Data processing method and apparatus, device, and storage medium

Abstract

The application discloses a data processing method, device and equipment and a storage medium. The method comprises the following steps: solving each sub-optimization model according to a first solving constraint condition respectively, and obtaining N residual quantity thresholds corresponding to N elements after mixed processing; traversing an i-th element in the N elements, determining M second solving constraint conditions based on a residual quantity threshold corresponding to the i-th element currently traversed and a current available quantity of the i-th element; solving a main optimization model according to the first solving constraint condition, a historical second solving constraint condition set and the M second solving constraint conditions corresponding to the i-th element to obtain M solving results; selecting target mixed information from mixed information corresponding to the M second solving constraint conditions corresponding to the i-th element; and outputting the target mixed information obtained in each traversal when the N elements are all traversed. The application converts a multi-target problem into multiple single-target problems, and solves multiple sets of satisfactory solutions, so that the mixed efficiency can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to data processing methods, data processing apparatus, data processing devices, and computer storage media. Background Technology

[0002] With intensifying market competition and the increasing size of various industrial equipment, the requirements for industrial batching are rising. Common business scenarios include sintering batching, cement raw material batching, and magnesium oxide powder batching. In each batching optimization field, multi-objective optimization challenges are frequently encountered, such as achieving the lowest average batching price while minimizing raw material inventory accumulation. Depending on the specific business scenario, there may be other desired objectives. Therefore, solving multi-objective optimization problems has become one of the hot research topics in the industrial production field. Summary of the Invention

[0003] This application provides a data processing method, a data processing apparatus, a data processing device, and a computer storage medium, which can decompose a multi-objective problem into multiple single-objective problems for solution and coordinate the processing of multi-objective problems.

[0004] On one hand, this application provides a data processing method executed by a data processing device. The data processing device is configured with a data optimization model for determining the mixed information of N elements. The data optimization model includes a primary optimization model, N secondary optimization models corresponding to the N elements, and a first solution constraint condition acting on the primary optimization model and the N secondary optimization models; N is an integer greater than 1. This data processing method includes:

[0005] Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element;

[0006] Traverse the i-th element among N elements, and determine M second solution constraints based on the remaining threshold corresponding to the i-th element and the current available quantity of the i-th element. Generate at least one mixed information from a second solution constraint corresponding to the i-th element, a set of historical second solution constraints, and the current available quantities of the Ni elements that have not been traversed; M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the set of historical second solution constraints includes the selected second solution constraints from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0007] The main optimization model is solved based on the first solution constraint, the historical set of second solution constraints, and the M second solution constraints corresponding to the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint corresponding to the i-th element, and each solution result includes: the amount of processing resources required to mix N elements according to the mixing information corresponding to the corresponding second solution constraint;

[0008] Based on M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added to the historical second solution constraint set as the selected second solution constraints.

[0009] When all N elements are traversed, output the target mixed information obtained in each traversal.

[0010] On one hand, this application provides a data processing apparatus, configured within a data processing device, which includes a data optimization model for determining mixed information of N elements. The data optimization model comprises a primary optimization model, N secondary optimization models corresponding to the N elements, and a first solution constraint acting on the primary optimization model and the N secondary optimization models; N is an integer greater than 1. The data processing apparatus includes:

[0011] The solving unit is used to solve each sub-optimization model according to the first solving constraint conditions to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one element corresponding to one residual threshold.

[0012] The determining unit is used to traverse the i-th element among the N elements, determine M second solution constraints based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element, and generate at least one mixing information from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities of the Ni elements that have not been traversed; M is an integer greater than 1, i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the second solution constraints selected from the M second solution constraints corresponding to each of the i-1 elements that have been traversed; the solving unit is further used to solve the main optimization model according to the first solution constraints, the historical set of second solution constraints, and the M second solution constraints corresponding to the i-th element to obtain M solution results; one solution result corresponds to a second solution constraint corresponding to the i-th element, and one solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraints;

[0013] The determining unit is further configured to select target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, and add the second solution constraint corresponding to the target mixed information as the selected second solution constraint to the historical second solution constraint set.

[0014] The output unit is used to output the target mixed information obtained in each traversal when all N elements have been traversed.

[0015] In one implementation, when the determining unit determines the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, it performs the following steps:

[0016] Based on the M solution results, determine the mixed information with the minimum processing resource requirement from the mixed information corresponding to the M second solution constraints of the i-th element; and determine the mixed information as the target mixed information.

[0017] In one implementation, when the determining unit determines the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, it performs the following steps:

[0018] Output the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and output the processing resource amount corresponding to each mixed information based on the M solution results; receive the first selection operation; determine the mixed information corresponding to the first selection operation as the target mixed information.

[0019] In one implementation, each of the M second solution constraints corresponding to the i-th element is represented by the remaining quantity interval of the i-th element when mixing the N elements. The remaining quantity interval includes the minimum remaining quantity and the maximum remaining quantity. The determining unit is further configured to use the remaining quantity threshold corresponding to the currently traversed i-th element as the minimum remaining quantity in the t-th remaining quantity interval; determine the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t; and add the minimum remaining quantity in the t-th remaining quantity interval and the remaining quantity increment of the t-th remaining quantity interval to obtain the maximum remaining quantity of the t-th remaining quantity interval. Here, t is an integer greater than or equal to 1 and less than M.

[0020] In one implementation, the solving unit is further configured to subtract the current available quantity of the i-th element from the remaining quantity threshold, and to multiply the M elements and t; and to divide the result of the subtraction operation with the result of the multiplication operation to obtain the remaining quantity increment of the t-th remaining quantity interval.

[0021] In one implementation, the determining unit is further configured to determine whether the i-th element is the first element to be traversed among the N elements; if so, the historical second solution constraint set is empty, and the solving unit is further configured to solve the main optimization model according to the first solution constraint and the M second solution constraints corresponding to the i-th element to obtain M solution results. If the determining unit determines that the i-th element is not the first element to be traversed among the N elements, the solving unit is further configured to solve the main optimization model according to the historical second solution constraint set, the first solution constraint, and the M second solution constraints corresponding to the i-th element to obtain M solution results.

[0022] In one implementation, after all N elements have been traversed and the target mixing information obtained from each traversal is output, the solving unit can also be used to receive a second selection operation; the N elements are mixed according to the mixing information in the output target mixing information corresponding to the second selection operation to obtain the target object.

[0023] In one implementation, the determining unit is further configured to determine the primary optimization model based on any one or more of the following: the wet ratio of each of the N elements, the amount of resources required to obtain a unit quantity of each element, and the chemical composition information contained in a unit quantity of target objects; and to determine the secondary optimization model corresponding to each element based on the current available amount of each element and the amount of each element required to generate a target quantity of target objects.

[0024] In one implementation, the first solution constraints determined by the determining unit include N wet ratio constraints corresponding to N elements and P chemical component constraints corresponding to P chemical components, where P is an integer greater than or equal to 1; the wet ratio constraints corresponding to each element include the minimum and maximum wet ratio values ​​for each element; and the chemical component constraints include the minimum and maximum values ​​of each chemical component contained in the target object generated by mixing the N elements.

[0025] On one hand, this application provides a data processing device, including a processor adapted to implement one or more computer programs; and a computer storage medium storing one or more computer programs, which are loaded and executed by the processor.

[0026] Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element.

[0027] Traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Generate at least one mixed information from a second solution constraint corresponding to the i-th element, a set of historical second solution constraints, and the current available quantities corresponding to the Ni elements that have not been traversed, where M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the set of historical second solution constraints includes the second solution constraints selected from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0028] The main optimization model is solved according to the first solution constraint and the M second solution constraints corresponding to the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint corresponding to the i-th element, and each solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraint;

[0029] Based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added as the selected second solution constraints to the historical second solution constraint set.

[0030] When all N elements have been traversed, output the target mixed information obtained in each traversal.

[0031] On one hand, this application provides a computer storage medium including a computer program that, when executed by a processor, can perform the following:

[0032] Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element.

[0033] Traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Generate at least one mixed information from a second solution constraint corresponding to each i-th element, a historical set of second solution constraints, and the current available quantities corresponding to the Ni elements that have not been traversed, where M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the selected second solution constraints from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0034] The main optimization model is solved according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint corresponding to the i-th element, and each solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraint;

[0035] Based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added as the selected second solution constraints to the historical second solution constraint set.

[0036] When all N elements have been traversed, output the target mixed information obtained in each traversal.

[0037] On one hand, this application provides a computer program product or computer program, the computer program product including a computer program stored in a computer storage medium. The processor of a data processing device reads the computer program from the computer storage medium, causing the data processing device to execute:

[0038] Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element.

[0039] Traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Generate at least one mixed information from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities of the Ni elements that have not been traversed; M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the selected second solution constraints from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0040] The main optimization model is solved according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint corresponding to the i-th element, and each solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraint;

[0041] Based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added as the selected second solution constraints to the historical second solution constraint set.

[0042] When all N elements have been traversed, output the target mixed information obtained in each traversal.

