A method and apparatus for production planning

Abstract

The application discloses a production plan compiling method and device, and the method comprises the following steps: obtaining candidate material data of a rolling production line and a to-be-produced strip steel, distributing the to-be-produced strip steel to a first to-be-produced set of a main rolling scheduling model or a second to-be-produced set of a transition scheduling model, judging whether the coil data fluctuation of the first to-be-produced set is greater than a set threshold value, when the coil data fluctuation is not greater than the set threshold value, distributing the production material of the first to-be-produced set according to the candidate material data, and generating a first production plan table of the rolling production line in a corresponding roll period; when the coil data fluctuation is greater than the set threshold value, arranging the target coil of the second to-be-produced set to the target position of the first to-be-produced set, making the production arrangement order of the coil transition smooth, and generating a second production plan table of the rolling production line in the corresponding roll period, so that the specification and process parameter fluctuation of adjacent coils are smaller, and the accuracy of the production plan compilation is improved.

Description

Technical Field

[0001] This application relates to the technical field of production planning, and more particularly to a method and apparatus for production planning. Background Technology

[0002] The steel production process includes multiple steps such as ironmaking, steelmaking, continuous casting, hot rolling, and cold rolling. These steps are sequential processes. There are not only logistical and financial balance issues among them, but also energy and time balance issues. The hot rolling process, situated between continuous casting and cold rolling, plays a crucial role in connecting the two processes. The scientific formulation and arrangement of the hot rolling production plan directly affects the production of both continuous casting and cold rolling. Hot rolling production involves multiple processes such as furnace reheating, roughing, finishing, cooling, and coiling, and is characterized by high temperature and high energy consumption.

[0003] Currently, major steel mills primarily use manual methods for scheduling production plans for hot-rolled production lines. Planners need to sift through thousands of materials to select those suitable for scheduling, and then sort them according to the hot-rolling production process rules to obtain a material sequence that meets the on-site production requirements. This process involves a large amount of manual operation, relies heavily on the planner's scheduling experience, and is prone to causing fluctuations in product quality, which is detrimental to the refined management of production plans.

[0004] Hot rolling production scheduling is essentially a multi-objective combinatorial optimization problem. Although some existing scheduling techniques consider multiple objectives, most of the solution algorithms are single-objective optimization algorithms based on weighted methods. The objective weights are often difficult to determine, especially when the order of magnitude of the objectives are inconsistent. This makes it impossible to effectively solve the Pareto optimization problem, which also means that there is currently no mature solution for hot rolling production scheduling algorithms. Therefore, it is of great significance to study the application of multi-objective optimization algorithms in hot rolling production scheduling problems.

[0005] Therefore, improving the accuracy of production planning is a pressing technical problem that needs to be solved. Summary of the Invention

[0006] The present invention provides a method and apparatus for preparing production plans, which can improve the accuracy of production plan preparation.

[0007] The embodiments of the present invention provide the following solutions:

[0008] In a first aspect, embodiments of the present invention provide a method for compiling a production plan, the method comprising:

[0009] Obtain candidate material data and strip steel to be produced for the steel rolling production line;

[0010] Based on the attribute information of the strip steel to be produced, the strip steel to be produced is assigned to the first set of to be produced in the main rolling scheduling model or the second set of to be produced in the transition scheduling model.

[0011] Determine whether the fluctuation of the steel coil data in the first set to be produced is greater than a set threshold.

[0012] If not, then the production materials for the first set to be produced are allocated according to the candidate material data, and the first production plan table for the steel rolling production line in the corresponding roll period is generated.

[0013] If so, the target steel coils of the second set to be produced are arranged to the target positions of the first set to be produced, so that the fluctuation of the steel coil data does not exceed the set threshold, and a second production plan table for the steel rolling production line in the corresponding roll period is generated.

[0014] In one optional embodiment, acquiring candidate material data for the steel rolling production line includes:

[0015] Retrieve material inventory data from the management database;

[0016] The material inventory data with zero or missing information attribute values ​​is cleaned up, and the constrained material data is filtered out to obtain the candidate material data.

[0017] In an optional embodiment, determining whether the fluctuation of the steel coil data in the first set to be produced exceeds a set threshold includes:

[0018] Based on the steel coil data of each coil in the first set to be produced, obtain the width difference, thickness difference, and rolling temperature difference between adjacent steel coils;

[0019] When the width difference is less than the first threshold, the thickness difference is less than the second threshold, and the rolling temperature difference is not greater than the third threshold, it is determined that the fluctuation of the steel coil data is not greater than the set threshold.

[0020] When the width difference is not less than the first threshold, or the thickness difference is not less than the second threshold, or the rolling temperature difference is not greater than the third threshold, the fluctuation of the steel coil data is determined to be greater than the set threshold.

[0021] In an optional embodiment, after determining that the width difference is less than a first threshold, the thickness difference is less than a second threshold, and the rolling temperature difference is not greater than a third threshold, the method further includes:

[0022] Based on the steel coil data of each coil in the first set to be produced, the number of specification reversals and the rolling length of the steel coil within a preset width range are obtained, wherein the number of specification reversals is the number of times the strip specifications change from small to large in the strip rolling sequence.

[0023] When the number of specification bounces is less than the fourth threshold and the rolling length of the steel coil is less than the fifth threshold, it is determined that the fluctuation of the steel coil data is not greater than the set threshold.

[0024] In an optional embodiment, allocating production materials for the first set to be produced based on the candidate material data includes:

[0025] Read the JSON strings of each steel coil in the first set to be produced;

[0026] Based on the target fields of the JSON string, determine the material requirements information for the strip steel to be produced;

[0027] Based on the candidate material data and the material demand information, the material allocation result of the first set to be produced is obtained.

