A Multi-Objective Design Method for Medium-Thick Plate Billets

Abstract

This invention provides a multi-objective design method for medium-thick plate billets, relating to the field of automation technology. The invention extracts the design process characteristics of medium-thick plate billets, transforms production objectives and process constraints into mathematical expressions, and establishes a multi-objective system optimization model for the design process of medium-thick plate billets. Orders are grouped according to thickness and steel quality codes. Heuristic algorithms are designed to generate initial medium-thick plate billet design schemes. Different heuristic algorithms are constructed according to objective priorities to cover the generation process of different initial feasible billet design schemes. A billet sub-plate crossover strategy and a billet sub-plate mutation strategy are designed to improve the initial feasible design schemes through crossover and mutation processes. A design scheme selection strategy selects the Pareto optimal scheme in the calculation process to form a scheme pool. A scheme reduction strategy is used to reduce the number of schemes in the scheme pool and adjust the sub-plate distribution on different medium-thick plates within the same design scheme. Finally, a design scheme within a limited number of schemes is given.

Description

Technical Field

[0001] This invention relates to the field of automation technology, and in particular to a multi-objective design method for medium-thick plate blanks. Background Technology

[0002] Billet design for medium and heavy plates is a key technology for improving production efficiency in hot rolling production of steel enterprises. The quality of the billet design plan has a significant impact on key production metrics such as production carryover, yield, and billet weight. Medium and heavy plates have numerous specifications and complex order types. Currently, billet design for medium and heavy plates in steel enterprises is mostly based on manual experience or an automated design process. However, manual operation requires planners to complete the assembly and billet design for hundreds of orders, resulting in extremely low efficiency and problems such as continuously increasing spot production rates and low yields. Automated design processes cannot address the various production constraints and finished product standard requirements during the billet design process, nor can they guarantee the superiority of the design scheme. Often, after automated design, manual adjustments are still required, which cannot effectively improve production efficiency. Therefore, researching and developing an intelligent assembly system that meets the needs of medium and heavy plate production, establishing relevant assembly and optimization models, and solving problems such as low production efficiency and high spot production rates are urgent tasks for the medium and heavy plate production process of steel enterprises.

[0003] In actual production, the diverse order demands of customers are difficult to reconcile with the large-scale production plans of steel enterprises, necessitating a scheduling method to manage different orders. This involves combining given customer order demands into different batches, with orders within the same batch sharing similar process characteristics, thus forming a design scheme. Multiple design schemes then form a plate assembly plan—this is the billet design problem, the core of which is the decision of the optimal combination. Medium-thick plate billet design refers to the billet design problem for the medium-thick plate production process. Besides billet design, master plate design is also a crucial step. Generally, master plate design requires decisions on the actual production specifications of the master plate and the cutting method of the order's sub-plates on the master plate, while billet design involves designing suitable billet cross-sectional specifications and lengths for the planned master plate. Since the master plate is rolled from the billet, in this paper, we collectively refer to these two processes as medium-thick plate billet design.

[0004] Research on the optimization design and utilization of medium and heavy plate billets can effectively improve production planning capabilities. Facing the changing characteristics of customer orders—personalized, small-batch, and fragmented—it enables rapid response and feedback to customer needs, meeting customized requirements. Through optimized design methods and the application of efficient algorithms, inventory billets can be effectively utilized, while billet design surplus can be effectively controlled, resulting in a significant improvement in yield. The optimized design results of medium and heavy plate billets form the basis for subsequent rolling and steelmaking batch planning. The quality of billet design also determines the raw material input and energy output of subsequent processes, which has a positive impact on enterprise production costs and environmental pollution control.

[0005] In addition to the design problem of medium and heavy plate billets, due to the diversified requirements of billet design objectives, automatic decision-making technology for single-objective optimization can no longer meet the actual production needs. It is necessary to calculate algorithms for multiple schemes based on multiple objectives to meet the requirements.

[0006] The design problem of medium and heavy plate billets can be categorized as a bin packing problem in mathematical optimization. Zhang Q, Liu S, and others established an integer programming model to study this problem in "Column generation algorithms for mother plate design in steel plants" and designed a corresponding solution method based on the column generation algorithm. However, the problems envisioned in the above research are relatively ideal and not suitable for actual production environments. In the field of related adaptability research, some researchers have successively conducted research on practical problems and related variations. Zheng Zhong et al., in patents such as "A collaborative design method and system for medium and heavy plate mother plates and billets oriented towards production order combination optimization" and "Hot-rolled medium and heavy plates considering the flexibility of non-fixed-length order specifications," designed automated methods for billet design problems with different characteristics, achieving certain performance improvements. Li Ziqiang et al., in patents "Intelligent steel billet assembly method and device" and "Method and device for medium and heavy plate billet design of adaptive equipment in steel enterprises," respectively designed an intelligent billet assembly method and a medium and heavy plate billet design method that can meet multiple rule constraints and cutting equipment requirements, improving the quality of billet assembly. However, these methods cannot meet the requirements of enterprises to simultaneously consider multiple design objectives to complete billet design. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention provides a multi-objective design method for medium-thick plate blanks.

[0008] A multi-objective-based design method for medium-thick plate billets includes the following steps:

[0009] Step 1: Extract the design process characteristics of medium and heavy plate billets, transform the production objectives and process constraints into mathematical expressions, and thus establish a multi-objective function and constraints for the design process of medium and heavy plate billets;

[0010] The design process features of the medium and heavy plate billet include: the matching relationship between billet and order, the limitations of production equipment, the production process from billet to order, customer needs, and the quality measurement indicators of production plan.

[0011] Step 1.1: Define the decision variable x in the design process of medium and heavy plate billets. ij x ij This represents the design quantity of order i on billet j, and its value range is x. ij >0, It is a set of integers;

[0012] Step 1.2: Define the optimization objectives in the design process of medium and heavy plate billets;

[0013] The optimization objectives include: minimizing the quantity of spot goods taken out, maximizing the yield rate, and maximizing the total weight of billets.

[0014] The spot goods are products whose production plan exceeds the order quantity in the design of medium and heavy plate billets, i.e. p i This refers to the number of items ordered in order i;

[0015] The yield rate is a standard for evaluating the production efficiency of a production plan designed for medium and heavy plate blanks. It represents the ratio of finished product quantity to raw material weight. Finished product quantity refers to the actual weight of sub-plates that can be cut, including both futures and spot sub-plate weights. The yield rate is expressed as Where q i It is the unit weight of order i, d i It is the length of order i.

[0016] The weight of the sub-billet is the designed raw material weight for medium-thick plate billets, i.e., the weight of the billet itself. The weight of the j-th billet is calculated based on its thickness, width, length, and steel density ρ. The designed thickness of billet j and the order thickness T are also considered. j Equivalently, the design width is the width w of the main order k. k and the amount of cutting edge S j The sum of the design length of the billet is the total length of the order plus the amount of the cut-off ends, H. j The sum of these two quantities makes the weight of the billet equal to the weight of the billet. The total weight of the raw materials is:

[0017] Step 1.3: Analyze the production process characteristics of medium and heavy plate billet design, and describe the mathematical expression of its constraints based on the defined decision variables, specifically including order constraints, production process constraints, and unique master order constraints.

[0018] The order constraint is that during the production of medium and heavy plates, the designed billet is rolled by a rolling mill, and the volume of the billet changes after rolling. Due to limitations in production technology and equipment, the thickness range T of the medium and heavy plates is limited. min and T max Width range W min and W max Therefore, there is T. min ≤T j ≤T max and W min ≤W j ≤W max .

[0019] The manufacturing process constraints include: each design scheme has at most one master order, i.e. z kkj This indicates whether order k is designed as a master order onto billet j; if so, it is 1, otherwise 0. In the design scheme for medium-thick plate billets, the width jump of sub-plates cannot exceed the upper limit. Width jump is a parameter affecting the yield of medium-thick plates, referring to the difference between the maximum and minimum width of all sub-plates in a design scheme. Therefore, |min{1,x uj}w u -min{1,x vj}w v |≤W c Where order u and order v are orders with different production order numbers, W c For the maximum jump width, w u It is the width of order u, w v This refers to the width of order v. The length of the design scheme cannot exceed the limit length required by the design of medium-thick plate blanks, mathematically expressed as... Where D j d is the limit length value. i The length of order i is given, and the yield rate of the design scheme must not be less than the lower limit of yield rate Y. MIN The mathematical expression is:

[0020]

[0021] The unique master order constraint means that each design scheme can only have one master order. Where M is a maximal number.