[0043] This application employs a data optimization model to determine the mixing information of N elements, so that the N elements can be mixed and processed to generate a target object based on the mixing information of the N elements. Specifically, the data optimization model includes a primary optimization model and N secondary optimization models corresponding to the N elements. That is, the process of determining the mixing information of N elements using the data optimization model is a process of optimizing multiple objectives. First, each secondary optimization model can be solved according to the first solution constraint to obtain the N residual thresholds that the N elements need to satisfy after mixing. Then, the i-th element among the N elements is traversed, and M second solution constraints are determined based on the residual threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. A second solution constraint corresponding to the i-th element... At least one mixed information can be generated by solving the constraints, the historical second set of constraints, and the current available quantities corresponding to the Ni elements that have not been traversed. The main optimization model is solved according to the first constraint, the historical second set of constraints, and M second constraints to obtain M solution results. Then, based on the M solution results, a target mixed information is selected from the mixed information corresponding to the M second constraints of the i-th element, and the second constraint corresponding to the target mixed information is added to the historical second constraint set as the selected second constraint. When all N elements have been traversed, the target mixed information obtained from each traversal is output. As can be seen from the above process, this embodiment transforms multi-objective optimization into single-objective optimization when solving multi-objective optimization problems. Furthermore, the constraints for solving the main optimization model can be reconstructed through the solution results of each sub-objective optimization, ultimately outputting multiple satisfactory solutions, i.e., multiple sets of target mixed information, effectively improving mixing efficiency and reducing costs. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in this application 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1A flowchart illustrating a data processing method provided in an embodiment of this application;

[0046] Figure 2 This is a schematic diagram of a target mixed information selection interface provided in an embodiment of this application;

[0047] Figure 3 This is a schematic diagram of an ingredient selection interface provided in an embodiment of this application;

[0048] Figure 4 A flowchart illustrating another data processing method provided in an embodiment of this application;

[0049] Figure 5 A schematic diagram illustrating a data processing method provided in this application for use in a sintering batching scenario;

[0050] Figure 6 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application;

[0051] Figure 7 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application. Detailed Implementation

[0052] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0053] Ingredient optimization technology (in this application embodiment, one ingredient can be referred to as an element) is currently widely used in industrial scenarios, such as sintering batching, cement raw material batching, steelmaking batching, and magnesium oxide powder batching. Ingredient optimization technology mainly transforms the actual industrial batching optimization problem into a data modeling problem, and solves it using operations research methods to obtain an optimal or satisfactory solution. Finally, the batching and mixing can be performed based on the optimal or satisfactory solution to obtain the corresponding industrial product.

[0054] In the field of batching optimization, multi-objective optimization challenges are frequently encountered. Multi-objectives refer to the indicators that need to be optimized. For example, in sintering batching applications, multi-objective optimization might mean minimizing both the average price of the batching and the inventory buildup of raw materials. This application proposes a data processing scheme based on batching optimization technology. It constructs a data optimization model by combining a primary optimization model, multiple secondary optimization models, and a first solution constraint. First, based on the first solution constraint, the multiple secondary optimization models are solved to obtain an optimal solution for each secondary optimization model to optimize its secondary objective. In this application embodiment, it is assumed that the secondary optimization objective used by each secondary optimization model is the remaining quantity of each element. Therefore, the optimal solution for each secondary optimization model is to minimize the remaining quantity of each element (here referred to as the remaining quantity threshold). Then, based on the solution results of each secondary optimization objective, a solution constraint is established for solving the primary optimization model. Finally, the primary optimization model is solved based on the newly established solution constraint and the first solution constraint, outputting multiple solution results. In this application, the master optimization model is used to optimize the master optimization objective, which can be the amount of processing resources (or cost) required to mix N elements.

[0055] When using the data processing scheme described above in this application, multi-objective optimization is transformed into single-objective optimization, and the constraints on the main optimization model can be reconstructed through the solution results of each sub-objective optimization. In the end, multiple satisfactory solutions can be output, that is, mixed information of multiple objectives, which effectively improves the mixing efficiency and reduces the cost.

[0056] The above data processing scheme can be executed by a data processing device, which can be a terminal, such as a smartphone, tablet computer, laptop computer, desktop computer, smart speaker, smartwatch, vehicle terminal, smart home appliance, smart voice interaction device, etc.; or, the data processing device can also be a server, such as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0057] In the aforementioned data processing scheme, the data optimization model and the current availability of each element can be stored in a cloud database. When the data processing scheme is executed, the data processing device retrieves this data from the cloud database. Alternatively, the data optimization model, the current availability of each element, or other data generated in the data processing scheme can also be stored in the blockchain. Blockchain is a new application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, it is a decentralized database, a chain of data blocks linked using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. Due to these characteristics of blockchain, the data stored on the blockchain is immutable, ensuring data security.

[0058] Based on the above data processing scheme, this application proposes a data processing method. Please refer to [link to relevant documentation]. Figure 1 , Figure 1 This is a flowchart illustrating a data processing method provided in an embodiment of this application. Figure 1 The data processing method described herein can be executed by a data processing device equipped with a data optimization model. The data optimization model is used to determine the mixed information of N elements, including a primary optimization model and N secondary optimization models corresponding to the N elements, as well as a first solution constraint condition acting on the primary optimization model and the N secondary optimization models. Here, N is an integer greater than 1. Figure 1 The data processing method includes, but is not limited to, the following steps:

[0059] Step S101: Solve each sub-optimization model according to the first solution constraint to obtain the N residual thresholds that need to be satisfied after the N elements are mixed.

[0060] The N elements can refer to the N kinds of raw materials required for the batching. The N elements are different in different application scenarios. For example, in the application of mixing and forming in sintering plants, the N elements can include raw materials such as slag powder, steel slag, high-silica cobalt, and iron oxide scale.

[0061] In one implementation, the sub-optimization model for each element is generated based on the current available quantity of each element and the usage quantity of each element required to generate the target number of objects. The role of the sub-optimization model is to solve for the residual quantity threshold after each element is mixed. Minimizing the residual quantity threshold that each element needs to satisfy after mixing can be taken as a secondary objective. Therefore, the role of N sub-optimization models is to optimize N secondary objectives. In the homogenization and sintering application in sintering plants, the residual quantity threshold can refer to the minimum inventory accumulation of each raw material after mixing.

[0062] In most applications involving the mixing of multiple elements, ensuring the quality of the resulting product requires constraints on the content of chemical elements. Therefore, the first constraint in this application can include chemical composition constraints. Assuming the mixed product has P chemical components, each corresponding to a chemical composition constraint, the first constraint includes P chemical composition constraints. Furthermore, during the mixing process, each required element is typically mixed with a liquid (such as water) in a specific ratio to form a wet mixture. Constraining the wet ratio of each element controls the amount used in the mixture. Therefore, the first constraint in this application can also include wet ratio constraints for each element. Since mixing requires N elements, the first constraint also includes N wet ratio constraints, one for each element. The wet ratio constraint for each element includes the minimum and maximum wet ratio values ​​for each element; the chemical composition constraint for each chemical component includes the minimum and maximum values ​​of each chemical component generated after mixing the N elements.

[0063] For example, if minimizing the residual threshold of each element after mixing is taken as a secondary objective, then the secondary objective corresponding to the i-th element can be expressed by the following formula:

[0064] Secondary_target(i) = S(X) i ), i = 1, 2, ..., N

[0065] Wherein, S(X) i This represents the threshold for the remaining amount of the i-th element after N elements are mixed to produce K tons of feed. The formula for calculating this threshold is as follows:

[0066] S(X i ) = R i -K*h(X i (V), i = 1, 2, ..., N

[0067] Among them, R i The remaining amount of the i-th element is indicated by the h function, which represents the amount of element used per unit of feed and can be calculated based on the wet ratio of the elements and the chemical composition contained in the elemental composition.

[0068] Based on this, the suboptimal model corresponding to the i-th element can be expressed as:

[0069]

[0070] The suboptimal model corresponding to the i-th element and the set of constraints used to solve the suboptimal model can be represented by the following mathematical equation:

[0071]

[0072]

[0073] In this context, the target object generated after mixing, i.e., the various chemical components contained in the ingredients, is denoted as c(j), where P represents the total number of P chemical components, L_b(j) represents the minimum content ratio of the j-th chemical component after mixing, and U_b(j) represents the maximum content ratio; typically, U_b(j) ≤ 100 and L_b(j) ≥ 0 are taken; X i Let L_X(i) represent the wet ratio of the i-th element, L_X(i) represent the minimum wet ratio of the i-th raw material, and U_X(i) represent the maximum wet ratio; typically, L_X(i) ≥ 0 and U_X(i) ≥ 0 are taken. In actual production, the content ratio of chemical components and the maximum and minimum wet ratios can be manually configured and adjusted according to business needs.

[0074] In the embodiments of this application, the smaller the remaining amount threshold of an element, the more the element is used during the mixing process.

[0075] Step S102: Traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining threshold corresponding to the i-th element and the current available quantity of the i-th element, where M is an integer greater than 1.

[0076] In this embodiment, at least one mixing information is generated from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities corresponding to the Ni elements that have not been traversed, where i is greater than 1 and less than or equal to N. The historical set of second solution constraints includes the selected second solution constraint from the M second solution constraints corresponding to each of the i-1 traversed elements. The mixing information may include wet proportion information corresponding to each of the N elements, the usage amount of the N elements required to generate each unit of feed, and other feed information. Furthermore, since the mixing process requires the use of N elements, and these N elements correspond to N sub-optimization models, and based on the N residual thresholds obtained from solving the N sub-optimization models, M second solution constraints can be generated. Therefore, it is necessary to traverse the N elements to determine the second solution constraints corresponding to each element.

[0077] In one implementation, each of the M second solution constraints corresponding to the i-th element is represented by the remaining quantity interval of the i-th element when mixing the N elements. The remaining quantity interval includes the minimum remaining quantity and the maximum remaining quantity. The t-th second solution constraint is determined based on the remaining quantity threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Specifically, this can be achieved by: using the remaining quantity threshold corresponding to the currently traversed i-th element as the minimum remaining quantity in the t-th remaining quantity interval; determining the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t; and adding the minimum remaining quantity in the t-th remaining quantity interval and the remaining quantity increment of the t-th remaining quantity interval to obtain the maximum remaining quantity of the t-th remaining quantity interval. Here, t is an integer greater than or equal to 1 and less than M.