[0028] In one optional embodiment, generating a production schedule for the steel rolling line in the corresponding roll period includes:

[0029] Based on the production materials corresponding to each steel coil in the production set, steel coil material groups are obtained by classifying them according to the same specifications.

[0030] The steel coil material group is input into a preset ant colony algorithm model and the corresponding calculation parameters are set to obtain a production order matrix that conforms to the continuous rolling process sorting rules. Each element in the production order matrix represents a different sorting result of the steel coil material group.

[0031] The production plan is obtained based on the target element in the production order matrix.

[0032] In an optional embodiment, obtaining the production schedule based on the target element in the production order matrix includes:

[0033] Obtain the width bounce count, thickness bounce count, average width transition value, average thickness transition value, and average temperature transition value for each element in the production order matrix;

[0034] According to the formula Obtain the connection cost P for each element, where t1 is the width bounce count and t2 is the thickness bounce count. The average width transition value. The average thickness transition value. The mean temperature transition value is given, where α, β, γ, κ, and λ are preset coefficients, and α > β > γ > κ > λ.

[0035] The matrix element corresponding to the minimum connection cost is determined as the target element;

[0036] The production plan table is generated based on the steel coil sorting results of the target elements.

[0037] Secondly, embodiments of the present invention also provide a production planning apparatus, the apparatus comprising:

[0038] The acquisition module is used to acquire candidate material data and strip steel to be produced for the steel rolling production line;

[0039] The allocation module is used to allocate the strip steel to be produced to the first set of production in the main rolling scheduling model or the second set of production in the transition scheduling model according to the attribute information of the strip steel to be produced.

[0040] The judgment module is used to determine whether the fluctuation of the steel coil data in the first set to be produced is greater than a set threshold.

[0041] The first generation module is used to allocate production materials for the first set to be produced according to the candidate material data when the fluctuation of the steel coil data is not greater than a set threshold, and generate the first production plan table of the steel rolling production line in the corresponding roll period.

[0042] The second generation module is used to arrange the target steel coils of the second production set to the target position of the first production set when the fluctuation of the steel coil data is greater than a set threshold, so that the fluctuation of the steel coil data is not greater than the set threshold, and to generate a second production plan table for the steel rolling production line in the corresponding roll period.

[0043] Thirdly, embodiments of the present invention also provide an electronic device, including a processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, cause the electronic device to perform the steps of any of the methods described in the first aspect.

[0044] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0045] The production planning method and apparatus of the present invention have the following advantages compared with the prior art:

[0046] The production planning method of this invention acquires candidate material data and strip steel to be produced for the rolling mill production line. Based on the attribute information of the strip steel to be produced, it allocates the strip steel to the first production set of the main rolling scheduling model or the second production set of the transition scheduling model. It determines whether the fluctuation of the coil data in the first production set exceeds a set threshold. When the fluctuation of the coil data does not exceed the set threshold, it indicates that the arrangement order of coil production is relatively reasonable. Then, the production materials of the first production set are allocated according to the candidate material data, and a first production plan table for the rolling mill production line in the corresponding roll period is generated. When the fluctuation of the coil data exceeds the set threshold, it indicates that there is a flaw in the arrangement order of coil production. Then, the target coils of the second production set are arranged to the target position of the first production set, so that the production arrangement order of the coils transitions smoothly, and a second production plan table for the rolling mill production line in the corresponding roll period is generated. The production plan no longer relies on manual experience for preparation, making the fluctuation of specifications and process parameters of adjacent coils smaller, thereby improving the accuracy of production plan preparation and helping the rolling mill production line reduce costs and increase efficiency. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 A flowchart illustrating a production planning method provided in an embodiment of the present invention;

[0049] Figure 2 A schematic diagram of the production order matrix provided in an embodiment of the present invention;

[0050] Figure 3 A schematic diagram of node probabilities in the ant colony algorithm model provided in this embodiment of the invention. Figure 1 ;

[0051] Figure 4 A schematic diagram of node probabilities in the ant colony algorithm model provided in this embodiment of the invention. Figure 2 ;

[0052] Figure 5 The taboo representation provided for embodiments of the present invention;

[0053] Figure 6 This is a schematic diagram of a production planning device provided in an embodiment of the present invention. Detailed Implementation

[0054] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the protection scope of the embodiments of the present invention.

[0055] Please see Figure 1 , Figure 1 A flowchart of a production planning method provided in an embodiment of the present invention, the method comprising:

[0056] S11. Obtain candidate material data and strip steel to be produced for the steel rolling production line.

[0057] Specifically, the strip steel to be produced refers to strip steel that has not yet been included in the production plan on the rolling mill production line. Candidate material data represents materials that can be matched with the strip steel to be produced, such as slabs in inventory. Both candidate material data and strip steel to be produced can be represented in a list format or other methods, without specific restrictions here.

[0058] In practical applications, due to the large scale of steel plant production systems, some inventory materials may be required for the production of specific steel coils. Allocating these materials to conventional strip steel awaiting production could lead to misallocation of materials and strip steel. Therefore, in one specific implementation, acquiring candidate material data for the steel rolling production line includes:

[0059] Obtain material inventory data from the management database; clean up material data with zero or missing information attribute values, and filter out constrained material data to obtain candidate material data.

[0060] Specifically, the management database can be an Oracle database on the steel plant's manufacturing management system server. The manufacturing management system is used for the overall material management of the steel plant. After connecting to the interface on the manufacturing management system server, material inventory data can be obtained. If the information attribute value of the material data is zero or missing, it means that the information attribute value of the material data is incomplete or missing, and whether it can match the strip steel to be produced is uncertain. Therefore, it needs to be cleaned. It can be understood that the cleanup method can be to specially mark and classify it, or move it to a specific category so that it does not participate in the material allocation of the strip steel to be produced. Constrained material data means that there is already a corresponding strip steel to be produced. The constraint methods of the material include single material constraints and dynamic additional constraints. After filtering, the material allocation can be prevented from being disordered. Of course, the material inventory data can also be deduplicated according to the material number to obtain accurate candidate material data. After obtaining the candidate material data and the strip steel to be produced, proceed to step S12.