[0022] Step 2: Group the orders according to thickness and steel code; the orders are divided into a primary set according to thickness, and the orders in the primary set have the same thickness; then the orders in the primary set are grouped a second time according to the steel code, and the orders with the same steel code are grouped into the same set, which is the secondary set of orders;

[0023] Step 2.1: Filter orders that meet the design specifications to form the original order set;

[0024] The order design specifications mentioned above refer to the requirements for thickness range, width range, length range, steel code, number of incomplete orders, stock quantity, and width difference in the design of medium and heavy plate blanks. Orders that do not meet the order design specifications cannot meet the finished product requirements for medium and heavy plate production.

[0025] Step 2.2: Construct an empty order set and load orders that meet the order design specifications;

[0026] Step 2.3: Select an unselected order from the original order set and search if a first-level order set exists in the order set with the same thickness as the selected order. If it exists, proceed to step 2.4; otherwise, proceed to step 2.5.

[0027] Step 2.4: Search within the primary order set of the same thickness for a secondary order set whose steel code is the same as the selected order's steel code. If they are the same, proceed to step 2.6; otherwise, proceed to step 2.7.

[0028] Step 2.5: Create a new primary order set and record the thickness of the current order as the thickness of all orders contained in this primary order set. Proceed to step 2.7.

[0029] Step 2.6: Add the order to the secondary order collection. Proceed to Step 2.8.

[0030] Step 2.7: Create a new secondary order set, recording the steel code of the current order as the steel code of the orders contained in this secondary order set. Add this secondary order set to the currently searched primary order set. Add the currently selected order to the newly created secondary order set. Proceed to Step 2.8.

[0031] Step 2.8: Check if there are any orders in the original order set that have not yet been selected. If so, proceed to step 2.1; otherwise, terminate the grouping process. Proceed to step 2.9.

[0032] Step 2.9: Perform sorting operations on the second-level order sets under each first-level order set;

[0033] The sorting operation is based on the order length and the number of orders that are missing from the order list. All orders in each secondary order set are sorted in descending order by order length, and orders with the same order length are sorted in descending order by the number of orders that are missing from the order list.

[0034] The quantity of orders that are not yet fulfilled refers to the quantity that was designed into the remaining demand of the order.

[0035] Step 3: Design heuristic algorithms to generate initial medium-thick plate billet design schemes; construct different heuristic algorithms according to target priorities to cover the generation process of different initial feasible billet design schemes;

[0036] Step 3.1: Construct a heuristic set of target priorities. The design of medium-thick plate billets is based on three objectives: minimizing the number of outgoing shipments, maximizing the yield, and maximizing the weight of the sub-bills. Six priority rankings are constructed: minimum outgoing shipments - maximum yield - maximum sub-bills weight; minimum outgoing shipments - maximum sub-bills weight - maximum yield; maximum yield - minimum outgoing shipments - maximum sub-bills weight; maximum yield - maximum sub-bills weight - minimum outgoing shipments; maximum sub-bills weight - maximum yield - minimum outgoing shipments; maximum sub-bills weight - minimum outgoing shipments - maximum yield.

[0037] Step 3.2: Execute a heuristic algorithm with the objective priority of "minimum quantity of spot goods - maximum weight of billets - highest yield", which is simply referred to as spot goods-bills-yield.

[0038] Step 3.2.1: Select a billet from the set of alternative billets as the billet to be designed; wherein the set of alternative billets includes one or more billets selected by the operator.

[0039] Step 3.2.2: Select a first-level order set from the order set. The number of second-level order sets in the first-level order set cannot be 0.

[0040] Step 3.2.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-combination quantity.

[0041] Step 3.2.4: Select an order from the secondary order set and designate this order as the master order. The master order confirms the thickness, width, and length range of the design scheme; the lower limit of the length range is determined by the lower limit of the billet length; the upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value.

[0042] Step 3.2.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order.

[0043] Step 3.2.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit of the design quantity for the order.

[0044] Step 3.2.7: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0045] Step 3.2.8: Design the order according to the design quantity. Then, check if the design requirements are met based on the design success conditions of spot-sub-billet-yield. If the design success requirements of spot-sub-billet-yield are met, update the order attributes, record the design scheme, and proceed to step 3.2.4. If the design failure conditions of spot-sub-billet-yield are met, reduce the design quantity of the order until the design quantity is reduced to 0 or the design success conditions of spot-sub-billet-yield are met, and proceed to step 3.2.4. If the design feasibility conditions of spot-sub-billet-yield are met, maintain the current design status, record each design attribute, and proceed to step 3.2.5.

[0046] Step 3.2.9: Execute step 3.2.3 until there are no orders in the second-level order set that have not been traversed in the first-level order set.

[0047] Step 3.2.10: Execute step 3.2.2 until there is no more first-level order set that has not been traversed.

[0048] Step 3.2.11: Execute step 3.2.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0.

[0049] The design success conditions for spot-billet-yield are: the design thickness with tolerance, the design width with shearing variable, and the design length with cutting head and tail are all between the minimum and maximum values ​​required by the design.

[0050] The design failure conditions for spot-billet-yield are: the design width value with cutting variables is less than 0 or greater than the upper limit of the width constraint, and the design length with cutting head and tail is less than 0 or greater than the upper limit of the length constraint.

[0051] The design feasibility conditions for spot-billet-yield are as follows: the design thickness with tolerance and the design width with shearing are both between the minimum and maximum values ​​required by the design, and the design length with shearing head and tail is greater than 0 and less than its lower limit.

[0052] Step 3.3: Execute a heuristic algorithm with the objective priority of "minimum quantity of spot goods - highest yield - largest weight of billet", which is simply referred to as spot goods-yield-billet;

[0053] Step 3.3.1: Select a blank from the blank set as the blank to be designed.

[0054] Step 3.3.2: Select a first-level order set from the order set. The number of second-level order sets in the first-level order set cannot be 0.

[0055] Step 3.3.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-combination quantity.

[0056] Step 3.3.4: Select an order from the secondary order set and designate it as the master order. The master order will confirm the thickness, width, and length range of the design scheme. The lower limit of the length range is determined by the lower limit of the billet length. The upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value.

[0057] Step 3.3.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order.

[0058] Step 3.3.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit of the design quantity for the order.

[0059] Step 3.3.7: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0060] Step 3.3.8: Design the order according to the design quantity of the design order.

[0061] Step 3.3.9: Then, check whether the design requirements are met based on the design success conditions of spot goods - yield rate - sub-billet. If the design success conditions of spot goods - yield rate - sub-billet are met, update the attributes of the order, record the design scheme, and proceed to step 3.3.4. If the design failure conditions of spot goods - yield rate - sub-billet are met, proceed to step 3.3.10. If the design feasibility conditions of spot goods - yield rate - sub-billet are met, maintain the current design state, record each design attribute, and proceed to step 3.3.5. If all orders in the secondary order set have been traversed, proceed to step 3.3.11.

[0062] Step 3.3.10: Replace the billet with a smaller billet index in the current design until no billet with a smaller index exists. If the design failure condition of spot-yield-sub-billet is met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success condition of spot-yield-sub-billet is met, then proceed to step 3.3.4.

[0063] Step 3.3.11: Execute step 3.3.3 until there are no orders in the second-level order set that have not been traversed in the first-level order set.

[0064] Step 3.3.12: Execute step 3.3.2 until there is no more first-level order set that has not been traversed.

[0065] Step 3.3.13: Execute step 3.3.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0.

[0066] The successful design conditions for the spot-yield-sub-billet design are as follows: the design thickness with tolerance, the design width with shearing variation, and the design length with shearing head and tail are all between the minimum and maximum values ​​required by the design. The width jump of the design scheme is less than the width jump limit value, and the yield is greater than the lower limit value of the yield.

[0067] The design failure conditions for spot-yield-sub-billet are the same as those for spot-sub-billet-yield.