[0078] In one implementation, the method for determining the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t may include: subtracting the current available quantity of the i-th element from the remaining quantity threshold, and multiplying M and t; dividing the result of the subtraction operation by the result of the multiplication operation to obtain the remaining quantity increment of the t-th remaining quantity interval. Optionally, this method for determining the remaining quantity increment of the t-th remaining quantity can be expressed mathematically as follows:

[0079]

[0080] Wherein, S(X) i R represents the threshold for the remaining quantity of the i-th element; i Let S(i) represent the current available quantity of the i-th element; S(i) represent the remaining quantity threshold; and t represent the number of remaining quantity intervals, where t represents the t-th remaining quantity interval. Based on this, the second solution constraint can be expressed as:

[0081]

[0082] It should be understood that the M second-order constraints corresponding to the i-th element are determined based on the remaining quantity threshold and the current available quantity of the i-th element. Therefore, it can be known that any second-order constraint corresponding to the i-th element is used to constrain the usage of the i-th element when mixing N elements. For example, any second-order constraint might constrain the minimum usage of the i-th element to be x, and the maximum usage to be x+y. Since the Ni elements have not yet been traversed, the second-order constraints corresponding to each of the Ni elements have not yet been determined; only the remaining quantity thresholds of each element are determined. That is, at this point, we only know the minimum inventory required for each of the Ni elements when mixing N elements, but we do not know the specific usage range of each element. The usage of each untraversed element during mixing has multiple possible values. Therefore, based on the selected second-order constraints corresponding to each of the i-1 traversed elements, any second-order constraint corresponding to the i-th element, and the current available quantity of each of the Ni untraversed elements, we can determine multiple mixing information for mixing N elements.

[0083] Step S103: Solve the main optimization model according to the first solution constraint, the historical second solution constraint set, and M second solution constraints to obtain M solution results.

[0084] As mentioned above, each of the M second solution constraints corresponding to the i-th element corresponds to one or more mixing information. Since a solution result corresponds to one second solution constraint of the i-th element, the M solution results obtained from solving the main optimization model can include the amount of processing resources required to mix the N elements according to the mixing information corresponding to the respective second solution constraints, information on each of the P chemical components contained in the target object generated by mixing the N elements according to the mixing information (such as the proportions of various chemical components), and the wet mixing ratio information corresponding to each of the N elements.

[0085] Optionally, minimizing the processing resources required to mix N elements can be the primary objective; that is, the master optimization model is used to optimize the primary objective. In the sintering plant's homogenization and stacking application, the processing resources can be cost. For example, in the sintering plant's homogenization and stacking application, the primary objective can be to minimize the cost of generating each unit of material, and the secondary objective can be to minimize the inventory accumulation of various elements after batching (i.e., after mixing). In step S102, any mixing information generated based on a second solution constraint corresponding to the i-th element, the historical set of second solution constraints, and the current available quantities of the Ni elements that have not been traversed can be a batching scheme, including batching information such as the usage of each element in the batching process; a solution result obtained from solving the master optimization model can be information such as the cost required to generate each unit of material according to a batching scheme.

[0086] Step S104: Based on the M solution results, select the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and add the second solution constraint corresponding to the target mixed information as the selected second solution constraint to the historical second solution constraint set.

[0087] As mentioned above, each mixed information corresponds to a second solution constraint of the i-th element. The second solution constraint corresponding to the selected target mixed information can be determined as the selected second solution constraint among the M second solution constraints corresponding to the i-th element, and added to the historical second solution constraint set.

[0088] In one implementation, the data processing device can determine the mixed information with the minimum processing resource requirement from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, and then set the determined mixed information as the target mixed information. The minimum processing resource requirement may include information such as the lowest cost per unit of material generated, which can be configured by the user according to the needs of the business scenario.

[0089] In one implementation, the data processing device can output the mixed information corresponding to the M second solution constraints for the i-th element, and output the processing resource amount corresponding to each mixed information based on the M solution results. Then, it receives a first selection operation and determines the mixed information corresponding to the first selection operation as the target mixed information. Optionally, the first selection operation can be executed by a first instruction configured by the user in the data processing device. The first instruction can be pre-configured in the data processing device according to the user's settings. For example, if the user wants to always select the mixed information with the smallest processing resource amount, then when the first instruction is executed, the data processing device can select the mixed information with the smallest processing resource amount from multiple mixed information options as the target mixed information. Alternatively, if the user wants to always select the mixed information with a moderate processing resource amount, then when the first instruction is executed, the data processing device can select the mixed information with a moderate processing resource amount from multiple mixed information options as the target mixed information. Optionally, the first selection operation can be input by the user in the data processing device. In the specific implementation, after obtaining multiple mixed information corresponding to M second solution constraints, the data processing device displays a target mixed information selection interface to receive the user's real-time selection operation, that is, the user selects a target mixed information through the target mixed information selection interface.

[0090] For example, please refer to Figure 2 , Figure 2 This diagram illustrates a target mixing information selection interface provided in an embodiment of this application. The diagram includes multiple mixing information items, displaying the wet mixing ratio, chemical composition, current available quantity, remaining quantity threshold, and processing resource quantity for each item. Users can view detailed information by clicking the corresponding frame. Users can also select a mixing information item as the target mixing information by clicking the corresponding number box.

[0091] Step S105: When all N elements have been traversed, output the target mixed information obtained in each traversal.

[0092] As can be seen from the above steps S101-S104, each time an element is traversed, one target mixed information can be output. When all N elements have been traversed, N target mixed information can be output.

[0093] In one implementation, the data processing device can receive a second selection operation and, according to the matching information in the output target mixing information, mix the N elements to obtain the target object. Optionally, the second selection operation can be executed by a second instruction configured by the user in the data processing device. This second instruction can be pre-configured in the data processing device according to the user's settings; for example, if the user wants to minimize the processing resources required for mixing based on the target mixing information, then when the second instruction is executed, the data processing device can select the target mixing information with the minimum processing resources from multiple candidate target mixing information as the basis for mixing. Optionally, the second selection operation can be input by the user in the data processing device. Specifically, after traversing all N elements and generating N target mixing information, the data processing device displays a batching scheme selection interface to receive the user's real-time selection operation; that is, the user selects a target mixing information for mixing through this batching scheme selection interface.

[0094] For example, please refer to Figure 3 , Figure 3 This diagram illustrates a formulation selection interface provided in an embodiment of this application. The diagram includes icons for adding a formulation, importing, and exporting, as well as formulation schemes 1 through 6. Clicking the "View More" icon allows users to view the remaining formulation schemes. Each formulation scheme has a "Select" icon and a "View" icon below it. Users can click the "Select" icon to designate the formulation as the target; clicking the "View" icon displays detailed information about the formulation, including the wet proportions of each element and chemical composition constraints. Clicking the "Add Formulation" icon allows users to import relevant data from the formulation; users can also click the "Export" icon and select one or more formulation schemes from the diagram to export the selected formulations as files from the data processing device.

[0095] This application employs a data optimization model to determine the mixing information of N elements, so that the N elements can be mixed and processed to generate a target object based on the mixing information of the N elements. Specifically, the data optimization model includes a primary optimization model and N secondary optimization models corresponding to the N elements. That is, the process of determining the mixing information of the N elements using the data optimization model is a multi-objective optimization process. First, the secondary optimization models are used to obtain the N residual thresholds corresponding to the N elements. Then, the i-th element among the N elements is traversed, and M second solution constraints are determined based on the residual threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. According to the first solution constraints, the historical set of second solution constraints, and the M second solution constraints corresponding to the i-th element, the primary optimization model is solved to obtain M solution results. Based on the M solution results, target mixing information is selected from the mixing information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixing information are added as selected second solution constraints to the historical set of second solution constraints. When all N elements are traversed, the target mixing information obtained from each traversal is output. This target mixing information includes the ingredient mixing scheme. In solving multi-objective optimization problems, this embodiment transforms multi-objective optimization into single-objective optimization. Furthermore, the constraints on the main optimization model can be reconstructed using the solution results of each sub-objective optimization, ultimately outputting multiple satisfactory solutions, i.e., multiple sets of target mixing information. This effectively improves mixing efficiency and reduces costs.

[0096] Please see Figure 4 , Figure 4 This is a flowchart illustrating another data processing method provided in this application embodiment, which can also be applied to similar industrial batching optimization scenarios with multi-objective optimization needs, such as sintering batching, cement raw material batching, and magnesium oxide powder batching. The execution entity of this data processing method is a data processing device configured with a data optimization model. The data optimization model is used to determine the mixed information of N elements, including a primary optimization model and N secondary optimization models corresponding to the N elements, as well as a first solution constraint condition acting on the primary optimization model and the N secondary optimization models. Here, N is an integer greater than 1. This data processing method includes, but is not limited to, the following steps:

[0097] Step S401: Construct the corresponding sub-optimization model based on N elements, and solve each sub-optimization model according to the first solution constraint.

[0098] In the embodiments of this application, the sub-optimization model corresponding to each element is generated based on the current available quantity of each element and the usage quantity of each element required to generate the target number of target objects. The sub-optimization model is used to solve the secondary objectives of each element and obtain the N remaining quantity thresholds that need to be satisfied after N elements are mixed and processed.