[0061] S12. Based on the attribute information of the strip steel to be produced, allocate the strip steel to the first set of production to be produced in the main rolling scheduling model or the second set of production to be produced in the transition scheduling model.

[0062] Specifically, attribute information can characterize information such as the production volume, specifications, delivery cycle, quality requirements, and coil type of the strip steel to be produced. Based on this, the strip steel to be produced can be allocated to the main rolling mill scheduling model or the transitional scheduling model. The main rolling mill scheduling model is typically used to categorize the production plans of batch strip steel, which has a large production volume, high delivery cycle requirements, and high surface quality requirements. The transitional scheduling model is used to categorize strip steel to be produced with smaller production volumes, special specifications, or lower surface quality requirements. After allocating the strip steel to be produced, each scheduling model can sequentially integrate the strip steel to be produced within its production set, for example, by arranging them in descending order based on the specifications of each coil. After allocating the strip steel to be produced to the production set of the corresponding scheduling model, the process proceeds to step S13.

[0063] S13. Determine whether the fluctuation of the steel coil data in the first set to be produced is greater than the set threshold.

[0064] Specifically, the steel coil data can include the specifications and production process data of the strip steel to be produced. The threshold can be set according to the specific steel coil data. It can be based on the difference between the steel coil data of adjacent steel coils and the set threshold to determine whether the fluctuation of the steel coil data is greater than the set threshold. Alternatively, a data curve can be generated by fitting the steel coil data, and the slope of the data curve can be used to determine whether its fluctuation is greater than the set threshold.

[0065] In one specific implementation, determining whether the fluctuation of the steel coil data in the first set to be produced exceeds a set threshold includes:

[0066] Based on the steel coil data of each coil in the first set to be produced, the width difference, thickness difference, and rolling temperature difference of adjacent steel coils are obtained; when the width difference is less than the first threshold, the thickness difference is less than the second threshold, and the rolling temperature difference is not greater than the third threshold, it is determined that the steel coil data fluctuation is not greater than the set threshold; when the width difference is not less than the first threshold, or the thickness difference is not less than the second threshold, or the rolling temperature difference is not greater than the third threshold, it is determined that the steel coil data fluctuation is greater than the set threshold.

[0067] Specifically, the width difference characterizes the width fluctuation range of adjacent steel coils, the thickness difference characterizes the thickness fluctuation range of adjacent steel coils, and the rolling temperature difference characterizes the rolling temperature fluctuation range of adjacent steel coils. When the width difference is less than the first threshold, the thickness difference is less than the second threshold, and the rolling temperature difference is not greater than the third threshold, it indicates that the specification transition and rolling temperature transition of the adjacent steel coils produced by the rolling mill are reasonable and suitable for production on the rolling mill. In this case, the steel coil data fluctuation is determined to be no greater than the set threshold. Conversely, if any value of the width difference, thickness difference, or rolling temperature difference exceeds its corresponding threshold, the specification transition or rolling temperature transition is not suitable for production on the rolling mill. In this case, the steel coil data fluctuation is determined to be greater than the set threshold. It should be noted that the first, second, and third thresholds can be determined based on the experience of technical personnel or through calibration experiments, and no specific restrictions are imposed here.

[0068] In practical applications, to ensure the overall production efficiency and stability of the steel rolling production line, limiting only the width difference, thickness difference, and rolling temperature difference may adversely affect the production line due to repeated changes in steel coil production specifications. Therefore, in one specific implementation, after determining that the width difference is less than a first threshold, the thickness difference is less than a second threshold, and the rolling temperature difference is not greater than a third threshold, the method further includes:

[0069] Based on the steel coil data of each coil in the first set to be produced, the number of specification bounces and the rolling length of the coil within the preset width range are obtained. The number of specification bounces is the number of times the strip specification changes from small to large in the strip rolling sequence. When the number of specification bounces is less than the fourth threshold and the rolling length of the coil is less than the fifth threshold, it is determined that the fluctuation of the steel coil data is not greater than the set threshold.

[0070] Specifically, the ideal rolling order for each steel coil in the first production set is from wide to narrow and from high temperature to low temperature. However, due to the specification limitations of the strip steel to be produced, specification bounces may occur, meaning that the strip steel specifications increase from small to large in the rolling sequence. The number of specification bounces can include the number of width bounces and the number of thickness bounces. When the number of specification bounces is less than the fourth threshold, it indicates that the number of specification bounces is small, and multiple adjustments to the width limiting device on the rolling line are not required during production, thus not adversely affecting production efficiency. The steel coil rolling length is the length of strip steel continuously rolled within a preset width range. If the steel coil rolling length is not less than the fifth threshold, it indicates that the length of continuously rolled strip steel is large, and the width boundary of the strip steel may cause wear marks on the rolls during the corresponding roll period. When rolling strip steel with a larger width boundary, the wear marks will affect the surface quality of the strip steel. Therefore, when the number of specification bounces is less than the fourth threshold and the steel coil rolling length is less than the fifth threshold, it can be determined that the steel coil data fluctuation is not greater than the set threshold. After determining whether the steel coil data fluctuation is greater than the set threshold, proceed to step S14 or S15.

[0071] S14. If not, allocate the production materials of the first set to be produced according to the candidate material data, and generate the first production plan table of the steel rolling production line in the corresponding roll period.