[0068] The design feasibility conditions for spot-yield-sub-billet are the same as those for spot-sub-billet-yield.

[0069] Step 3.4: Execute a heuristic algorithm with the objective priority of "highest yield - largest billet weight - minimum number of spot items", which is simply referred to as yield-billet-spot;

[0070] Step 3.4.1: Select a blank from the blank set as the blank to be designed.

[0071] Step 3.4.2: Select a primary order set from the order set. The number of secondary order sets in the primary order set cannot be 0.

[0072] Step 3.4.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-combination quantity.

[0073] Step 3.4.4: Select an order from the secondary order set and designate it as the master order. The master order confirms the thickness, width, and length range of the design scheme. The lower limit of the length range is determined by the lower limit of the billet length. The upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value.

[0074] Step 3.4.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order.

[0075] Step 3.4.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit of the design quantity for the order.

[0076] Step 3.4.7: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0077] Step 3.4.8: Design the order according to the design quantity of the design order.

[0078] Step 3.4.9: Then, check whether the design requirements are met based on the design success conditions of yield rate-sub-billet-spot. If the design success conditions of yield rate-sub-billet-spot are met, update the attributes of the order, record the design scheme, and execute step 3.4.4. If the design failure conditions of yield rate-sub-billet-spot are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of yield rate-sub-billet-spot are met. If the design feasibility conditions of yield rate-sub-billet-spot are met, maintain the current design state, record each design attribute, and execute step 3.4.5. If all orders in the secondary order set have been traversed, execute step 3.4.11.

[0079] Step 3.4.10: If the design scheme always meets the design feasibility conditions of yield-sub-bill-spot, but cannot meet the design success conditions of yield-sub-bill-spot, then execute the "maximum width spot carry-out strategy" to add spot products;

[0080] Step 3.4.10.1: Construct the optional order set brought out by the spot goods. Orders brought out by the spot goods must meet two conditions: first, the order exists in the order set already designed for this solution; second, the order's shortfall is less than 1. Then, sort the orders in the optional order set by width in descending order, and for orders of the same width, sort them by length in descending order.

[0081] Step 3.4.10.2: Record the total length of the current design scheme. Select the first order from the set of available orders. Calculate the remaining length based on the maximum limit length of the current design scheme and the total length of the current design scheme. Calculate the maximum design quantity limit for the order based on the remaining length and the length of the current order.

[0082] Step 3.4.10.3: Bring out the maximum quantity of goods, then check the design conditions of the current solution. If successful, terminate the loop; otherwise, erase the current operation, decrement the quantity of goods brought out by 1, and continue to try designing until the design is successful or the quantity of goods brought out is 0.

[0083] Step 3.4.10.4: Update the current solution attributes and select the next order to continue trying to bring out the spot goods, until all orders in the order candidate set have been traversed.

[0084] Step 3.4.11: Execute step 3.4.3 until there are no orders in the second-level order set that have not been traversed in the first-level order set.

[0085] Step 3.4.12: Execute step 3.4.2 until there is no more first-level order set that has not been traversed.

[0086] Step 3.4.13: Execute step 3.4.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0.

[0087] The design success conditions for the yield-sub-bill-spot are the same as those for the spot-sub-bill-yield design.

[0088] The design failure conditions for the yield-sub-billet-spot are the same as those for the spot-sub-billet-yield design failure.

[0089] The design feasibility conditions for the yield-billet-spot are the same as those for the spot-billet-yield design.

[0090] Step 3.5: Execute a heuristic algorithm with the objective priority of "highest yield - minimum number of spot items - maximum billet weight", which is simply referred to as yield-spot-billet;

[0091] Step 3.5.1: Select a blank from the blank set as the blank to be designed.

[0092] Step 3.5.2: Select a first-level order set from the order set. The number of second-level order sets in the first-level order set cannot be 0.

[0093] Step 3.5.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-pair quantity.

[0094] Step 3.5.4: Select an order from the secondary order pool and designate it as the master order. The master order confirms the thickness, width, and length range of the design scheme. The lower limit of the length range is determined by the lower limit of the billet length. The upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value.

[0095] Step 3.5.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order.

[0096] Step 3.5.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit of the design quantity for the order.

[0097] Step 3.5.7: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0098] Step 3.5.8: Design the order according to the design quantity of the design order.

[0099] Step 3.5.9: Then, check whether the design requirements are met based on the design success conditions of yield rate - spot goods - sub-bills. If the design success requirements are met, update the attributes of the order, record the design scheme, and execute step 3.5.4. If the design failure conditions of yield rate - spot goods - sub-bills are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of yield rate - spot goods - sub-bills are met. If the design feasibility conditions of yield rate - spot goods - sub-bills are met, maintain the current design state, record each design attribute, and execute step 3.5.5. If all orders in the secondary order set have been traversed, execute step 3.5.11.

[0100] Step 3.5.10: If the design scheme always meets the design feasibility conditions of yield rate-spot goods-sub-blank, but cannot meet the design success conditions of yield rate-spot goods-sub-blank, then execute the "minimum spot quantity spot carry-out strategy" and add spot products.

[0101] The minimum spot quantity spot take-out strategy is as follows: if the current scheme does not meet the design success conditions of yield-spot-sub-billet, but meets the design feasibility conditions of yield-spot-sub-billet, then try to take out spot;

[0102] Step 3.5.10.1: Construct the optional order set brought out by the spot goods. Orders brought out by the spot goods must meet two conditions: first, the order exists in the order set already designed for this solution; second, the order's shortfall is less than 1. Then, sort the orders in the optional order set by unit weight in descending order, and within the same unit weight, sort them by length in descending order.

[0103] Step 3.5.10.2: Record the total length of the current design scheme. Select the first order from the set of available orders. Calculate the remaining length based on the maximum limit length of the current design scheme and the total length of the current design scheme. Calculate the maximum design quantity limit for the order based on the remaining length and the length of the current order.

[0104] Step 3.5.10.3: Design the order onto the solution with the maximum design quantity, update the current solution length, check if the design success condition is met. If it is met, terminate the design. If it is not met, remove the current design operation, decrement the design quantity of the order by 1, design it onto the solution again, and re-check the design success condition until the design quantity of the order drops to 0.

[0105] Step 3.5.10.4: Return to step 3.5.2 and select another order from the order set until all orders in the order set have been traversed.

[0106] Step 3.5.10.5: Determine the success conditions for the design. If successful, record the attributes of this design; otherwise, release all orders related to this design.

[0107] Step 3.5.11: Execute step 3.5.3 until there are no orders in the second-level order set that have not been traversed in the first-level order set.

[0108] Step 3.5.12: Execute step 3.5.2 until there is no more first-level order set that has not been traversed.

[0109] Step 3.5.13: Execute step 3.5.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0.

[0110] Step 3.6: Execute a heuristic algorithm with the objective priority of "maximum billet weight - highest yield - minimum quantity of spot goods", which is simply referred to as billet-yield-spot goods;

[0111] Step 3.6.1: Select the first billet in the billet set as the design billet.

[0112] Step 3.6.2: Select a first-level order set from the order set. The number of second-level order sets in the first-level order set cannot be 0.

[0113] Step 3.6.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-combination quantity.

[0114] Step 3.6.4: Select an order from the secondary order set and designate this order as the master order. The master order confirms the thickness, width, and length range of the design scheme. The lower limit of the length range is determined by the lower limit of the billet length. The upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value.

[0115] Step 3.6.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order.

[0116] Step 3.6.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit of the design quantity for the order.

[0117] Step 3.6.7: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0118] Step 3.6.8: Design the order according to the design quantity of the design order.

[0119] Step 3.6.9: Then, check whether the design requirements are met based on the design success conditions of billet-yield-spot goods. If the design success requirements are met, update the attributes of the order, record the design scheme, and proceed to step 3.6.4. If the design failure conditions of billet-yield-spot goods are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of billet-yield-spot goods are met. If the design feasibility conditions of billet-yield-spot goods are met, maintain the current design state, record each design attribute, and proceed to step 3.6.5. If all orders in the secondary order set have been traversed, proceed to step 3.6.11.

[0120] Step 3.6.10: If the design scheme consistently meets the design feasibility conditions of billet-yield-spot availability but fails to meet the design success conditions of billet-yield-spot availability, then execute the "bill replacement strategy".