[0099] Optionally, the first set of constraints includes the wet ratio constraint for each of the N elements and the P chemical component constraints corresponding to the P chemical components. The wet ratio constraint for each element includes the minimum and maximum wet ratio values ​​for that element; each chemical component constraint includes the minimum and maximum proportions of each chemical component contained in the target object generated from the mixture of the N elements.

[0100] For example, please refer to Table 1, which is an example table of the chemical composition, wet ratio and unit price of each element required in the business scenario of sintering feedstock.

[0101] Table 1. Elemental chemical composition, wet blending ratio, and unit valence

[0102]

[0103]

[0104] For example, Table 2 shows the wet proportion constraints of the required elements in the business scenario of sintering batching. Table 3 shows the chemical composition constraints of the batching after batching, i.e., after mixing.

[0105] Table 2 Constraints on Wet Mixing Ratio

[0106]

[0107] Table 3 Chemical Composition Constraints

[0108] upper limit 59 4.8 100 100 1.98 100 100 100 lower limit 0 4.6 0 0 0 0 0 0

[0109] For example, if minimizing the residual threshold of each element after mixing is taken as a secondary objective, then the secondary objective corresponding to the i-th element can be expressed by the following formula:

[0110] Secondary_target(i) = S(X) i ), i = 1, 2, ..., N

[0111] Wherein, S(X) i () represents the threshold of the remaining amount of the i-th element after N elements are mixed to produce K tons of feed. The calculation formula is as follows:

[0112] S(Xi ) = R i -K*h(X i (V), i = 1, 2, ..., N

[0113] Among them, R i Let represent the remaining amount of the i-th element; the h function represents the amount of element used per unit of feed, which can be calculated based on the wet ratio of the elements and the chemical composition of the feed. Based on this, the suboptimal model corresponding to the i-th element can be expressed by the following formula:

[0114]

[0115] The suboptimal model and the set of constraints used to solve it can be represented by the following mathematical equations:

[0116]

[0117]

[0118] In this context, the target object generated after mixing, i.e., the various chemical components contained in the ingredients, is denoted as c(j), where P represents the total number of P chemical components, L_b(j) represents the minimum content ratio of the j-th chemical component after mixing, and U_b(j) represents the maximum content ratio; typically, U_b(j) ≤ 100 and L_b(j) ≥ 0 are taken; X i Let L_X(i) represent the wet ratio of the i-th element, L_X(i) represent the minimum wet ratio of the i-th raw material, and U_X(i) represent the maximum wet ratio; typically, L_X(i) ≥ 0 and U_X(i) ≥ 0 are taken. In actual production, the content ratio of chemical components and the maximum and minimum wet ratios can be manually configured and adjusted according to business needs.

[0119] In this embodiment, a smaller remaining quantity threshold for an element indicates that the element is used more during the mixing process. The remaining quantity threshold may include the inventory accumulation of each element after the mixing process; N elements may refer to the raw materials required for N different ingredients.

[0120] Step S402: Sort the N elements from highest to lowest according to their mixed importance.

[0121] In this embodiment, the mixed importance of each element can be set by the user. Optionally, the mixed importance can be sorted according to the demand of the elements in the mixture, the current availability of each element, or the amount of resources required by each element. Sort according to the mixed importance before proceeding to the next step, which is conducive to the standardized execution of the data processing method.

[0122] Step S403: Traverse the N elements in descending order of importance. When traversing the i-th element, determine M second solution constraints based on the remaining threshold and the current available quantity corresponding to the i-th element, where M is an integer greater than 1.

[0123] In this embodiment, at least one mixed information is generated from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities corresponding to the Ni elements that have not been traversed, where i is greater than 1 and less than or equal to N. The historical set of second solution constraints includes the selected second solution constraint from the M second solution constraints corresponding to each of the i-1 traversed elements.

[0124] In one implementation, each of the M second solution constraints corresponding to the i-th element is represented by the remaining quantity interval of the i-th element when mixing the N elements. The remaining quantity interval includes the minimum remaining quantity and the maximum remaining quantity. The t-th second solution constraint is determined based on the remaining quantity threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Specifically, this can be achieved by: using the remaining quantity threshold corresponding to the currently traversed i-th element as the minimum remaining quantity in the t-th remaining quantity interval; determining the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t; and adding the minimum remaining quantity in the i-th remaining quantity interval and the remaining quantity increment of the t-th remaining quantity interval to obtain the maximum remaining quantity of the t-th remaining quantity interval. Here, t is an integer greater than or equal to 1 and less than M. Optionally, each second solution constraint can be divided into M remaining quantity intervals by average partitioning.

[0125] In one implementation, the method for determining the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t may include: subtracting the current available quantity of the i-th element from the remaining quantity threshold, and multiplying M and t; dividing the result of the subtraction operation by the result of the multiplication operation to obtain the remaining quantity increment of the t-th remaining quantity interval. Optionally, this method for determining the remaining quantity increment of the i-th remaining quantity can be expressed mathematically as follows:

[0126]

[0127] Wherein, S(X) i R represents the threshold for the remaining quantity of the i-th element; i Let S(i) represent the current available quantity of the i-th element; S(i) represent the remaining quantity threshold; and t represent the number of remaining quantity intervals, where t represents the t-th remaining quantity interval. Based on this, the second solution constraint can be expressed as:

[0128]

[0129] It should be understood that the M second-order constraints corresponding to the i-th element are determined based on the remaining quantity threshold and the current available quantity of the i-th element. Therefore, it can be known that any second-order constraint corresponding to the i-th element is used to constrain the usage of the i-th element when mixing N elements. For example, any second-order constraint might constrain the minimum usage of the i-th element to be x and the maximum usage to be x+y. Since the Ni elements have not yet been traversed, the second-order constraints corresponding to each of the Ni elements have not yet been determined. Only the remaining quantity threshold of each element has been determined. That is to say, at this time, we only know the minimum inventory required for each of the Ni elements when mixing N elements, but we do not know the specific usage range of each element. The second-order constraint corresponding to the usage of each untraversed element during mixing has multiple possible values. Based on the selected second solution constraints corresponding to each of the i-1 elements that have been traversed, any second solution constraint corresponding to the i-th element, and the current available quantities of the Ni untraversed elements, multiple mixing information for mixing N elements can be determined.

[0130] Step S404: Determine whether the i-th element in the current traversal is the first element in the sorted sequence of the N elements.

[0131] In this embodiment, the element at the top of the sorted list has the highest mixed importance and is processed differently from other elements. If the current element is the first element in the sorted list of N elements, then step S405 is executed; if the current element is not the first element in the sorted list of N elements, then step S406 is executed.

[0132] S405: Solve the main optimization model according to the first solution constraint and the M second solution constraints corresponding to the i-th element to obtain the M solution results corresponding to the i-th element, and then execute step S408.

[0133] In this embodiment, when the i-th element is the first element in the mixed importance ranking, the historical second solution constraint set is empty. The first solution constraint and the M second solution constraints corresponding to the i-th element can constitute the constraint set of the master optimization model, used to solve the master optimization model, which is used to optimize the main objective. For the element with the mixed importance ranking first, for example, the master optimization model can be represented by the following formula:

[0134]

[0135] Here, f(X,P,V) is the equation for solving the main objective. The specific calculation formula can be obtained by combining the business scenario and sorting out the relationship between the wet ratio X, unit price P, and content ratio V of each chemical component corresponding to the raw materials.

[0136] Based on the master optimization model, and the M second and first solution constraints corresponding to this element, the equations of the master optimization model are solved as follows:

[0137]

[0138]

[0139] Where i = 1 represents the first element after sorting by mixing importance; t = z represents the residual threshold of the z-th residual interval of the second solution constraint corresponding to this element. As shown in the equation above, one element corresponds to M residual intervals, and each residual interval corresponds to one equation. Solving the master optimization model requires solving the M equations shown above to obtain M solution results corresponding to the current element. One solution result includes the amount of processing resources required to mix N elements according to the mixing information corresponding to the second solution constraint. This amount of processing resources can be the unit price of each element.

[0140] Step S406: Obtain the set of historical second solution constraints before the current traversal.

[0141] In this embodiment, if the current element is not the element with the highest importance among the N elements, in order to ensure the rationality and completeness of the final generated target mixture information, it is necessary to integrate the solution results of the master optimization model for all N elements. Therefore, it is necessary to obtain the second solution constraints corresponding to the target mixture information obtained from all previous traversals. In step S408, these second solution constraints are added to the historical second solution constraint set as selected second solution constraints. Then, the obtained historical second solution constraint set is also used as the constraint condition for solving the master optimization model corresponding to the current element.

[0142] Step S407: Solve the main optimization model according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain the M solution results corresponding to the i-th element in the current traversal.

[0143] In this embodiment, the acquired historical second solution constraint set, the first solution constraint, and M second solution constraints are combined to form a new constraint set, which is used to solve the main optimization model. The M solution results include the mixing information corresponding to the second constraint of the current element, and the mixing information corresponding to the second solution constraints of other elements. Each mixing information may include the wet mix ratio information, chemical composition information, current available quantity, remaining quantity threshold, and processing resource quantity corresponding to each element. For example, if the element is the second most important element in the mixing ranking, the main optimization model and the constraint set used to solve the main optimization model can be represented by the following equation:

[0144]

[0145]

[0146] Where i = 2 represents the second element after sorting by mixed importance; t = z represents the residual threshold of the z-th residual interval of the second solution constraint corresponding to this element. The second solution constraint corresponding to the element ranked first in the mixed importance sorting is expressed by the following formula:

[0147]

[0148] Where t1 represents the residual threshold of the t1th residual interval of the second solution constraint corresponding to the element. Similarly, when traversing the elements corresponding to i = 3, 4, 5, ..., N, the historical set of second solution constraints needs to be used to solve the main optimization model.