[0072] Specifically, when the fluctuation of the steel coil data does not exceed a set threshold, production materials are allocated to the strip steel to be produced in the first set of production, and a first production plan is generated to represent the production order of the strip steel to be produced in the first set of production. It should be noted that the first production plan represents the production plan of the strip steel to be produced in one corresponding roll period, where the roll period is the continuous production cycle after the rolling mill line changes rolls.

[0073] In one specific implementation, the production materials for the first set to be produced are allocated based on candidate material data, including:

[0074] Read the JSON strings of each steel coil in the first set to be produced; determine the material requirements information of the strip steel to be produced based on the target fields of the JSON strings; obtain the material allocation results of the first set to be produced based on the candidate material data and the material requirements information.

[0075] Specifically, the name or marking information of the strip steel to be produced can be encoded using JSON strings. These JSON strings represent the attribute information of the strip steel, including production specifications, quality requirements, and coil type. The target fields can be determined through the corresponding key values ​​in the JSON string. Parsing these target fields reveals the material requirements for the strip steel to be produced. For example, the production specifications of the strip steel can be determined through the corresponding target fields, further determining the slab specifications for rolling the strip steel. Matching the candidate material data with the material requirements information yields the material allocation results for each coil of strip steel to be produced. In short, JSON strings are a lightweight data exchange format, easy for computers to encode, read, and parse, making them well-suited for production planning.

[0076] S15. If so, the target steel coils of the second set to be produced are arranged to the target position of the first set to be produced, so that the fluctuation of the steel coil data is not greater than the set threshold, and a second production plan table of the rolling mill in the corresponding roll period is generated.

[0077] Specifically, when the fluctuation in steel coil data exceeds a set threshold, it indicates that the specification or rolling temperature fluctuations of adjacent steel coils in the second production set are too large and cannot be well adapted to continuous production on the steel rolling line. In this case, the target steel coil is moved to the target position in the first production set, ensuring that the specification or rolling temperature fluctuations of adjacent steel coils are suitable for continuous production on the steel rolling line. The target position represents the location in the first production set where the production order is unreasonable. After the steel coil data fluctuation is no greater than the set threshold, a second production plan table for the corresponding rolling period is generated.

[0078] In one specific implementation, generating a production schedule for the steel rolling line in the corresponding roll period includes:

[0079] Based on the production materials corresponding to each steel coil in the production set, steel coil material groups are obtained by classifying them according to the same specifications. The steel coil material groups are then input into a preset ant colony algorithm model and the corresponding calculation parameters are set to obtain a production order matrix that conforms to the continuous rolling process sorting rules. Each element in the production order matrix represents a different sorting result of the steel coil material group. Based on the target element in the production order matrix, a production plan table is obtained.

[0080] Specifically, grouping production materials by specifications allows for continuous rolling of identical strip steel, reducing the impact of frequent specification changes on production efficiency. The ant colony algorithm model is used to attempt to arrange the order of the strip steel to be produced; please refer to [link to relevant documentation]. Figure 2 In the production order matrix, each element represents a sorting result for each group of steel coil materials. Initial values ​​for the ant colony algorithm model can be set to determine the number of ants. Deploying ants to groups of materials of the same specification may require deploying ants to groups of materials of the same specification that meet specific characteristics, thus satisfying the set scheduling rules. Ants start from the current node and crawl towards subsequent nodes. First, candidate crawling nodes are searched in the connectivity cost matrix, and then the probability of crawling to each crawling node is calculated. A schematic diagram of the crawling node probability is shown below. Figure 3 and Figure 4 As shown, repeat the above process until no candidate crawling nodes can be found, then use virtual nodes to conclude; all nodes visited by the ant are recorded in the taboo list, and the taboo represents the intent as follows. Figure 5 As shown, the tabu list stores all the nodes visited by each ant, i.e., the path. A path must end with a dummy node; "-1" means that the node position is invalid. For example, ant #1 generates a path that includes node 3, node 5, node 7, node 9, and the dummy node, resulting in a total of 5 valid nodes.

[0081] Each element in the production sequence matrix includes comparing the received candidate material ID sequence with historical records, removing candidate materials that appear in the historical records; setting head and tail candidate materials, which can come from the tail material of the previous main rolling mill model, the head material of the subsequent main rolling mill model, or selected according to specific criteria; generating a list of same-specification groups based on whether the specifications are the same, and modifying the list of same-specification material groups to meet constraints such as the total number of kilometers of the same width. Since there are many strips of the same or similar width produced continuously after being categorized by specification, the width boundary of the strip causes wear marks on the rolls. Therefore, after categorizing by specification, the length of the strip to be produced corresponding to that specification can be read to determine if it exceeds the set length limit; if so, it is segmented so that the length of the segmented strip is less than the length limit. For example, if the number of members of the same specification in a steel coil material group is 7, and the total rolling kilometers exceed the limit, then the number of members in this steel coil material group is divided into 3, 3, and 1 to ensure that the total rolling kilometers do not exceed the limit and are as close as possible to the set upper limit, thereby reducing the frequency of strip specification changes. Elements in the production sequence matrix that conform to the production specifications of the steel rolling line are identified as target elements, and a production plan is obtained based on the strip production sequence corresponding to the target elements.

[0082] In one specific implementation, a production plan is obtained based on the target elements in the production sequence matrix, including:

[0083] Obtain the width bounce count, thickness bounce count, average width transition value, average thickness transition value, and average temperature transition value for each element in the production order matrix; according to the formula... Obtain the connection cost P for each element, where t1 is the number of width bounces and t2 is the number of thickness bounces. The average width transition value. The average thickness transition value. The mean value for temperature transition is given. α, β, γ, κ, and λ are all preset coefficients, where α > β > γ > κ > λ.