[0121] The billet replacement strategy is as follows: the index values ​​of all billets in the candidate billet set are greater than the index of the billet corresponding to the current design. Starting with the billet with the smallest index value in the candidate billet set, the process iterates through the billets in the candidate set and compares them with the current design scheme. If the design success condition is met, the design scheme is updated, and the billet replacement process terminates. Otherwise, the billet traversal continues until all billets in the candidate set have been traversed, at which point the billet replacement process terminates.

[0122] Step 3.6.11: If the design success conditions of sub-blank-yield-spot cannot be met in step 3.6.10, then execute the "maximum width spot carry-out strategy" to add spot sub-boards.

[0123] Step 3.6.11: Execute step 3.6.3 until there are no orders in the second-level order set that have not been traversed in the first-level order set.

[0124] Step 3.6.12: Execute step 3.6.2 until there is no more first-level order set that has not been traversed.

[0125] Step 3.7: Execute a heuristic algorithm with the objective priority of "maximum billet weight - minimum quantity of spot goods - highest yield", which is simply referred to as billet-spot goods-yield.

[0126] Step 3.7.1: Select the first blank in the blank set as the design blank.

[0127] Step 3.7.2: Select a first-level order set from the order set. The number of second-level order sets in the first-level order set cannot be 0.

[0128] Step 3.7.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-combination quantity.

[0129] Step 3.7.4: Select an order from the secondary order set and designate it as the master order. The master order confirms the thickness, width, and length range of the design scheme. The lower limit of the length range is determined by the lower limit of the billet length. The upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value.

[0130] Step 3.7.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order.

[0131] Step 3.7.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit of the design quantity for the order.

[0132] Step 3.7.7: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0133] Step 3.7.8: Design the order according to the design quantity of the design order.

[0134] Step 3.7.9: Then, check whether the design requirements are met based on the design success conditions of billet-spot goods-yield rate. If the design success requirements are met, update the attributes of the order, record the design scheme, and proceed to step 3.7.4. If the design failure conditions of billet-yield rate-spot goods are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of billet-yield rate-spot goods are met. If the design feasibility conditions of billet-yield rate-spot goods are met, maintain the current design state, record each design attribute, and proceed to step 3.7.5. If all orders in the secondary order set have been traversed, proceed to step 3.7.11.

[0135] Step 3.7.10: If the design scheme always meets the design feasibility conditions of billet-yield-spot, but cannot meet the design success conditions of billet-yield-spot, then execute the "maximum width spot carry-out strategy".

[0136] Step 3.7.11: Execute step 3.7.3 until there are no orders in the second-level order set that have not been traversed in the first-level order set.

[0137] Step 3.7.12: Execute step 3.7.2 until there is no more first-level order set that has not been traversed.

[0138] Step 3.7.13: Implement the spot elimination strategy.

[0139] The strategy for eliminating spot inventory is as follows: if the current solution has spot inventory, then attempt to eliminate the spot inventory;

[0140] Step 3.7.13.1: Construct a "Spot Optimization Candidate Set" based on the original set of optimization solutions. The original set of optimization solutions refers to the set of design solutions generated in steps 3.7.1-3.7.12. The "Spot Optimization Candidate Set" includes all design solutions with spot availability.

[0141] Step 3.7.13.2: Remove the design schemes from the "Spot Optimization Candidate Scheme Set" from the original design scheme set.

[0142] Step 3.7.13.3: Divide the candidate solutions in the candidate spot optimization solution set into multiple subsets according to their thickness.

[0143] Step 3.7.13.4: Select one subset from multiple subsets and release all orders contained in the schemes within it. Delete the spot orders from the released orders, and the remaining released orders constitute the "redesigned order set".

[0144] Step 3.7.13.5: Use the heuristic method of "largest billet weight - highest yield - smallest quantity of goods shipped" to redesign the orders in the "redesign order set" to obtain a new set of design schemes.

[0145] Step 3.7.13.6: Merge the set of new design schemes into the set of original design schemes.

[0146] Step 4: Design a crossover strategy for the billet sub-plates to improve the initial feasible design scheme. The specific crossover strategy is as follows:

[0147] Step 4.1: Division of the cross-cutting scheme set; All design schemes included in the billet design plan are divided according to thickness, and for the same thickness, they are divided according to steel code. A cross-cutting scheme set is established for design schemes of the same thickness and steel code. The schemes in the cross-cutting scheme set are sorted from largest to smallest according to the billet index; the schemes at the top of the sort are those requiring improvement, and the schemes at the bottom of the sort are the basis for improvement.

[0148] Step 4.2: Select the solutions that need to be used as the basis for improvement in ascending order;

[0149] Step 4.3: Filter the cross-cutting solutions that have the same sub-board as the solution to be improved, and sort these solutions down as the preferred solutions to be improved.

[0150] Step 4.4: Select billets with a billet index greater than the billet index designed in the improved basis scheme to construct a reduced-order billet set for the improved basis scheme and create a candidate set for order adjustment.

[0151] Step 4.5: Subtract sub-boards from the improved scheme, reducing the number of sub-boards in the scheme sequentially, and checking the design success conditions of sub-bill-yield-spot stock. If the design success conditions of sub-bill-yield-spot stock are met, update the attributes of the current scheme, record the reduced orders, and add them to the order adjustment candidate set. If the design success conditions of sub-bill-yield-spot stock are not met, replace the billet and continue checking.

[0152] Step 4.6: Release all design sub-boards of the improved scheme and load them into the order adjustment candidate set.

[0153] Step 4.7: Query the primary and secondary order sets to which the orders in the current order adjustment candidate set belong. Query the orders in these sets; if any order has a shortfall of more than 0 units, add that order to the candidate order set.

[0154] Step 4.8: Construct an upgraded billet set for the improved solution, and fill the billet candidate set with billets whose indices are less than those of the billets belonging to the improved solution.

[0155] Step 4.9: Adjust the order candidates. Sort the orders in the collection in descending order of width and length.

[0156] Step 4.10: Select the billets from the upgraded billet set of the improved scheme in sequence to prepare for the design process.

[0157] Step 4.11: Select a billet from the improved billet set in ascending order.

[0158] Step 4.12: Select an order from the order adjustment candidate set as the master order, and determine the width, thickness, limit length, and steel code of the solution.

[0159] Step 4.13: Select orders from the order adjustment candidate set as the orders to be attempted for assembly.

[0160] Step 4.14: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit for the design quantity of the order.

[0161] Step 4.15: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0162] Step 4.16: Design the order according to the design quantity of the design order.

[0163] Step 4.17: Then check whether the design requirements are met according to the design success conditions. If the design success requirements are met, update the order attributes, record the design scheme, and proceed to step 4.12. If the design failure conditions are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions are met. If the design feasibility conditions are met, maintain the current design state, record each design attribute, and proceed to step 4.13.

[0164] Step 4.18: If the design scheme consistently meets the design feasibility conditions but fails to meet the design success conditions, then replace the billet. If replacing the billet still fails to meet the design success conditions, then execute the in-stock product carry-out strategy and add in-stock products.

[0165] Step 4.19: Check if the current scheme is the same as the initial scheme. If they are the same, do not update the design flag.

[0166] Step 5: Design a variation strategy for the billet sub-plate and improve the initial feasible design scheme;

[0167] Step 5.1: Division of the variable scheme set: All design schemes included in the billet design plan are divided according to thickness, and for the same thickness, they are divided according to steel code;

[0168] Step 5.2: Randomly select 3 schemes from the set of mutable schemes with the same thickness and steel code as the schemes to participate in the mutation;

[0169] Step 5.3: Release all orders on the three selected schemes to form a candidate order set, and then sort the orders in the candidate order set in descending order of width and length.

[0170] Step 5.4: Select one billet from the billet set that has not been selected before;

[0171] Step 5.5: Select an order from the candidate order set as the master order, determine the thickness, width, limit length and steel code of the design scheme based on the attributes of the master order and the selected billet, and calculate the upper limit of the design quantity of the master order.