[0149] In one implementation, the solution result obtained after solving the main optimization model can be judged. If the solution result does not meet the preset conditions, the main optimization model solution can be skipped, and the traversal of the next element can be executed. The preset conditions can be determined by the data processing device, or by the user after the data processing device outputs the results; the user can also change the configuration of the preset conditions.

[0150] For example, in the homogenization and stacking application of a sintering plant, the primary objective may be to minimize the cost of generating each unit of material, and the secondary objective may be to minimize the inventory accumulation of various elements after batching. Then, the mixing information obtained by solving the secondary optimization model can be a batching scheme, including the usage of each element in the batching process and other batching information. The solution obtained by solving the primary optimization model can be information such as the cost required to generate each unit of material according to a batching scheme, as well as the mixing information corresponding to the second solution constraint used to solve the primary optimization model, that is, the batching information of the batching scheme.

[0151] Step S408: Obtain the first selection operation, determine the target mixing information corresponding to the i-th element of the current traversal based on the M solution results, and add the second solution constraint corresponding to the target mixing information as the selected second solution constraint to the historical second solution constraint set.

[0152] As mentioned above, each mixed information corresponds to a second solution constraint of the i-th element. The second solution constraint corresponding to the selected target mixed information can be determined as the selected second solution constraint among the M second solution constraints corresponding to the i-th element, and added to the historical second solution constraint set.

[0153] In this embodiment, for each element obtained by executing step S405 or step S407, M solution results are generated. Among them, a second solution constraint corresponding to the i-th element, a set of historical second solution constraints, and the current available quantities corresponding to the Ni elements that have not been traversed generate at least one mixture information. Each time the main optimization model is solved, the data processing device can obtain a first selection operation and determine the mixture information corresponding to the first selection operation as the target mixture information. Optionally, the first selection operation can be executed by a first instruction configured by the user in the data processing device; the first instruction can be pre-configured in the data processing device according to the user's settings. For example, if the user wants to always select the mixture information with the smallest processing resource amount, then when the first instruction is executed, the data processing device can select the mixture information with the smallest processing resource amount from multiple candidate mixture information as the target mixture information; or, if the user wants to always select the mixture information with a moderate processing resource amount, then when the first instruction is executed, the data processing device can select the mixture information with a moderate processing resource amount from multiple candidate mixture information as the target mixture information. Optionally, the first selection operation can be input by the user in the data processing device. In practical implementation, after obtaining multiple mixed information corresponding to M second solution constraints, the data processing device displays a... Figure 2 The target mixed information selection interface shown is used to receive the user's real-time selection operation, that is, the user selects a target mixed information through this target mixed information selection interface.

[0154] In one implementation, the data processing device can determine the mixed information with the minimum processing resource requirement from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, and then determine the determined mixed information as the target mixed information.

[0155] Step S409: If all N elements have been traversed, proceed to step S410; if there are still elements among the N elements that have not been traversed, continue traversing from step S403.

[0156] When the last element out of N elements is reached, the principal optimization model and the set of constraints used to solve the principal optimization model can be represented as follows:

[0157]

[0158]

[0159] Among them, t i This represents the threshold value of the t-th residual interval corresponding to the second solution constraint condition of the element.

[0160] Step S410: Output the target mixed information obtained in each traversal.

[0161] As can be seen from steps S401-S409 above, each time an element is traversed, one target mixed information can be output. When all N elements have been traversed, N target mixed information can be output.

[0162] Step S411: Obtain the second selection operation, and perform a mixed processing on N elements to obtain the target object.

[0163] Optionally, after outputting the target mixing information obtained in each iteration, the data processing device can also receive a second selection operation; the N elements are then mixed according to the mixing information corresponding to the second selection operation in the output target mixing information to obtain the target object. Optionally, the second selection operation can be executed by a second instruction configured by the user in the data processing device. The second instruction can be pre-configured in the data processing device according to the user's settings. For example, if the user wants to minimize the processing resources required for mixing based on the target mixing information, then when the second instruction is executed, the data processing device can select the target mixing information with the minimum processing resources from multiple candidate target mixing information as the basis for mixing. Optionally, the second selection operation can be input by the user in the data processing device. In specific implementation, after traversing all N elements and generating N target mixing information, the data processing device displays an array such as... Figure 3 The ingredient selection interface shown receives real-time selections from users, allowing them to select a target mixture for mixing.

[0164] To facilitate understanding of the data processing method provided in the embodiments of this application, please refer to Table 4, which is an example of the generated batching scheme in the sintering batching scenario.

[0165] Table 4 Ingredient Scheme

[0166]

[0167]

[0168] This application employs a data optimization model to determine the mixing information of N elements, so that the N elements can be mixed and processed to generate a target object based on the mixing information of the N elements. Specifically, the data optimization model includes a primary optimization model and N secondary optimization models corresponding to the N elements. That is, the process of determining the mixing information of N elements using the data optimization model is a multi-objective optimization process. First, the N elements are sorted from high to low according to their mixing importance. Then, according to their mixing importance from high to low, the secondary optimization models are used to obtain the N residual thresholds corresponding to each of the N elements, and M second solution constraints are determined. If the current element is the most important element, the primary optimization model is solved according to the first solution constraint and the M second solution constraints corresponding to that element to obtain M solution results. Then, according to the obtained first selection operation, the target mixing information is determined from the mixing information corresponding to the M second solution constraints. If the current element is not the element with the highest mixing importance, then the historical second set of solution constraints needs to be obtained and used to solve the main optimization to obtain M solution results. The target mixing information is then determined from these results based on the first selection operation. When all N elements have been traversed, the target mixing information obtained from each traversal is output. Finally, the N elements are mixed according to the obtained second selection operation to obtain the target object. The target mixing information includes the ingredient ratio. In solving multi-objective optimization problems, this embodiment transforms multi-objective optimization into single-objective optimization. Furthermore, the constraints for solving the main optimization model can be reconstructed using the solution results of each sub-objective optimization, ultimately outputting multiple satisfactory solutions, i.e., multiple sets of target mixing information, effectively improving mixing efficiency and reducing costs. Simultaneously, it can obtain the user's selection operation to instruct the determination of target mixing information and the mixing of N elements, facilitating user control and management of the data processing process.

[0169] The above Figure 1 and Figure 4The data processing method described can be applied to the business scenario of sintering batching. In this scenario, "element" refers to the raw materials required for batching, "residual quantity threshold" refers to the minimum inventory accumulation of each raw material after batching, and "processing resource quantity" refers to the average cost of generating each unit of batching. During the sintering production process, the chemical composition and inventory accumulation of raw materials may change over time, and the quality detection of sintered ore has a lag. As the first step in sintering, batching typically uses mathematical models as an important tool to solve the batching ratio. The solution of the ratio is closely related to the sintering process, which includes a large amount of uncertain information. Therefore, a simple numerical model cannot describe the overly complex process, and the results of mathematical models are only temporarily effective. When the composition of the sintered ore fluctuates, adjustments still need to be made based on human experience, thus the adaptability of mathematical models is poor. Therefore, applying the data processing method in this embodiment to the sintering batching scenario, adding theoretical guidance from data optimization models to human experience, can more effectively optimize the sintering batching. The following is a combination of... Figure 5 This section will introduce the specific implementation process of data processing methods in the sintering batching scenario:

[0170] The primary objective is to minimize the average cost per unit of feedstock, while the secondary objective is to minimize the inventory accumulation of each raw material after feedstock preparation. The first solution constraint for each raw material includes chemical composition constraints for each chemical component in the feedstock after mixing, and wet proportion constraints for each element. Each second solution constraint is represented by the inventory accumulation range obtained from the sub-optimization model. Based on this, the data processing method applied to the sintering feedstock scenario includes:

[0171] First, the wet blending ratio of each raw material is used as the decision variable. Based on this decision variable, a primary objective function and a secondary objective function are constructed, along with constraints on chemical composition and wet blending ratio. These constraints are used to construct a secondary optimization model for the N raw materials. The raw materials are ranked according to their mixing importance, and the minimum inventory accumulation of the N raw materials is obtained by solving the model sequentially from highest to lowest mixing importance. This allows us to determine the M secondary constraints for each raw material. If the current raw material has the highest mixing importance, the primary optimization model is solved based on the primary constraints and the M secondary constraints for that raw material to obtain the corresponding M solutions. If the current raw material is not the one with the highest mixing importance, for example... Figure 5The solution steps for the main optimization model 2 require obtaining the second solution constraint 1 corresponding to the target mixture information 1 obtained before the current traversal. Combining this with the second solution constraint 2, the main optimization model is solved to obtain M solution results. The target mixture information 2 is then determined from these results based on the first selection operation. Each traversal then determines a batching scheme based on the obtained second selection operation and the target mixture information. Traversing each raw material generates M batching schemes. Each scheme includes the average cost per unit of batching, the minimum inventory accumulation for each raw material, the wet mixing ratio for each raw material, and the chemical composition content ratio after batching. Users can select the appropriate batching scheme according to different business needs.

[0172] This application employs a data optimization model to determine the mixing information of N raw materials, enabling the subsequent mixing of these materials to generate a batching scheme. When solving multi-objective optimization problems, this application transforms multi-objective optimization into single-objective optimization. Furthermore, the constraints on the main optimization model can be reconstructed using the solution results of each sub-objective optimization, ultimately outputting multiple satisfactory solutions, i.e., multiple sets of objective mixing information. This effectively improves mixing efficiency and reduces costs. Simultaneously, it can acquire user selection operations to instruct the determination of objective mixing information and the mixing of the N elements, facilitating user control and management of the data processing process.