[0084] The matrix element corresponding to the minimum connection cost is determined as the target element; a production plan table is generated based on the steel coil sorting result of the target element.

[0085] Specifically, each element in the production order matrix can represent the evaluation result of each connection cost. Each element of the matrix is ​​the connection result from the i-th same specification material group to the j-th same specification material group. The connection cost of each element can be calculated using the above formula, and the matrix element corresponding to the minimum connection cost is determined as the target element to generate the production plan.

[0086] It should be noted that a heuristic backtracking algorithm can also be set to control the parameter values ​​of the ant colony algorithm model; the heuristic backtracking function searches for feasible solutions in the transition segment; the search starts from the root node of the backtracking solution space tree and proceeds in a depth-first manner; the search process is guided by forward conditions, backtracking conditions, success conditions, and failure conditions, and ends only when either the success condition or the failure condition is met; the forward condition is the existence of unexplored child nodes; the backtracking condition is that all child nodes have been explored and no feasible solution has been found; the success condition is that the child nodes contain the endpoint node and do not violate any constraints; the failure condition is that the root node has been backtracked to, all child nodes of the root node have been explored, and no feasible solution has been found; after completion, the success or failure of the transition segment scheduling attempt and the scheduling results are saved to the corresponding location.

[0087] Connection cost P = f(t1 is the number of width bounces, t2 is the number of thickness bounces), The average width transition value. The average thickness transition value. (This is the average value for temperature transition);

[0088] Heuristic factor correction M = pheromone increase intensity coefficient c1 / connection cost P;

[0089] Pheromon concentration N = (1 - pheromone evaporation coefficient c2) × pheromone concentration θ + heuristic factor correction M.

[0090] The production planning method provided in this application will be fully described below with reference to a specific embodiment. Taking the production planning of a hot-rolled 1580 tinplate rolling mill as an example, please refer to Table 1 for the various parameters of this production line.

[0091] Table 1

[0092]

[0093]

[0094] The production planning for this steel rolling production line includes steps one through five.

[0095] Step 1: Extract candidate material data from the material inventory based on production line information. Connect to the Oracle database on the manufacturing management system server via a database driver to obtain real-time material data from the material inventory table view. This data includes attributes such as material length, width, thickness, weight, tapping temperature, material number, grade, tapping mark, surface sorting degree, contract number, warehouse area number, warehousing mark, material status, production time, delivery date, subsequent process code, key orders, difficult orders, urgent markings, and overdue markings. Clean up material data containing zero or missing values ​​in attributes such as material length, width, thickness, weight, and tapping temperature, and remove duplicates based on the material number. Then, filter the data according to dynamic additional constraints such as warehouse area restrictions, logistics status, process restrictions, material status, specification restrictions, and restrictions on individual materials and various material information, to obtain candidate material data that meets the requirements of the intelligent algorithm.

[0096] Step two involves generating serialized material data and constraint rule data lists based on scheduling rules. According to the main rolling scheduling model established for the 1580 production line, candidate material data is filtered, calculated, and serialized. This allows for the segmentation of candidate material data into main rolling section and transition section data. Information such as width jumps, thickness jumps, and tapping temperature jumps of the strip to be produced is obtained. Additionally, rolling period sequence constraint scheduling rule data, including adjacent material constraints, same-roll period mileage limits, roll period material block limits, same-width mileage limits, attribute bounce count limits, and surface sorting degree segmentation limits, is also acquired. This data is processed and serialized to obtain processed candidate material data. Finally, the candidate material data is integrated with the scheduling rules to obtain interface JSON data.

[0097] Step 3: Receive the model name (key "model") and the candidate material information list (key "stock") from the JSON format string, and parse out the material ID, width, thickness, temperature, rolling length, and main rolling section number for each candidate material. Decompose the candidate material information list into multiple individual candidate material information; iterate through all candidate material information, extracting the main rolling section or transition section number to which the candidate material belongs; if it belongs to the current main rolling section or transition section, receive the scheduling rule-related information for that candidate material; otherwise, do not receive it. Specifically, four main rolling sections need to be scheduled. If the first main rolling section is currently being scheduled, and its main rolling section number is "1000", then the material is received as a candidate material for the first main rolling section.

[0098] Receive a list of scheduling rules from a JSON string, including searching and reading width transition rules by key "MS_WT" (WT means width transition), searching and reading thickness transition rules by key "MS_ThT" (ThT means thickness transition), searching and reading temperature transition rules by key "MS_TeT" (TeT means temperature transition), searching and reading same-width rolling length limit rules by key "SW_RL_LIMIT", searching and reading width bounce count limit rules by key "WT_NJ_C_LIMIT", searching and reading thickness bounce count limit rules by key "ThT_NJ_C_LIMIT", and searching and reading the first material specification standard rule for the transition segment by key "TS_HS".

[0099] Step 4: Call the corresponding scheduling calculation function to generate the production plan.

[0100] The maximum number of rolling units is set according to the model name. The main rolling scheduling model for tinplate is a batch production model, and the maximum number of rolling units is set to 3. Based on the main rolling scheduling model, the main rolling section is scheduled. The received candidate material IDs are compared with historical records, and candidate material information appearing in the historical records is removed. A list of material groups with the same specification is generated according to whether the specifications are the same. Materials with the same width, thickness, and temperature value belong to the same steel coil material group. The list of steel coil material groups with the same specification is modified to meet constraints such as the total number of kilometers for the same width. For material groups whose total rolling kilometers exceed the maximum value of the total rolling kilometers for the same width, they are split into groups with fewer members, so that each split group meets the total number of kilometers for the same width.