[0172] Step 5.6: Select an order from the candidate order set, and calculate the maximum number of designs for the selected order based on the remaining length of the current design scheme and the length of the selected order;

[0173] Step 5.7: Design orders according to the maximum design quantity limit, update the current design scheme, check the design success conditions. If the design is successful, record the design scheme and terminate the design. If the design failure conditions are met, release all current orders and execute step 5.5. If the design feasibility conditions are met, retain the current design status and execute step 5.6.

[0174] Step 5.8: If the design scheme still meets the design feasibility conditions after all orders in the candidate order set have been traversed, then keep the current design scheme unchanged, change the blank material designed for the new design scheme, and check the design judgment conditions at the same time. If the design success conditions are met, record the design scheme; otherwise, continue to change the blank material until the design success conditions are met.

[0175] Step 5.9: If the design success conditions cannot be met after changing the billet, then implement the spot-out strategy;

[0176] Step 6: Implement the design scheme selection strategy; the selection strategy includes three selection criteria: replacement criteria, pooling criteria, and rejection criteria.

[0177] Replacement criteria: The three objectives of spot quantity, billet weight and yield can all control the schemes in the scheme pool.

[0178] Inclusion criteria: At least one of the three objectives—the quantity of spot goods taken out, the weight of billets, and the yield rate—cannot be dominated by the schemes in the scheme pool.

[0179] Discard criteria: The three objectives of spot quantity, billet weight and yield can all be controlled by a scheme in the scheme pool.

[0180] Step 6.1: Select solutions from the initial solution set, the crossover solution set, and the mutation solution set in turn to try to enter the pool, until all solutions in the three sets have been selected.

[0181] Step 6.2: If the proposed solution meets the replacement criteria, replace one of the proposed solutions in the pool and then proceed to step 6.1; if the replacement criteria are not met, proceed to step 6.3.

[0182] Step 6.3: If the proposed solution meets the discard criteria, it cannot be added to the pool and proceed to step 6.1; otherwise, proceed to step 6.4.

[0183] Step 6.4: If the proposed solution meets the inclusion criteria, compare whether the solution already exists in the solution pool. If it does not exist, add the solution directly to the solution pool and proceed to step 6.1. If it exists, do not add the solution to the solution pool and proceed to step 6.1.

[0184] Step 7: Reduce the number of solutions in the solution pool according to the solution reduction strategy.

[0185] Step 7.1: Implement the minimum order balance strategy: Check the solutions in the solution pool. If a solution contains orders that have not been fully utilized, delete that solution.

[0186] Step 7.2: Implement the minimum available stock strategy: Check the solutions in the solution pool. If a solution has available stock, delete the solution until the number of solutions in the solution pool equals the limit number of solutions.

[0187] Step 7.3: Implement the strategy of minimizing the number of medium and heavy plates: Sort the schemes in the scheme pool in descending order of the number of medium and heavy plates designed. Delete schemes in the scheme pool with a large number of medium and heavy plates designed, until the number of medium and heavy plates designed for all schemes in the scheme pool is equal to the minimum number of medium and heavy plates or the number of schemes in the scheme pool equals the limit number of schemes.

[0188] Step 7.4: Implement a uniform billet distribution strategy: Calculate the difference between the maximum and minimum billet numbers of the design schemes in the scheme pool, sort them in descending order according to the difference, and then delete the schemes in the scheme pool in sequence until the billet difference of all schemes in the scheme pool is equal to the minimum billet difference or the number of schemes in the scheme pool is equal to the limit number of schemes.

[0189] Step 7.5: Implement the minimum billet average strategy: Calculate the average billet serial number of the design schemes in the scheme pool, and sort all schemes in descending order according to this average. Then, delete schemes in the scheme pool sequentially until the average billet serial number of all schemes in the scheme pool is equal to the minimum average billet serial number, or the number of schemes in the scheme pool equals the limit number of schemes.

[0190] Step 7.6: Implement the minimum billet quantity strategy: Count the total number of billets for each scheme in the scheme pool and sort them in descending order based on this total number of billets. Then, delete schemes from the scheme pool sequentially until the total number of billets for all schemes in the scheme pool is equal to the minimum total number of billets, or the number of schemes in the scheme pool equals the limit number of schemes.

[0191] Step 8: Adjust the distribution of sub-plates on different medium-thick plates in the same design scheme, and finally give the design scheme within the limited number of schemes.

[0192] Step 8.1: Execute the sub-board optimization strategy to optimize the distribution of sub-boards in each scheme, so that sub-boards with the same order number are distributed on the same medium-thick plate. That is, query the medium-thick plates in pairs, medium-thick plate A and medium-thick plate B. If there are two orders on medium-thick plate A, namely order a and order b, where order a and order b have the same specifications and the same order number as order c on medium-thick plate B, and the number of sub-boards of a is not less than the number of sub-boards of c, then swap the same number of sub-boards of order c on medium-thick plate B and order b on medium-thick plate A.

[0193] Step 8.2: Implement the minimum combined rolling strategy: Count the number of combined rolled slabs for each scheme in the scheme pool and sort them in descending order of the number of combined rolled slabs. Then, delete schemes from the scheme pool sequentially until the total number of combined rolled slabs for all schemes in the scheme pool equals the minimum number of combined rolled slabs or the total number of schemes in the scheme pool equals the limit number of schemes.

[0194] The beneficial effects of adopting the above technical solution are as follows:

[0195] This invention provides a multi-objective design method for medium and heavy plate billets. By considering various production constraints and finished product standard requirements during the billet design process, a multi-objective optimization model for medium and heavy plate billet design is established. This method solves problems such as low production efficiency, continuously increasing spot rate, and low yield. Based on multiple optimization indicators (production carry-over, yield, and billet weight), multiple design schemes with their own advantages are optimized to meet the design needs of different designers. Attached Figure Description

[0196] Figure 1 This is a flowchart of the medium-thick plate design method according to an embodiment of the present invention;

[0197] Figure 2This is a schematic diagram of the medium-thick plate billet design process according to an embodiment of the present invention;

[0198] Figure 3 This is an overall schematic diagram of the medium-thick plate billet design process according to an embodiment of the present invention;

[0199] Figure 4 This is a gene structure diagram of an embodiment of the present invention;

[0200] Figure 5 This is a gene partitioning diagram according to an embodiment of the present invention;

[0201] Figure 6 This is a diagram illustrating the cross-operation of an embodiment of the present invention;

[0202] Figure 7 This is a schematic diagram of a cross-example of an embodiment of the present invention;

[0203] Figure 8 This is a variation operation diagram of an embodiment of the present invention. Detailed Implementation

[0204] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0205] The specific implementation of this invention includes mathematical models, algorithms, computing hardware, and software conditions. The computing hardware conditions mainly include a local area network (LAN) consisting of at least two servers, at least two IPv4 network interfaces, and at least one router. One of the two servers carries a business data processing system, which processes business orders, slab data, etc., and then sends them to the other server undertaking the computing tasks via the ERP system. The ERP system is a digital management platform based on information technology and a systematic management philosophy, providing decision-making tools for enterprises and employees. The computing software conditions mainly include IBM WebSphere MQ communication software for remote data exchange and SQL Server 2016 database software for data management and storage. The relevant configurations of the communication software include queue manager, queues, channels, ports, and IP addresses of both communicating parties. The key relevant configurations of the database include server address, server port, database name, username, and password.

[0206] The billet and order data involved in the design of medium and heavy plate billets are transmitted from the ERP system. The billet and order data are then downloaded from the local server. Billet data includes: billet thickness, billet width, minimum billet length, and maximum billet length. Order data includes: production order number, billet requisition plan number, steel code, order thickness, order width, order length, order unit weight, number of ordered pieces, order quantity, number of missing sub-plates, missing quantity, length type code, and product edge status code.

[0207] A multi-objective design method for medium-thick plate billets, such as... Figure 1 As shown, it includes the following steps:

[0208] Step 1: Extract the design process characteristics of medium and heavy plate billets, transform the production objectives and process constraints into mathematical expressions, and thus establish a multi-objective function and constraints for the design process of medium and heavy plate billets;

[0209] The design and process characteristics of the medium-thick slab billet are as follows: Figure 2 , Figure 3 As shown, this includes: the matching relationship between billets and orders, the limitations of production equipment, the production process from billet to order, customer needs, and indicators for measuring the quality of production plans;

[0210] Step 1.1: Define the decision variable x in the design process of medium and heavy plate billets. ij x ij This represents the design quantity of order i on billet j, and its value range is x. ij >0, It is a set of integers;

[0211] Step 1.2: Define the optimization objectives in the design process of medium and heavy plate billets;

[0212] The optimization objectives include: minimizing the quantity of spot goods taken out, maximizing the yield rate, and maximizing the total weight of billets.