[0173] Based on the above-described data processing method embodiments, this application provides a data processing apparatus, see [link to relevant documentation]. Figure 6 This is a schematic diagram of a data processing device provided in an embodiment of this application. The data processing device can be a computer program (including program code) running on a data processing device, such as an application software. The data processing device can be used to execute the corresponding steps in the method provided in the embodiment of this application. The data processing device can be configured with a data optimization model, which can be used to determine the mixed information of N elements. The data optimization model includes a primary optimization model, N secondary optimization models corresponding to the N elements, and a first solution constraint condition acting on the primary optimization model and the N secondary optimization models; N is an integer greater than 1. Figure 6 The data processing device 600 may include: a solving unit 601, a determining unit 602, and an output unit 603. Wherein:

[0174] The solving unit 601 is used to solve each sub-optimization model according to the first solving constraint conditions to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one element corresponding to one residual threshold.

[0175] The determining unit 602 is used to traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining amount threshold corresponding to the currently traversed i-th element and the current available amount of the i-th element. At least one mixed information is generated by a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available amounts corresponding to the Ni elements that have not been traversed, where M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the second solution constraints selected from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0176] The solving unit 601 is further configured to solve the main optimization model according to the first solving constraint, the historical set of second solving constraints, and the M second solving constraints corresponding to the i-th element to obtain M solution results; one solution result corresponds to one second solving constraint corresponding to the i-th element, and one solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solving constraint;

[0177] The determining unit 602 is further configured to select target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, and add the second solution constraint corresponding to the target mixed information as the selected second solution constraint to the historical second solution constraint set.

[0178] The output unit 603 is used to output the target mixed information obtained in each traversal when all N elements have been traversed.

[0179] In one implementation, when determining the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, the determining unit 602 performs the following steps:

[0180] Based on the M solution results, determine the mixed information with the minimum processing resource requirement from the mixed information corresponding to the M second solution constraints of the i-th element; and determine the mixed information as the target mixed information.

[0181] In one implementation, when determining the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, the determining unit 602 performs the following steps:

[0182] Output the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and output the processing resource amount corresponding to each mixed information based on the M solution results; receive the first selection operation; determine the mixed information corresponding to the first selection operation as the target mixed information.

[0183] In one implementation, each of the M second solution constraints corresponding to the i-th element is represented by the remaining quantity interval of the i-th element when mixing the N elements. The remaining quantity interval includes the minimum remaining quantity and the maximum remaining quantity. The determining unit 602 is further configured to use the remaining quantity threshold corresponding to the currently traversed i-th element as the minimum remaining quantity in the t-th remaining quantity interval; determine the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t; and add the minimum remaining quantity in the t-th remaining quantity interval and the remaining quantity increment of the t-th remaining quantity interval to obtain the maximum remaining quantity of the t-th remaining quantity interval. Here, t is an integer greater than or equal to t and less than M.

[0184] In one implementation, the solving unit 601 is further configured to subtract the current available quantity of the i-th element from the remaining quantity threshold, and multiply the M elements and t; and divide the result of the subtraction operation with the result of the multiplication operation to obtain the remaining quantity increment of the t-th remaining quantity interval.

[0185] In one implementation, the determining unit 602 is further configured to determine whether the i-th element is the first element to be traversed among the N elements; if so, the historical second solution constraint set is empty, and the solving unit 601 is further configured to solve the main optimization model using the first solution constraint and the M second solution constraints corresponding to the i-th element to obtain M solution results. If the determining unit 602 determines that the i-th element is not the first element to be traversed among the N elements, the solving unit 601 is further configured to obtain the historical second solution constraint set before the current traversal; and solve the main optimization model using the obtained historical second solution constraint set, the first solution constraint, and the M second solution constraints corresponding to the i-th element to obtain M solution results.

[0186] In one implementation, when all N elements are traversed, after outputting the target mixing information obtained from each traversal, the solving unit 601 can also be used to receive a second selection operation; the N elements are mixed according to the mixing information in the output target mixing information corresponding to the second selection operation to obtain the target object.

[0187] In one implementation, the determining unit 602 is further configured to determine a primary optimization model based on any one or more of the following: the wet ratio of each of the N elements, the amount of resources required to obtain a unit quantity of each element, and the chemical composition information contained in a unit quantity of target objects; and to determine a secondary optimization model corresponding to each element based on the current available amount of each element and the amount of each element required to generate a target quantity of target objects.

[0188] In one implementation, the first solution constraint determined by the determining unit 602 includes N wet ratio constraints corresponding to N elements and P chemical component constraints corresponding to P chemical components, where P is an integer greater than or equal to 1; the wet ratio constraint corresponding to each element includes the minimum and maximum wet ratio of each element; the chemical component constraint includes the minimum and maximum values ​​of each chemical component contained in the target object generated by mixing N elements.

[0189] According to one embodiment of this application, Figure 1 and Figure 4 The data processing method shown can involve various steps that can be derived from... Figure 6 This is performed by each unit in the data processing apparatus shown. For example, Figure 1 Both steps S101 and S103 can be performed by... Figure 6 The solution unit 601 in the data processing device shown is used to execute steps S102 and S104, which can be performed by... Figure 6 The determination unit 602 in the data processing device shown is responsible for executing step S105, which can be performed by... Figure 6 The output unit 603 in the data processing device shown is used to perform this; for example, Figure 4 Steps S401, S405, S407, and S411 can be derived from... Figure 6 The solution unit 601 in the data processing device shown executes the steps S402, S403, S404, S406, S408, and S409, which can be performed by... Figure 6 The determination unit 602 in the data processing device shown is responsible for executing step S410, which can be performed by... Figure 6 The output unit 603 in the data processing device shown is used to perform this operation.

[0190] According to another embodiment of this application, Figure 6The data processing apparatus shown can be composed of individual or combined units into one or more other units, or some of the units can be further divided into multiple functionally smaller units. This achieves the same operation without affecting the technical effects of the embodiments of this application. The above-mentioned units are based on logical function division. In practical applications, the function of one unit can be implemented by multiple units, or the function of multiple units can be implemented by one unit. In other embodiments of this application, the data processing apparatus may also include other units. In practical applications, these functions can also be implemented with the assistance of other units, and can be implemented collaboratively by multiple units.

[0191] According to another embodiment of this application, the following can be achieved by running on a general-purpose computing device, such as a computer, which includes processing elements and storage elements such as a central processing unit (CPU), random access memory (RAM), and read-only memory (ROM), a device capable of performing operations such as... Figure 1 and Figure 4 The computer program (including program code) for each step involved in the corresponding method shown, to construct such... Figure 6 The data processing apparatus shown herein, and the data processing method for implementing the embodiments of this application, are described. The computer program may be recorded on, for example, a computer-readable storage medium, loaded onto the data processing apparatus via the computer-readable storage medium, and executed therein.

[0192] This application embodiment uses a data optimization model to determine the mixing information of N elements, so that the N elements can be mixed and processed to generate a target object based on the mixing information of the N elements. Specifically, the data optimization model includes a primary optimization model and N secondary optimization models corresponding to the N elements. That is, the process of using the data optimization model to determine the mixing information of N elements is a process of optimizing multiple objectives. First, each secondary optimization model can be solved according to the first solution constraint to obtain the N residual thresholds that the N elements need to satisfy after mixing. Then, the i-th element among the N elements is traversed, and M second solution constraints are determined based on the residual threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. The second solution constraint of the i-th element... At least one mixed information can be generated by solving the constraint conditions, the historical second solution constraint set, and the current available quantities of Ni untraversed elements. The main optimization model is solved based on the first solution constraint condition, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain M solution results. Then, based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraint corresponding to the target mixed information is added to the historical second solution constraint set as the selected second solution constraint. When all N elements have been traversed, the target mixed information obtained from each traversal is output. As can be seen from the above process, this embodiment transforms multi-objective optimization into single-objective optimization when solving multi-objective optimization problems. Furthermore, the constraint conditions for solving the main optimization model can be reconstructed through the solution results of each sub-objective optimization, ultimately outputting multiple satisfactory solutions, i.e., multiple target mixed information, effectively improving mixing efficiency and reducing costs.

[0193] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a data processing device provided in an embodiment of this application. Figure 7 The data processing device is equipped with a data optimization model, which is used to determine the mixed information of N elements. The data optimization model includes a primary optimization model, N secondary optimization models corresponding to the N elements, and a first solution constraint condition acting on the primary optimization model and the N secondary optimization models. N is an integer greater than 1. Figure 7 The data processing device includes a processor 701, an input interface 702, an output interface 703, and a computer storage medium 704. The processor 701, input interface 702, output interface 703, and computer storage medium 704 can be connected via a bus or other means.

[0194] Computer storage medium 704 can be stored in the memory of a data processing device. The computer storage medium 704 is used to store computer programs, and the processor 701 is used to execute the computer programs stored in the computer storage medium 704. The processor 701 (or CPU (Central Processing Unit)) is the computing and control core of the data processing device, suitable for implementing one or more computer programs, specifically suitable for loading and executing:

[0195] Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element.

[0196] Traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Generate at least one mixed information from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities of the Ni elements that have not been traversed, where M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the selected second solution constraints from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0197] The main optimization model is solved according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint corresponding to the i-th element, and each solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraint;

[0198] Based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added as the selected second solution constraints to the historical second solution constraint set.

[0199] When all N elements have been traversed, output the target mixed information obtained in each traversal.

