. A method for predicting water tank parameters, characterized in that, include: Obtain current condition data for the steel to be cooled, as well as historical condition data for the historical cooling process; Calculate the similarity between the current condition data and the historical condition data, and filter the historical condition data based on the comparison result between the similarity and a preset similarity threshold; Based on the screening results and the current condition data, the water tank parameters under the current condition data are predicted, and the steel to be cooled is cooled according to the water tank parameters under the current condition data. If there are multiple similarities, the process of filtering the historical conditional data based on the comparison results of the similarities and the preset similarity threshold includes: Select a similarity as the target similarity. If the target similarity is greater than the preset similarity threshold, then the historical conditional data of the target similarity will be calculated and used as the target conditional data. If the target similarity is less than or equal to the preset similarity threshold, then another similarity is selected as the target similarity and compared with the preset similarity threshold until all the similarities are compared with the preset similarity threshold, and all target condition data are obtained. All target condition data are used as the filtering results.

2. The method for predicting water tank parameters according to claim 1, characterized in that, The process of predicting the water tank parameters under the current conditions, based on the screening results and the current condition data, includes: Obtain the water tank parameters under the target conditions; The target condition data, the tank parameters under the target condition data, and the current condition data are input into the tank parameter prediction model to obtain the tank parameters under the current condition data. The tank parameter prediction model is obtained by training a preset tank parameter prediction model based on the historical condition data and the tank parameters under the historical condition data.

3. The method for predicting water tank parameters according to claim 2, characterized in that, If the preset water tank parameter prediction model includes a preset steel feature construction model and a preset water tank parameter regression model, then the process of training the preset water tank parameter prediction model based on the historical condition data and the water tank parameters under the historical condition data to obtain the water tank parameter prediction model includes: A first preset number of data points are selected from the historical condition data as sample condition data, and a second preset number of data points are selected from the historical condition data as test condition data. The test condition data is different from the sample condition data. The preset steel feature construction model and the preset water tank parameter regression model are trained using the sample condition data to obtain the trained feature construction model and the trained parameter regression model. The test condition data is input into the trained feature construction model to obtain the test vector of the steel to be cooled; and the test vector is input into the trained parametric regression model to obtain the test result. Calculate the error value between the test results and the water tank parameters under the test conditions; Based on the comparison result between the error value and the preset error threshold, it is determined whether to stop training the preset steel feature construction model and the preset water tank parameter regression model, and the feature construction model obtained when training stops and the parameter regression model obtained when training stops are combined to obtain the water tank parameter prediction model.

4. The water tank parameter prediction method according to claim 3, characterized in that, The process of determining whether to stop training the preset steel feature construction model and the preset water tank parameter regression model based on the comparison result of the error value and the preset error threshold includes: If the error value is less than the preset error threshold, then training of the preset steel feature construction model and the preset water tank parameter regression model shall be stopped. If the error value is greater than or equal to the preset error threshold, then the training of the preset steel feature construction model and the preset water tank parameter regression model continues until the error value between the test result obtained after training and the water tank parameters under the test conditions is less than the preset error threshold, and then the training of the preset steel feature construction model and the preset water tank parameter regression model stops.

5. The method for predicting water tank parameters according to claim 2, characterized in that, If the water tank parameter prediction model includes a steel feature construction model and a water tank parameter regression model, then the process of inputting the target condition data, the water tank parameters under the target condition data, and the current condition data into the water tank parameter prediction model to obtain the water tank parameters under the current condition data includes: The target condition data, the water tank parameters under the target condition data, and the current condition data are input together into the steel feature construction model to obtain the feature representation vector of the steel to be cooled. The feature representation vector is input into the water tank parameter regression model to obtain the water tank parameters under the current conditions.

6. The method for predicting water tank parameters according to claim 5, characterized in that, If the steel feature construction model includes a preset feature representation vector construction model, then the process of inputting the target condition data, the water tank parameters under the target condition data, and the current condition data into the steel feature construction model to obtain the feature representation vector of the steel to be cooled includes: The target condition data is represented in vector form to obtain the target condition vector; the current condition data is represented in vector form to obtain the current condition vector; and the water tank parameters under the target condition data are represented in vector form to obtain the water tank parameter vector. Calculate the difference between the target condition vector and the current condition vector to obtain the difference vector; The difference vector is combined with the water tank parameter vector to obtain a combined vector, and the combined vector is concatenated with the current condition vector to obtain a concatenated vector; The concatenated vector is input into the preset feature representation vector to construct the model, thereby obtaining the feature representation vector.

7. A water tank parameter prediction device, characterized in that, include: The data acquisition module is used to acquire the current condition data of the steel to be cooled, as well as the historical condition data of the historical cooling process. The data filtering module is used to calculate the similarity between the current condition data and the historical condition data, and to filter the historical condition data based on the comparison result of the similarity with a preset similarity threshold. If there are multiple similarities, the process of filtering the historical condition data based on the comparison results of the similarities with a preset similarity threshold includes: selecting a similarity as a target similarity; if the target similarity is greater than the preset similarity threshold, then the historical condition data with the calculated target similarity is used as the target condition data; if the target similarity is less than or equal to the preset similarity threshold, then another similarity is selected as the target similarity and compared with the preset similarity threshold, until all similarities are compared with the preset similarity threshold to obtain all target condition data; and all target condition data is used as the filtering result. The parameter prediction module is used to combine the screening results and the current condition data to predict the water tank parameters under the current condition data, and to cool the steel to be cooled according to the water tank parameters under the current condition data.

8. An electronic device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the water tank parameter prediction method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, It stores computer-readable instructions that, when executed by the processor of a computer, cause the computer to perform the water tank parameter prediction method according to any one of claims 1 to 6.

Images