[0213] The spot goods are products whose production plan exceeds the order quantity in the design of medium and heavy plate billets, i.e. p i This refers to the number of units ordered in order i. These products cannot be included in existing orders and cannot be treated as futures products. They can only be processed in the spot market. Since the price of products in the spot market is lower than that in the futures market, bringing in less spot goods can increase the company's production revenue.

[0214] The yield rate is a standard for evaluating the production efficiency of a production plan designed for medium and heavy plate blanks. It represents the ratio of finished product quantity to raw material weight. Finished product quantity refers to the actual weight of sub-plates that can be cut, including both futures and spot sub-plate weights. The yield rate is expressed as Where q i It is the unit weight of order i, d i It is the length of order i.

[0215] The weight of the sub-billet is the designed raw material weight for medium-thick plate billets, i.e., the weight of the billet itself. The weight of the j-th billet is calculated based on its thickness, width, length, and steel density ρ. The designed thickness of billet j and the order thickness T are also considered. j Equivalently, the design width is the width w of the main order k.k and the amount of cutting edge S j The sum of the design length of the billet is the total length of the order plus the amount of the cut-off ends, H. j The sum of these two quantities makes the weight of the billet equal to the weight of the billet. The total weight of the raw materials is:

[0216] Step 1.3: Analyze the production process characteristics of medium and heavy plate billet design, and describe the mathematical expression of its constraints based on the defined decision variables, specifically including order constraints, production process constraints, and unique master order constraints.

[0217] The order constraints are as follows: during the production of medium and heavy plates, the designed billet is rolled by a rolling mill, and the volume of the billet changes after rolling. Therefore, the thickness, width, and length will vary depending on the order requirements. Due to limitations in production processes and equipment, the thickness range T of medium and heavy plates is... min and T max Width range W min and W max Therefore, there is T. min ≤T j ≤T max and W min ≤W j ≤W max .

[0218] The manufacturing process constraints include: each design scheme has at most one master order, i.e. z kkj This indicates whether order k is designed as a master order onto billet j; if so, it is 1, otherwise 0. In the design scheme for medium-thick plate billets, the width jump of sub-plates cannot exceed the upper limit. Width jump is a parameter affecting the yield of medium-thick plates, referring to the difference between the maximum and minimum width of all sub-plates in a design scheme. Therefore, |min{1,x uj}w u -min{1,x vj}w v |≤W c Where order u and order v are orders with different production order numbers, W c For the maximum jump width, w u It is the width of order u, w v This refers to the width of order v. The length of the design scheme cannot exceed the limit length required by the design of the medium-thick plate billet. During the rolling process, the medium-thick plate billet passes through multiple production equipment, such as rolling mills. These machines have their own billet length handling capacity limits, similar to the maximum rolling length limit of a rolling mill. Therefore, the minimum value among the upper limits of the length handling capacity of all production equipment is considered the limit length value required by the design of the medium-thick plate billet. Alternatively, this limit length value can be manually specified, but it is usually taken as the minimum of all length values, mathematically expressed as... Where D j d is the limit length value. i This refers to the length of order i. A solution with an excessively low yield rate will actually reduce production efficiency; therefore, there is a lower limit to the yield rate of the design solution. The calculated yield rate of the design solution must not be less than the lower limit Y. MIN The mathematical expression is:

[0219]

[0220] The unique master order constraint stipulates that each design scheme can only have one master order, used to define process rules, including steel code, design thickness, billet index code, shearing identifier, classification society code, etc. Where M is a maximal number.

[0221] Step 2: Group the orders according to thickness and steel code; the orders are divided into a primary set according to thickness, and the orders in the primary set have the same thickness; then the orders in the primary set are grouped a second time according to the steel code, and the orders with the same steel code are grouped into the same set, which is the secondary set of orders;

[0222] Step 2.1: Filter orders that meet the design specifications to form the original order set;

[0223] The order design specifications mentioned above refer to the requirements for thickness range, width range, length range, steel code, number of incomplete orders, stock quantity, and width difference in the design of medium and heavy plate blanks. Orders that do not meet the order design specifications cannot meet the finished product requirements for medium and heavy plate production.

[0224] Step 2.2: Construct an empty order set and load orders that meet the order design specifications;

[0225] Step 2.3: Select an unselected order from the original order set and search if a first-level order set exists in the order set with the same thickness as the selected order. If it exists, proceed to step 2.4; otherwise, proceed to step 2.5.

[0226] Step 2.4: Search within the primary order set of the same thickness for a secondary order set whose steel code is the same as the selected order's steel code. If they are the same, proceed to step 2.6; otherwise, proceed to step 2.7.

[0227] Step 2.5: Create a new primary order set and record the thickness of the current order as the thickness of all orders contained in this primary order set. Proceed to step 2.7.

[0228] Step 2.6: Add the order to the secondary order collection. Proceed to Step 2.8.

[0229] Step 2.7: Create a new secondary order set, recording the steel code of the current order as the steel code of the orders contained in this secondary order set. Add this secondary order set to the currently searched primary order set. Add the currently selected order to the newly created secondary order set. Proceed to Step 2.8.

[0230] Step 2.8: Check if there are any orders in the original order set that have not yet been selected. If so, proceed to step 2.1; otherwise, terminate the grouping process. Proceed to step 2.9.

[0231] Step 2.9: Perform sorting operations on the second-level order sets under each first-level order set;

[0232] The sorting operation is based on the order length and the number of orders that are missing from the order list. All orders in each secondary order set are sorted in descending order by order length, and orders with the same order length are sorted in descending order by the number of orders that are missing from the order list.

[0233] The quantity of orders that are not yet fulfilled refers to the quantity that was designed into the remaining demand of the order.

[0234] Step 3: Design a heuristic algorithm to generate an initial medium-thick plate billet design scheme;

[0235] Step 3.1: Construct a heuristic set of target priorities. The design of medium-thick plate billets is based on three objectives: minimizing the number of outgoing shipments, maximizing the yield, and maximizing the weight of the sub-bills. Six priority rankings are constructed: minimum outgoing shipments - maximum yield - maximum sub-bills weight; minimum outgoing shipments - maximum sub-bills weight - maximum yield; maximum yield - minimum outgoing shipments - maximum sub-bills weight; maximum yield - maximum sub-bills weight - minimum outgoing shipments; maximum sub-bills weight - maximum yield - minimum outgoing shipments; maximum sub-bills weight - minimum outgoing shipments - maximum yield.

[0236] Step 3.2: Execute a heuristic algorithm with the objective priority of "minimum quantity of spot goods - maximum weight of billets - highest yield", which is simply referred to as spot goods-bills-yield.

[0237] Step 3.3: Execute a heuristic algorithm with the objective priority of "minimum quantity of spot goods - highest yield - largest weight of billet", which is simply referred to as spot goods-yield-billet;

[0238] Step 3.4: Execute a heuristic algorithm with the objective priority of "highest yield - largest billet weight - minimum number of spot items", which is simply referred to as yield-billet-spot;

[0239] Step 3.5: Execute a heuristic algorithm with the objective priority of "highest yield - minimum number of spot items - maximum billet weight", which is simply referred to as yield-spot-billet;

[0240] Step 3.6: Execute a heuristic algorithm with the objective priority of "maximum billet weight - highest yield - minimum quantity of spot goods", which is simply referred to as billet-yield-spot goods;

[0241] Step 3.7: Execute a heuristic algorithm with the objective priority of "maximum billet weight - minimum quantity of spot goods - highest yield", which is simply referred to as billet-spot goods-yield.

[0242] Different heuristic algorithms are constructed according to target priorities to cover the generation process of different initial feasible billet design schemes. In this method, the concepts of "large genes" and "small genes" are specifically proposed. Small genes are gene points; numerous small genes form large genes; and numerous large genes form gene sequences. For example... Figure 4 As shown, the length of a large gene is fixed, while the length of its gene sequence is variable. Sub-plates are small genes, and sub-plates distributed across a large plate are large genes. A complete design scheme is represented by a single gene sequence, or individual. The population is N, indicating that there are N individuals in the population.