[0200] This application uses a data optimization model to determine the mixing information of N elements, so that the N elements can be mixed and processed to generate a target object based on the mixing information of the N elements. In specific implementation, the data optimization model includes a primary optimization model and N secondary optimization models corresponding to N elements. That is, the process of determining the mixed information of N elements using the data optimization model is a process of multi-objective optimization. First, each secondary optimization model can be solved according to the first solution constraint to obtain the N residual thresholds that need to be satisfied after the N elements are mixed. Then, the i-th element among the N elements is traversed, and M second solution constraints are determined based on the residual threshold corresponding to the i-th element and the current available quantity of the i-th element. Each second solution constraint corresponding to the i-th element can generate at least one mixed information. The primary optimization model is solved according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain M solution results. Then, based on the M solution results, the target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraint corresponding to the target mixed information is added as the selected second solution constraint to the historical second solution constraint set. When all N elements are traversed, the target mixture information obtained from each traversal is output. As can be seen from the above process, when solving multi-objective optimization problems, the embodiments of this application transform multi-objective optimization into single-objective optimization, and the constraints on the main optimization model can be reconstructed through the solution results of each sub-objective optimization, and finally multiple satisfactory solutions, that is, multiple target mixture information, can be output, which effectively improves the mixing efficiency and reduces the cost.

[0201] This application embodiment also provides a computer storage medium (memory), which is a memory device of a data processing device or server, used to store programs and data. It is understood that the computer storage medium here may include the built-in storage medium of the data processing device or data processing apparatus, or it may include extended storage media supported by the data processing device or data processing apparatus. The computer storage medium provides storage space that stores the operating system of the data processing device or data processing apparatus. Furthermore, the storage space also stores one or more computer programs suitable for loading and execution by the processor 701. It should be noted that the computer storage medium here may be a high-speed RAM memory, or a non-volatile memory, such as at least one disk storage device; optionally, it may also be at least one computer storage medium located remotely from the aforementioned processor.

[0202] In one embodiment, one or more computer programs stored in the computer storage medium can be loaded and executed by the processor 701:

[0203] Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element.

[0204] Traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Generate at least one mixed information from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities of the Ni elements that have not been traversed, where M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the selected second solution constraints from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0205] The main optimization model is solved according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint corresponding to the i-th element, and each solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraint;

[0206] Based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added as the selected second solution constraints to the historical second solution constraint set.

[0207] When all N elements have been traversed, output the target mixed information obtained in each traversal.

[0208] In one implementation, when the computer storage medium determines the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, it performs the following steps: determining the mixed information with the minimum processing resource requirement from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results; and determining the determined mixed information as the target mixed information.

[0209] In one implementation, when the computer storage medium determines the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, it performs the following steps: outputting the mixed information corresponding to the M second solution constraints, and outputting the processing resource amount corresponding to each mixed information based on the M solution results; receiving a first selection operation; and determining the mixed information corresponding to the first selection operation as the target mixed information.

[0210] In one implementation, each of the M second solution constraints corresponding to the i-th element is represented by the remaining quantity interval of the i-th element when mixing the N elements. The remaining quantity interval includes the minimum remaining quantity and the maximum remaining quantity. One or more computer programs stored in the computer storage medium can be loaded by the processor 701 and executed as follows: taking the remaining quantity threshold corresponding to the currently traversed i-th element as the minimum remaining quantity in the t-th remaining quantity interval; determining the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t; adding the minimum remaining quantity in the t-th remaining quantity range and the remaining quantity increment of the t-th remaining quantity interval to obtain the maximum remaining quantity of the t-th remaining quantity interval. Here, t is an integer greater than or equal to 1 and less than M.

[0211] In one implementation, one or more computer programs stored in the computer storage medium can be loaded by processor 701 and executed as follows: subtracting the current available amount of the i-th element from the remaining amount threshold, and multiplying M elements and t; dividing the result of the subtraction operation with the result of the multiplication operation to obtain the remaining amount increment of the t-th remaining amount interval.

[0212] In one implementation, one or more computer programs stored in the computer storage medium can be loaded by processor 701 and executed as follows: determining whether the i-th element is the first of the N elements to be traversed; if so, the historical second solution constraint set is empty, and the main optimization model is solved based on the first solution constraint and the M second solution constraints corresponding to the element to obtain M solution results. If it is determined that the i-th element is not the first of the N elements to be traversed, the historical second solution constraint set before the current traversal is obtained; the main optimization model is solved based on the obtained historical second solution constraint set, the first solution constraint, and the M second solution constraints corresponding to the i-th element to obtain M solution results.

[0213] In one implementation, one or more computer programs stored in the computer storage medium can be loaded by processor 701 and executed as follows: when all N elements are traversed, after outputting the target mixing information obtained from each traversal, a second selection operation is received; the N elements are mixed according to the mixing information in the output target mixing information corresponding to the second selection operation to obtain the target object.

[0214] In one implementation, one or more computer programs stored in the computer storage medium can be loaded by processor 701 and executed with the following steps: determining a primary optimization model based on any one or more of the following: the wet ratio of each of the N elements, the amount of resources required to obtain a unit quantity of each element, and the chemical composition information contained in a unit quantity of target objects; and determining a secondary optimization model corresponding to each element based on the current available amount corresponding to each element and the amount of each element required to generate a target quantity of target objects.

[0215] In one implementation, the first solution constraint includes N wet ratio constraints corresponding to N elements and P chemical component constraints corresponding to P chemical components, where P is an integer greater than or equal to 1; the wet ratio constraint for each element includes the minimum and maximum wet ratio for each element; a chemical component constraint includes the minimum and maximum values ​​of each chemical component contained in the target object generated by mixing N elements.

[0216] This application uses a data optimization model to determine the mixing information of N elements, so that the N elements can be mixed and processed to generate a target object based on the mixing information of the N elements. In its implementation, the data optimization model includes a primary optimization model and N secondary optimization models corresponding to the N elements. That is, the process of determining the mixed information of the N elements using the data optimization model is a multi-objective optimization process. First, each secondary optimization model can be solved according to the first solution constraint to obtain the N residual thresholds that the N elements need to satisfy after mixing. Then, the i-th element among the N elements is traversed, and based on the residual threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element, M secondary solution constraints corresponding to the i-th element are determined. The primary optimization model is solved according to the first solution constraint, the historical set of secondary solution constraints, and the M secondary solution constraints corresponding to the i-th element to obtain M solution results. Then, based on the M solution results, the target mixed information is selected from the mixed information corresponding to the M secondary solution constraints corresponding to the i-th element, and the secondary solution constraints corresponding to the target mixed information are added as selected secondary solution constraints to the historical set of secondary solution constraints. When all N elements have been traversed, the target mixed information obtained in each traversal is output. As can be seen from the above process, when solving multi-objective optimization problems, the embodiments of this application transform multi-objective optimization into single-objective optimization, and the constraints on the main optimization model can be reconstructed through the solution results of each sub-objective optimization, and finally multiple sets of satisfactory solutions can be output, that is, multiple objective mixed information, which effectively improves the mixing efficiency and reduces the cost.

[0217] This application also provides a computer program product or a computer program, wherein the computer program product includes a computer program stored in a computer storage medium; the computer program is loaded and executed by a processor 701.

[0218] Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element.

[0219] Traverse the i-th element among the N elements, and determine M second solution constraints based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element. Generate at least one mixed information from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities of the Ni elements that have not been traversed, where M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the selected second solution constraints from the M second solution constraints corresponding to each of the i-1 elements that have been traversed.

[0220] The main optimization model is solved according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints of the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint of the i-th element, and each solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraint;

[0221] Based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added as the selected second solution constraints to the historical second solution constraint set.

[0222] When all N elements have been traversed, output the target mixed information obtained in each traversal.

[0223] In one implementation, the computer program is loaded by processor 701 and executes the following steps: based on the M solution results, determine the mixed information with the minimum processing resource requirement from the mixed information corresponding to the M second solution constraints corresponding to the i-th element; and determine the determined mixed information as the target mixed information.

[0224] In one implementation, the computer program is loaded by processor 701 and executes the following steps: outputting the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and outputting the processing resource amount corresponding to each mixed information based on the M solution results; receiving a first selection operation; and determining the mixed information corresponding to the first selection operation as the target mixed information.

[0225] In one implementation, each of the M second solution constraints corresponding to the i-th element is represented by the remaining quantity interval of the i-th element when mixing N elements. The remaining quantity interval includes the minimum remaining quantity and the maximum remaining quantity. The computer program is loaded by the processor 701 and executes the following steps: taking the remaining quantity threshold corresponding to the currently traversed i-th element as the minimum remaining quantity in the t-th remaining quantity interval; determining the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t; and adding the minimum remaining quantity in the t-th remaining quantity interval and the remaining quantity increment of the t-th remaining quantity interval to obtain the maximum remaining quantity of the t-th remaining quantity interval. Here, t is an integer greater than or equal to 1 and less than M.

[0226] In one implementation, a computer program is loaded by a processor 701 and executes the following steps: subtracting the current available amount of the i-th element from the remaining amount threshold, and multiplying M elements and t; dividing the result of the subtraction operation by the result of the multiplication operation to obtain the remaining amount increment of the t-th remaining amount interval.

[0227] In one implementation, a computer program is loaded by a processor 701 and executes the following steps: determining whether the i-th element is the first element to be traversed among the N elements; if so, the historical second solution constraint set is empty, and the main optimization model is solved based on the first solution constraint and M second solution constraints to obtain M solution results. If it is determined that the i-th element is not the first element to be traversed among the N elements, the historical second solution constraint set before the current traversal is obtained; the main optimization model is solved based on the obtained historical second solution constraint set, the first solution constraint, and the M second solution constraints corresponding to the i-th element to obtain M solution results.