[0101] A production order matrix (or connection cost assessment matrix) is generated based on the revised steel coil material groups. Each element of this matrix represents the connection order from one material group of the same specification to another. The connection results are evaluated, and the evaluation items include width transition assessment, width transition range, thickness transition assessment, thickness transition range, temperature transition assessment, and temperature transition range. The calculation of width transition assessment, thickness transition assessment, and temperature transition assessment is as follows: Based on the width, thickness, and temperature values ​​of the materials following in the sequence, the corresponding open interval position or fixed scattered position is calculated. After determining the position, the range of width, thickness, and temperature transition values ​​for that position is read. If the difference between the width, thickness, and temperature of the materials following in the sequence and the width, thickness, and temperature of the materials preceding in the sequence is within the transition value range, then the width, thickness, and temperature transition conforms to the transition rules.

[0102] Add virtual nodes to the production order matrix, and simultaneously add connection evaluations between virtual and non-virtual nodes. The evaluation of connections from non-virtual nodes to virtual nodes is set to valid, and the evaluation of connections from virtual nodes to non-virtual nodes is also set to valid. Set initial values ​​for the ant colony algorithm control parameters. The total number of ants can be set to 3 to 5 times the total number of material groups of the same specification, and the number of iteration rounds can be set to 5-10 rounds. Use the ant colony algorithm to search for candidate main rolling section scheduling results that meet the minimum criteria. Deploy ants to material groups of the same specification, with 3 to 5 ants deployed on each material group.

[0103] An ant starts from the current node and crawls towards the next node. First, candidate crawling nodes are searched in the connectivity cost matrix. Then, the probability of crawling to each crawling node is calculated. Finally, a roulette wheel algorithm is used to select a crawling node. The roulette wheel algorithm selects a locally optimal crawling node with a high probability, but sometimes it may select a locally suboptimal crawling node.

[0104] The calculation process of the input roulette wheel betting algorithm is as follows: α is 10000000000, β is 100000000, γ is 1000000, κ is 1000, and λ is 1. Setting these coefficients maximizes the penalty for width bounce and minimizes the penalty for temperature transition amplitude. The probability value is equal to the reciprocal of the connection cost value; the heavier the connection cost penalty, the lower the probability value of the crawling node, thus distinguishing between local optima and poor solutions. Furthermore, the roulette wheel algorithm is implemented to select locally optimal crawling nodes with a high probability, but sometimes it will also select locally poor crawling nodes, avoiding getting trapped in local optima and missing the global optimum.

[0105] Repeat the above process until the ants can no longer find any candidate crawling nodes, then terminate with virtual nodes. All nodes visited by the ants are recorded in a taboo list for use in subsequent steps. Filter all candidate main rolling section scheduling results according to pass-through scheduling rules, keeping only those that do not violate these rules. Pass-through rules can be such as the total number of strip steel materials to be produced not being less than the minimum, the total number of rolling kilometers of strip steel to be produced not being less than the minimum, or the total number of width bounces not exceeding the maximum, or the total number of thickness bounces not exceeding the maximum.

[0106] The scheduling results of the first production set that passed the previous round of screening are selected using the superior scheduling rules, and candidate results that do not violate these rules are retained. Superior scheduling rules could include maximizing the total number of materials or the total number of rolling kilometers. All candidate scheduling results that passed the previous round of screening are then selected using the optimal scheduling rules, and the final scheduling result is retained. A selection-type scheduling rule could be that the larger the total number of rolling kilometers, the better; this result is then saved, becoming the production plan for the first production set.

[0107] This iteration is about to end. Before ending, update control variables such as pheromone and heuristic factor corrections, and reward better candidate production plans. If there is another iteration, execute the ant crawling backward from the starting position again; otherwise, end the iteration. Save the scheduling success status and scheduling results (candidate material ID sequence) to the appropriate location for use in subsequent steps.

[0108] It should be noted that the production plan compilation method for the strip steel to be produced in the main rolling mill scheduling model is also applicable to the transition scheduling model. If the production plans for both the first and second sets of strip steel to be produced are successfully compiled, the scheduling attempt for the next rolling unit continues. Otherwise, step four ends.

[0109] Step 5: Determine whether the scheduling is successful based on the scheduling results of the main rolling mill scheduling model and the transition scheduling model, and return the corresponding scheduling results.

[0110] Based on the number of successful scheduling attempts from both the main rolling mill scheduling model and the transition scheduling model, the total number of rolling units to be output is determined. The minimum of the two success counts is taken as the total number of rolling units to be output. For example, if the number of successful main rolling mill scheduling attempts is 3 and the number of successful transition scheduling attempts is 2, it means that the first transition section and the first main rolling mill section have been scheduled, as have the second transition section and the second main rolling mill section. However, after scheduling the third main rolling mill section, the third transition section fails because the candidate scheduling result does not meet the scheduling rules. In this case, the total number of rolling units to be output is 2, that is, the first transition section, the first main rolling mill section, the second transition section, and the second main rolling mill section are output in the order from front to back. Complete material data is filtered from the inventory data according to the material ID number order, and the complete material data is saved to the local database. The background pushes the results to the interface for display.

[0111] Based on the same inventive concept as the production planning method, embodiments of the present invention also provide a production planning apparatus. Please refer to [link to relevant documentation]. Figure 6 The device includes:

[0112] The acquisition module 601 is used to acquire candidate material data and strip steel to be produced in the steel rolling production line;

[0113] The allocation module 602 is used to allocate the strip steel to be produced to the first set of production in the main rolling schedule model or the second set of production in the transition schedule model according to the attribute information of the strip steel to be produced.

[0114] The judgment module 603 is used to determine whether the fluctuation of the steel coil data in the first set to be produced is greater than a set threshold.

[0115] The first generation module 604 is used to allocate production materials for the first set to be produced according to the candidate material data when the fluctuation of the steel coil data is not greater than a set threshold, and generate the first production plan table of the steel rolling production line in the corresponding roll period.