[0243] Step 4: Design a cross-design strategy for billet sub-plates and improve the initial feasible design scheme.

[0244] Step 4.1: Division of the cross-cutting scheme set; All design schemes included in the billet design plan are divided according to thickness, and for the same thickness, they are divided according to steel code. A cross-cutting scheme set is established for design schemes of the same thickness and steel code. The schemes in the cross-cutting scheme set are sorted from largest to smallest according to the billet index; the schemes at the top of the sort are those requiring improvement, and the schemes at the bottom of the sort are the basis for improvement.

[0245] Step 4.2: Select the solutions that need to be used as the basis for improvement in ascending order;

[0246] Step 4.3: Filter the cross-cutting solutions that have the same sub-board as the solution to be improved, and sort these solutions down as the preferred solutions to be improved.

[0247] Step 4.4: Select billets with a billet index greater than the billet index designed in the improved basis scheme to construct a reduced-order billet set for the improved basis scheme and create a candidate set for order adjustment.

[0248] Step 4.5: Subtract sub-boards from the improved scheme, reducing the number of sub-boards in the scheme sequentially, and checking the design success conditions of sub-bill-yield-spot stock. If the design success conditions of sub-bill-yield-spot stock are met, update the attributes of the current scheme, record the reduced orders, and add them to the order adjustment candidate set. If the design success conditions of sub-bill-yield-spot stock are not met, replace the billet and continue checking.

[0249] Step 4.6: Release all design sub-boards of the improved scheme and load them into the order adjustment candidate set.

[0250] Step 4.7: Query the primary and secondary order sets to which the orders in the current order adjustment candidate set belong. Query the orders in these sets; if any order has a shortfall of more than 0 units, add that order to the candidate order set.

[0251] Step 4.8: Construct an upgraded billet set for the improved solution, and fill the billet candidate set with billets whose indices are less than those of the billets belonging to the improved solution.

[0252] Step 4.9: Adjust the order candidates. Sort the orders in the collection in descending order of width and length.

[0253] Step 4.10: Select the billets from the upgraded billet set of the improved scheme in sequence to prepare for the design process.

[0254] Step 4.11: Select a billet from the improved billet set in ascending order.

[0255] Step 4.12: Select an order from the order adjustment candidate set as the master order, and determine the width, thickness, limit length, and steel code of the solution.

[0256] Step 4.13: Select orders from the order adjustment candidate set as the orders to be attempted for assembly.

[0257] Step 4.14: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order. Compare this maximum design quantity with the order's under-packing quantity, and take the minimum of the two as the actual limit for the design quantity of the order.

[0258] Step 4.15: Select the design quantity of the design order as the actual upper limit of the design quantity.

[0259] Step 4.16: Design the order according to the design quantity of the design order.

[0260] Step 4.17: Then check whether the design requirements are met according to the design success conditions. If the design success requirements are met, update the order attributes, record the design scheme, and proceed to step 4.12. If the design failure conditions are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions are met. If the design feasibility conditions are met, maintain the current design state, record each design attribute, and proceed to step 4.13.

[0261] Step 4.18: If the design scheme consistently meets the design feasibility conditions but fails to meet the design success conditions, then replace the billet. If replacing the billet still fails to meet the design success conditions, then execute the in-stock product carry-out strategy and add in-stock products.

[0262] Step 4.19: Check if the current scheme is the same as the initial scheme. If they are the same, do not update the design flag.

[0263] In this embodiment, as shown Figure 4 , Figure 5 As shown, some individuals have the following genes: Figure 5 As shown in the diagram above, there are a total of 6 large genes, representing the 6 large plates produced by this scheme. Plates 2, 4, and 5 have the same thickness and width, while plates 1, 3, and 6 have the same thickness and width. Therefore, the individuals are divided into two sets: (2,4,5) and (1,3,6), and the plates within each set are arranged in descending order according to the cross-sectional size of their designed blanks. After this arrangement, the order of the plates in the sets becomes (5,2,4) and (3,1,6). Next, we perform crossover operations within each set.

[0264] like Figure 6 As shown, if there are a total of 8 orders participating in this design, then the gene lengths of large genes 5 and 4 are 8. If 3 orders are assembled on the large board represented by large gene 5, resulting in a total of 5 sub-boards, and 2 orders are assembled on the large board represented by large gene 4, resulting in 2 sub-boards, then the crossover method is to crossover the gene points on large gene 4 whose gene code is 0 with the gene points on large gene 5 whose gene code is not 0. The principle of crossover is to take one portion of the genes on large gene 5 and put it into an external gene point pool so that the sub-board situation on large gene 5 can still meet the design conditions. At the same time, the orders (gene points) that are not involved in the design are also put into the gene point pool, and then the gene points in the gene point pool are added to large gene 4 one by one, so that the billet cross-section of large gene 4 grows upward.

[0265] This method proposes a "gene merging" operation after performing gene crossover operations on various large genes. Specific details can be found in... Figure 7Let me explain. This method uses 6 heuristic algorithms, resulting in a population of 6 individuals. Individuals are divided into gene segments based on thickness and width. The gene segments of each individual are then compared vertically. The results show that the first individual has a superior gene segment (segment 2), the second has a superior gene segment (segment 3), the third has no superior gene, the fourth has a superior gene segment (segment 4,5), the fifth has no superior gene segment, and the sixth has a superior gene segment (segment 1). The superior gene segments from these 6 individuals are then concatenated to create an individual that incorporates all the superior genes in the current population. This individual is then added to the current population.

[0266] Step 5: Design a variation strategy for the billet sub-plate to improve the initial feasible design scheme; such as... Figure 8 As shown, the mutation operation in this method involves scrambling the original individual at the gene point level. Specifically, three major genes are randomly selected from the individual. The gene points on these three major genes—that is, the sub-genes on the three major gene plates—are removed and placed into a gene point pool along with the unassembled sub-genes. The gene points in the gene point pool are then used to splice the major genes together, forming new major genes. The number of new major genes may not be three. The purpose of the mutation operation is primarily to increase the directions of exploration.

[0267] Step 5.1: Division of the set of mutable schemes: All design schemes included in the billet design plan are divided according to thickness, and those of the same thickness are divided according to steel code; only design schemes of the same thickness and the same steel code are eligible to undergo common mutation.

[0268] Step 5.2: Randomly select 3 schemes from the set of mutable schemes with the same thickness and steel code as the schemes to participate in the mutation;

[0269] Step 5.3: Release all orders on the three selected schemes to form a candidate order set, and then sort the orders in the candidate order set in descending order of width and length.

[0270] Step 5.4: Select one billet from the billet set that has not been selected before;

[0271] Step 5.5: Select an order from the candidate order set as the master order, determine the thickness, width, limit length and steel code of the design scheme based on the attributes of the master order and the selected billet, and calculate the upper limit of the design quantity of the master order.

[0272] Step 5.6: Select an order from the candidate order set, and calculate the maximum number of designs for the selected order based on the remaining length of the current design scheme and the length of the selected order;

[0273] Step 5.7: Design orders according to the maximum design quantity limit, update the current design scheme, check the design success conditions. If the design is successful, record the design scheme and terminate the design. If the design failure conditions are met, release all current orders and execute step 5.5. If the design feasibility conditions are met, retain the current design status and execute step 5.6.

[0274] Step 5.8: If the design scheme still meets the design feasibility conditions after all orders in the candidate order set have been traversed, then keep the current design scheme unchanged, change the blank material designed for the new design scheme, and check the design judgment conditions at the same time. If the design success conditions are met, record the design scheme; otherwise, continue to change the blank material until the design success conditions are met.

[0275] Step 5.9: If the design success conditions cannot be met after changing the billet, then implement the spot-out strategy;

[0276] Step 6: Implement the design scheme selection strategy. The selection strategy includes three selection criteria: replacement criteria, inclusion criteria, and rejection criteria.

[0277] Replacement criteria: The three objectives of spot quantity, billet weight and yield can all control the schemes in the scheme pool.