[0228] In one implementation, the computer program is loaded by the processor 701 and executes the following steps: when all N elements are traversed, the target mixing information obtained from each traversal is output, and then a second selection operation is received; the N elements are mixed according to the mixing information in the output target mixing information corresponding to the second selection operation to obtain the target object.

[0229] In one implementation, a computer program is loaded by a processor 701 and executes the following steps: determining a primary optimization model based on one or more of the following: the wet ratio of each of the N elements, the amount of resources required to obtain a unit quantity of each element, and information on the chemical composition contained in a unit quantity of target objects; and determining a secondary optimization model for each element based on the current available amount of each element and the amount of each element required to generate a target quantity of target objects.

[0230] In one implementation, the computer program is loaded by the processor 701 and executes the following steps: the first solution constraint includes N wet ratio constraints corresponding to N elements and P chemical component constraints corresponding to P chemical components, the wet ratio constraint for each element includes the minimum and maximum wet ratio for each element; a chemical component constraint includes the minimum and maximum values ​​of each chemical component contained in the target object generated by mixing N elements.

[0231] This application uses a data optimization model to determine the mixing information of N elements, so that the N elements can be mixed and processed to generate a target object based on the mixing information of the N elements. In its implementation, the data optimization model includes a primary optimization model and N secondary optimization models corresponding to the N elements. That is, the process of determining the mixed information of the N elements using the data optimization model is a multi-objective optimization process. First, each secondary optimization model can be solved according to the first solution constraint to obtain the N residual thresholds that the N elements need to satisfy after mixing. Then, the i-th element among the N elements is traversed, and based on the residual threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element, M secondary solution constraints corresponding to the i-th element are determined. The primary optimization model is solved according to the first solution constraint, the historical set of secondary solution constraints, and the M secondary solution constraints corresponding to the i-th element to obtain M solution results. Then, based on the M solution results, the target mixed information is selected from the mixed information corresponding to the M secondary solution constraints corresponding to the i-th element, and the secondary solution constraints corresponding to the target mixed information are added as selected secondary solution constraints to the historical set of secondary solution constraints. When all N elements have been traversed, the target mixed information obtained in each traversal is output. As can be seen from the above process, when solving multi-objective optimization problems, the embodiments of this application transform multi-objective optimization into single-objective optimization, and the constraints on the main optimization model can be reconstructed through the solution results of each sub-objective optimization, and finally multiple sets of satisfactory solutions can be output, that is, multiple objective mixed information, which effectively improves the mixing efficiency and reduces the cost.

[0232] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A data processing method, characterized in that, The data processing method is executed by a data processing device, which is equipped with a data optimization model. The data optimization model is used to determine the mixed information of N elements. The data optimization model includes a primary optimization model, N secondary optimization models corresponding to the N elements, and a first solution constraint condition acting on the primary optimization model and the N secondary optimization models. N is an integer greater than 1, and the data processing method includes: Each sub-optimization model is solved according to the first solution constraint to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one residual threshold corresponding to one element. Traverse the i-th element among the N elements, and determine M second solution constraints corresponding to the i-th element based on the remaining threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element; generate at least one mixed information from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available quantities corresponding to the Ni elements that have not been traversed, where M is an integer greater than 1, and i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the selected second solution constraints from the M second solution constraints corresponding to each of the i-1 elements that have been traversed. The main optimization model is solved according to the first solution constraint, the historical second solution constraint set, and the M second solution constraints corresponding to the i-th element to obtain M solution results; each solution result corresponds to a second solution constraint of the i-th element, and each solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solution constraint; Based on the M solution results, target mixed information is selected from the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and the second solution constraints corresponding to the target mixed information are added as the selected second solution constraints to the historical second solution constraint set. When all N elements have been traversed, output the target mixed information obtained in each traversal.

2. The method as described in claim 1, characterized in that, The step of determining the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results includes: Based on the M solution results, the mixed information with the minimum processing resource requirement is determined from the mixed information corresponding to the M second solution constraints corresponding to the i-th element; The identified mixed information is designated as the target mixed information.

3. The method as described in claim 1, characterized in that, The step of determining the target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results includes: Output the mixed information corresponding to the M second solution constraints corresponding to the i-th element, and output the processing resource amount corresponding to each mixed information based on the M solution results; Receive the first selection operation and determine the mixed information corresponding to the first selection operation as the target mixed information.

4. The method as described in claim 1, characterized in that, Each of the M second solution constraints corresponding to the i-th element is represented by the remaining quantity interval of the i-th element when mixing the N elements, and the remaining quantity interval includes the minimum remaining quantity and the maximum remaining quantity; the t-th second solution constraint is determined based on the remaining quantity threshold corresponding to the currently traversed i-th element and the current available quantity of the i-th element, including: The threshold value of the remaining quantity corresponding to the i-th element currently being traversed is taken as the minimum remaining quantity in the t-th remaining quantity interval; t is an integer greater than or equal to 1 and less than M; The remaining quantity increment of the t-th remaining quantity interval is determined based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t. The minimum remaining amount in the t-th remaining amount range and the increment of the remaining amount in the t-th remaining amount interval are added together to obtain the maximum remaining amount in the t-th remaining amount interval.

5. The method as described in claim 4, characterized in that, The determination of the remaining quantity increment of the t-th remaining quantity interval based on the current available quantity of the i-th element, the remaining quantity threshold corresponding to the i-th element, M, and t includes: Subtract the current available quantity of the i-th element from the remaining quantity threshold, and multiply the M elements and t. The remainder increment of the t-th remainder interval is obtained by dividing the result of the subtraction operation by the result of the multiplication operation.

6. The method as described in claim 1, characterized in that, The process of solving the main optimization model based on the first solution constraint and the M second solution constraints corresponding to the i-th element to obtain M solution results includes: Determine whether the i-th element is the first of the N elements to be traversed; If so, the set of historical second solution constraints is empty, and the main optimization model is solved according to the first solution constraints and the M second solution constraints corresponding to the i-th element to obtain M solution results; If the i-th element is not the first one to be traversed among the N elements, then the main optimization model is solved according to the historical second solution constraint set, the first solution constraint, and the M second solution constraints corresponding to the i-th element to obtain M solution results.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: Sort the N elements in descending order of their mixed importance.

8. The method as described in claim 1, characterized in that, After outputting the target mixture information obtained from each traversal when all N elements have been traversed, the method further includes: Receive the second option operation; The target object is obtained by mixing the N elements according to the mixing information corresponding to the second selection operation in the output target mixing information.

9. The method as described in claim 8, characterized in that, The primary optimization model is determined based on one or more of the following: the wet ratio of each of the N elements, the amount of resources required to obtain a unit quantity of each element, and the chemical composition information contained in a unit quantity of target objects. The secondary optimization model corresponding to each element is generated based on the current available amount of each element and the amount of each element required to generate a target quantity of target objects.

10. The method as described in claim 9, characterized in that, The first solution constraint includes N wet ratio constraints corresponding to N elements and P chemical component constraints corresponding to P chemical components, where P is an integer greater than or equal to 1; the wet ratio constraint for each element includes the minimum and maximum wet ratio for each element; the chemical component constraint includes the minimum and maximum values ​​of each chemical component contained in the target object generated by mixing the N elements.

11. A data processing apparatus, characterized in that, The data processing device is configured in the data processing equipment, which is equipped with a data optimization model. The data optimization model is used to determine the mixed information of N elements. The data optimization model includes a primary optimization model, N secondary optimization models corresponding to the N elements, and a first solution constraint condition acting on the primary optimization model and the N secondary optimization models. N is an integer greater than 1, and the data processing device includes: The solving unit is used to solve each sub-optimization model according to the first solving constraint conditions to obtain N residual thresholds that need to be satisfied after N elements are mixed and processed, with one element corresponding to one residual threshold. A determining unit is used to traverse the i-th element among the N elements, determine M second solution constraints based on the remaining amount threshold corresponding to the currently traversed i-th element and the current available amount of the i-th element, and generate at least one mixed information from a second solution constraint corresponding to the i-th element, a historical set of second solution constraints, and the current available amounts corresponding to the Ni elements that have not been traversed; M is an integer greater than 1, i is greater than 1 and less than or equal to N; the historical set of second solution constraints includes the second solution constraints selected from the M second solution constraints corresponding to each of the i-1 elements that have been traversed; The solving unit is further configured to solve the main optimization model according to the first solving constraint, the historical second solving constraint set, and the M second solving constraints corresponding to the i-th element to obtain M solution results; one solution result corresponds to one second solving constraint corresponding to the i-th element, and one solution result includes: the amount of processing resources required to mix the N elements according to the mixing information corresponding to the corresponding second solving constraint; The determining unit is further configured to select target mixed information from the mixed information corresponding to the M second solution constraints corresponding to the i-th element based on the M solution results, and add the second solution constraint corresponding to the target mixed information as the selected second solution constraint to the historical second solution constraint set. The output unit is used to output the target mixed information obtained in each traversal when all N elements have been traversed.

12. A data processing device, characterized in that, include: A processor, suitable for implementing one or more computer programs; and, A computer storage medium storing one or more computer programs, which are loaded by the processor and executed as described in any one of claims 1-10.

13. A computer storage medium, characterized in that, The computer storage medium includes a computer program, which, when executed by a processor, is used to perform the data processing method as described in any one of claims 1-10.

14. A computer program product or computer program, characterized in that, The computer program product includes a computer program stored in a computer storage medium, which, when executed by a processor, is used to implement the data processing method as described in any one of claims 1-10.

Images