[0116] The second generation module 605 is used to arrange the target steel coils of the second production set to the target position of the first production set when the fluctuation of the steel coil data is greater than a set threshold, so that the fluctuation of the steel coil data is not greater than the set threshold, and to generate a second production plan table of the rolling mill in the corresponding roll period.

[0117] In one optional embodiment, the acquisition module includes:

[0118] The acquisition submodule is used to acquire material inventory data from the management database;

[0119] The first acquisition submodule is used to clean up material data in the material inventory data that has information attribute values ​​of zero or missing, and to filter out constrained material data to obtain the candidate material data.

[0120] In one optional embodiment, the determining module includes:

[0121] The second acquisition submodule is used to obtain the width difference, thickness difference and rolling temperature difference between adjacent steel coils based on the steel coil data of each steel coil in the first set to be produced.

[0122] The first determining submodule is used to determine that the fluctuation of the steel coil data is not greater than a set threshold when the width difference is less than a first threshold, the thickness difference is less than a second threshold, and the rolling temperature difference is not greater than a third threshold.

[0123] The second determining submodule is used to determine that the fluctuation of the steel coil data is greater than a set threshold when the width difference is not less than the first threshold, or the thickness difference is not less than the second threshold, or the rolling temperature difference is not greater than the third threshold.

[0124] In an optional embodiment, the determining module further includes:

[0125] The third submodule is used to obtain the number of specification reversals and the rolling length of the steel coil within a preset width range based on the steel coil data of each steel coil in the first set to be produced. The number of specification reversals is the number of times the strip specifications change from small to large in the strip rolling sequence.

[0126] The third determining submodule is used to determine that the fluctuation of the steel coil data is not greater than a set threshold when the number of specification bounces is less than a fourth threshold and the rolling length of the steel coil is less than a fifth threshold.

[0127] In one optional embodiment, the first generation module includes:

[0128] The reading submodule is used to read the JSON strings of each steel coil in the first set to be produced;

[0129] The fourth determination submodule is used to determine the material requirements information of the strip steel to be produced based on the target field of the JSON string;

[0130] The fourth obtaining submodule is used to obtain the material allocation result of the first set to be produced based on the candidate material data and the material demand information.

[0131] In one optional embodiment, the first generation module or the second generation module includes:

[0132] The fifth submodule is used to classify steel coils into groups of the same specifications based on the production materials corresponding to each steel coil in the production set.

[0133] The sixth submodule is used to input the steel coil material group into a preset ant colony algorithm model and set the corresponding calculation parameters to obtain a production order matrix that conforms to the continuous rolling process sorting rules, wherein each element in the production order matrix represents a different sorting result of the steel coil material group.

[0134] The seventh submodule is used to obtain the production plan table based on the target element in the production order matrix.

[0135] In one optional embodiment, the seventh obtaining submodule includes:

[0136] The acquisition unit is used to acquire the width bounce count, thickness bounce count, width transition mean, thickness transition mean, and temperature transition mean for each element in the production order matrix.

[0137] Obtaining a unit, used according to the formula Obtain the connection cost P for each element, where t1 is the width bounce count and t2 is the thickness bounce count. The average width transition value. The average thickness transition value. The mean temperature transition value is given, where α, β, γ, κ, and λ are preset coefficients, and α > β > γ > κ > λ.

[0138] A determining unit is configured to determine the matrix element corresponding to the minimum connection cost as the target element;

[0139] The generation unit is used to generate the production plan table based on the steel coil sorting result of the target element.

[0140] Based on the same inventive concept as the programming method, embodiments of the present invention also provide an electronic device, including a processor and a memory, wherein the memory is coupled to the processor and stores instructions that, when executed by the processor, cause the electronic device to perform the steps of any of the programming methods.

[0141] Based on the same inventive concept as the compilation method, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the compilation methods.

[0142] The technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0143] By acquiring candidate material data and strip steel to be produced for the rolling mill production line, and based on the attribute information of the strip steel to be produced, the strip steel to be produced is allocated to the first production set of the main rolling scheduling model or the second production set of the transition scheduling model. It is determined whether the fluctuation of the coil data in the first production set exceeds a set threshold. When the fluctuation of the coil data does not exceed the set threshold, it indicates that the arrangement order of coil production is relatively reasonable. Then, the production materials of the first production set are allocated according to the candidate material data, and the first production plan table of the rolling mill production line in the corresponding roll period is generated. When the fluctuation of the coil data exceeds the set threshold, it indicates that there is a flaw in the arrangement order of coil production. Then, the target coils of the second production set are arranged to the target position of the first production set, so that the production arrangement order of the coils is smoothly transitioned, and the second production plan table of the rolling mill production line in the corresponding roll period is generated. The production plan no longer relies on manual experience for preparation, making the fluctuation of specifications and process parameters of adjacent coils smaller, thereby improving the accuracy of production plan preparation and helping the rolling mill production line to reduce costs and increase efficiency.

[0144] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0145] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (modules, systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0146] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0147] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0148] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0149] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for preparing a production plan, characterized in that, The method includes: Obtain candidate material data and strip steel to be produced for the steel rolling production line; Based on the attribute information of the strip steel to be produced, the strip steel to be produced is assigned to the first set of to be produced in the main rolling scheduling model or the second set of to be produced in the transition scheduling model. Determine whether the fluctuation of the steel coil data in the first set to be produced is greater than a set threshold. If not, then the production materials for the first set to be produced are allocated according to the candidate material data, and the first production plan table for the steel rolling production line in the corresponding roll period is generated. If so, the target steel coils of the second set to be produced are arranged to the target positions of the first set to be produced, so that the fluctuation of the steel coil data is not greater than the set threshold, and a second production plan table for the steel rolling production line in the corresponding roll period is generated. The step of determining whether the fluctuation of the steel coil data in the first set to be produced exceeds a set threshold includes: Based on the steel coil data of each coil in the first set to be produced, obtain the width difference, thickness difference, and rolling temperature difference between adjacent steel coils; When the width difference is less than the first threshold, the thickness difference is less than the second threshold, and the rolling temperature difference is not greater than the third threshold, it is determined that the fluctuation of the steel coil data is not greater than the set threshold. When the width difference is not less than the first threshold, or the thickness difference is not less than the second threshold, or the rolling temperature difference is not greater than the third threshold, it is determined that the fluctuation of the steel coil data is greater than the set threshold. Generate the production schedule for the corresponding roll period of the steel rolling production line, including: Based on the production materials corresponding to each steel coil in the production set, steel coil material groups are obtained by classifying them according to the same specifications. The steel coil material group is input into a preset ant colony algorithm model and the corresponding calculation parameters are set to obtain a production order matrix that conforms to the continuous rolling process sorting rules. Each element in the production order matrix represents a different sorting result of the steel coil material group. The production plan table is obtained based on the target elements in the production order matrix; The step of obtaining the production plan table based on the target element in the production order matrix includes: Obtain the width bounce count, thickness bounce count, average width transition value, average thickness transition value, and average temperature transition value for each element in the production order matrix; According to the formula Obtain the connection cost of each element. P ,in, t 1 represents the number of width bounces. t 2 represents the number of thickness bounces. The average width transition value. The average thickness transition value. The average temperature transition value, , , , and All are preset coefficients. > > > > ; The matrix element corresponding to the minimum connection cost is determined as the target element; The production plan table is generated based on the steel coil sorting results of the target elements.

2. The method for preparing a production plan according to claim 1, characterized in that, The acquisition of candidate material data for the steel rolling production line includes: Retrieve material inventory data from the management database; The material inventory data with zero or missing information attribute values ​​is cleaned up, and the constrained material data is filtered out to obtain the candidate material data.

3. The method for preparing a production plan according to claim 1, characterized in that, After determining that the width difference is less than a first threshold, the thickness difference is less than a second threshold, and the rolling temperature difference is not greater than a third threshold, the method further includes: Based on the steel coil data of each coil in the first set to be produced, the number of specification reversals and the rolling length of the steel coil within a preset width range are obtained, wherein the number of specification reversals is the number of times the strip specifications change from small to large in the strip rolling sequence. When the number of specification bounces is less than the fourth threshold and the rolling length of the steel coil is less than the fifth threshold, it is determined that the fluctuation of the steel coil data is not greater than the set threshold.

4. The method for preparing a production plan according to claim 1, characterized in that, The step of allocating production materials for the first set to be produced based on the candidate material data includes: Read the JSON strings of each steel coil in the first set to be produced; Based on the target fields of the JSON string, determine the material requirements information for the strip steel to be produced; Based on the candidate material data and the material demand information, the material allocation result of the first set to be produced is obtained.

5. A production planning apparatus, characterized in that, The device includes: The acquisition module is used to acquire candidate material data and strip steel to be produced for the steel rolling production line; The allocation module is used to allocate the strip steel to be produced to the first set of production in the main rolling scheduling model or the second set of production in the transition scheduling model according to the attribute information of the strip steel to be produced. The judgment module is used to determine whether the fluctuation of the steel coil data in the first set to be produced is greater than a set threshold. The first generation module is used to allocate production materials for the first set to be produced according to the candidate material data when the fluctuation of the steel coil data is not greater than a set threshold, and generate the first production plan table of the steel rolling production line in the corresponding roll period. The second generation module is used to arrange the target steel coils of the second production set to the target position of the first production set when the fluctuation of the steel coil data is greater than the set threshold, so that the fluctuation of the steel coil data is not greater than the set threshold, and to generate the second production plan table of the steel rolling production line in the corresponding roll period. The judgment module includes: The second acquisition submodule is used to obtain the width difference, thickness difference and rolling temperature difference between adjacent steel coils based on the steel coil data of each steel coil in the first set to be produced. The first determining submodule is used to determine that the fluctuation of the steel coil data is not greater than a set threshold when the width difference is less than a first threshold, the thickness difference is less than a second threshold, and the rolling temperature difference is not greater than a third threshold. The second determining submodule is used to determine that the fluctuation of the steel coil data is greater than a set threshold when the width difference is not less than the first threshold, or the thickness difference is not less than the second threshold, or the rolling temperature difference is not greater than the third threshold. The first generation module or the second generation module includes: The fifth submodule is used to classify steel coils into groups of the same specifications based on the production materials corresponding to each steel coil in the production set. The sixth submodule is used to input the steel coil material group into a preset ant colony algorithm model and set the corresponding calculation parameters to obtain a production order matrix that conforms to the continuous rolling process sorting rules, wherein each element in the production order matrix represents a different sorting result of the steel coil material group. The seventh submodule is used to obtain the production plan table based on the target element in the production order matrix; The seventh obtaining submodule includes: Obtaining unit, used according to formula Obtain the connection cost of each element. P ,in, t 1 represents the number of width bounces. t 2 represents the number of thickness bounces. The average width transition value. The average thickness transition value. This is the average temperature transition value. , , , and All are preset coefficients. > > > > ; A determining unit is configured to determine the matrix element corresponding to the minimum connection cost as the target element; The generation unit is used to generate the production plan table based on the steel coil sorting result of the target element.

6. An electronic device, characterized in that, It includes a processor and a memory, the memory being coupled to the processor, the memory storing instructions that, when executed by the processor, cause the electronic device to perform the steps of the method of any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method described in any one of claims 1-4.

Images