[0278] Inclusion criteria: At least one of the three objectives—the quantity of spot goods taken out, the weight of billets, and the yield rate—cannot be dominated by the schemes in the scheme pool.

[0279] Discard criteria: The three objectives of spot quantity, billet weight and yield can all be controlled by a scheme in the scheme pool.

[0280] Step 6.1: Select solutions from the initial solution set, the crossover solution set, and the mutation solution set in turn to try to enter the pool, until all solutions in the three sets have been selected.

[0281] Step 6.2: If the proposed solution meets the replacement criteria, replace one of the proposed solutions in the pool and then proceed to step 6.1; if the replacement criteria are not met, proceed to step 6.3.

[0282] Step 6.3: If the proposed solution meets the discard criteria, it cannot be added to the pool and proceed to step 6.1; otherwise, proceed to step 6.4.

[0283] Step 6.4: If the proposed solution meets the inclusion criteria, compare whether the solution already exists in the solution pool. If it does not exist, add the solution directly to the solution pool and proceed to step 6.1. If it exists, do not add the solution to the solution pool and proceed to step 6.1.

[0284] Step 7: Reduce the number of solutions in the solution pool according to the solution reduction strategy.

[0285] Step 7.1: Implement the minimum order balance strategy: Check the solutions in the solution pool. If a solution contains orders that have not been fully utilized, delete that solution.

[0286] Step 7.2: Implement the minimum available stock strategy: Check the solutions in the solution pool. If a solution has available stock, delete the solution until the number of solutions in the solution pool equals the limit number of solutions.

[0287] Step 7.3: Implement the strategy of minimizing the number of medium and heavy plates: Sort the schemes in the scheme pool in descending order of the number of medium and heavy plates designed. Delete schemes in the scheme pool with a large number of medium and heavy plates designed, until the number of medium and heavy plates designed for all schemes in the scheme pool is equal to the minimum number of medium and heavy plates or the number of schemes in the scheme pool equals the limit number of schemes.

[0288] Step 7.4: Implement a uniform billet distribution strategy: Calculate the difference between the maximum and minimum billet numbers of the design schemes in the scheme pool, sort them in descending order according to the difference, and then delete the schemes in the scheme pool in sequence until the billet difference of all schemes in the scheme pool is equal to the minimum billet difference or the number of schemes in the scheme pool is equal to the limit number of schemes.

[0289] Step 7.5: Implement the minimum billet average strategy: Calculate the average billet serial number of the design schemes in the scheme pool, and sort all schemes in descending order according to this average. Then, delete schemes in the scheme pool sequentially until the average billet serial number of all schemes in the scheme pool is equal to the minimum average billet serial number, or the number of schemes in the scheme pool equals the limit number of schemes.

[0290] Step 7.6: Implement the minimum billet quantity strategy: Count the total number of billets for each scheme in the scheme pool and sort them in descending order based on this total number of billets. Then, delete schemes from the scheme pool sequentially until the total number of billets for all schemes in the scheme pool is equal to the minimum total number of billets, or the number of schemes in the scheme pool equals the limit number of schemes.

[0291] Step 8: Adjust the distribution of sub-plates on different medium-thick plates within the same design scheme, and finally provide a design scheme within the limited number of schemes.

[0292] Step 8.1: Execute the sub-board optimization strategy to optimize the distribution of sub-boards in each scheme, so that sub-boards with the same order number are distributed on the same medium-thick plate. That is, query the medium-thick plates in pairs, medium-thick plate A and medium-thick plate B. If there are two orders on medium-thick plate A, namely order a and order b, where order a and order b have the same specifications and the same order number as order c on medium-thick plate B, and the number of sub-boards of a is not less than the number of sub-boards of c, then swap the same number of sub-boards of order c on medium-thick plate B and order b on medium-thick plate A.

[0293] Step 8.2: Implement the minimum combined rolling strategy: Count the number of combined rolled slabs for each scheme in the scheme pool and sort them in descending order of the number of combined rolled slabs. Then, delete schemes from the scheme pool sequentially until the total number of combined rolled slabs for all schemes in the scheme pool equals the minimum number of combined rolled slabs or the total number of schemes in the scheme pool equals the limit number of schemes.

[0294] Compared to manual design methods, this method can generate multiple design schemes, each with its own advantages, for billet designers to choose from. Compared to manual design methods, the default design schemes provided by this method increase the yield by 2.12%, decrease the quantity of available billets by 1.72%, and increase the weight of the finished billets by 1.92%.

[0295] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A multi-objective based design method for plate blank, characterized in that, Includes the following steps: Step 1: Extract the design process characteristics of medium and heavy plate billets, transform the production objectives and process constraints into mathematical expressions, and thus establish a multi-objective function and constraints for the design process of medium and heavy plate billets; The design process features of the medium and heavy plate billet include: the matching relationship between billet and order, the limitations of production equipment, the production process from billet to order, customer needs, and the quality measurement indicators of production plan. Step 2: Group the orders according to thickness and steel code; the orders are divided into a primary set according to thickness, and the orders in the primary set have the same thickness; then the orders in the primary set are grouped a second time according to the steel code, and the orders with the same steel code are grouped into the same set, which is the secondary set of orders; Step 3: Design heuristic algorithms to generate initial medium-thick plate billet design schemes; construct different heuristic algorithms according to target priorities to cover the generation process of different initial feasible billet design schemes; Step 4: Design a cross-design strategy for billet sub-plates and improve the initial feasible design scheme; Step 5: Design a variation strategy for the billet sub-plate and improve the initial feasible design scheme; Step 6: Design scheme selection strategy; The selection strategy includes three selection criteria: replacement criteria, inclusion criteria, and rejection criteria. Replacement criteria: The three objectives of spot quantity, billet weight, and yield can all influence the solutions in the solution pool; Inclusion criteria: At least one of the three objectives—the quantity of spot goods taken out, the weight of billets, and the yield rate—cannot be dominated by the schemes in the scheme pool; Discard criteria: The three objectives of spot quantity, billet weight, and yield can all be controlled by a certain scheme in the scheme pool; The solutions in the initial solution set, the cross solution set, and the mutated solution set are sequentially tried to enter the pool. If a solution to be tried to enter the pool meets the replacement criterion, a solution in the solution pool is replaced. If the solution meets the entry criteria, it is compared whether the solution already exists in the solution pool. If not, the solution is directly added to the solution pool. If the solution meets the discard criteria, the solution cannot be entered into the pool, and the next solution in the cross solution set is compared. Step 7: Reduce the number of solutions in the solution pool according to the solution reduction strategy; Step 8: Adjust the distribution of sub-plates on different medium-thick plates in the same design scheme, and finally give the design scheme within the limited number of schemes.

2. A multi-objective based design method for plate blank according to claim 1, wherein, Step 1 specifically includes the following steps: Step 1.1: Define decision variables x in the design process of the medium plate blank ij , x ij denotes the designed quantity of order i on blank j, which takes value in the range of x ij > 0, is an integer set; Step 1.2: Define the optimization objectives in the design process of medium and heavy plate billets; The optimization objectives include: minimizing the quantity of spot goods taken out, maximizing the yield rate, and maximizing the total weight of billets. The spot goods are products produced in the production plan for medium and heavy plate billets that exceed the number of orders placed. p i This refers to the number of items ordered in order i; The yield rate is a standard for evaluating the production efficiency of a production plan designed for medium and heavy plate blanks. It represents the ratio of finished product quantity to raw material weight. The finished product quantity refers to the actual weight of sub-plates that can be cut, including the weight of futures sub-plates and the weight of spot sub-plates. The yield rate is expressed as Where q i It is the unit weight of order i, d i It is the length of order i; The weight of the sub-billet is the raw material weight designed for medium-thick plate billets, i.e., the weight of the billet itself; the weight of the j-th billet is calculated from the billet's thickness, width, length, and steel density ρ; the design thickness of billet j and the order thickness T. j Equivalently, the design width is the width w of the main order k. k and the amount of cutting edge S j The sum of the design length of the billet is the total length of the order plus the amount of the cut-off ends, H. j The sum of these two quantities makes the weight of the billet equal to the weight of the billet. The total weight of the raw materials is: Step 1.3: Analyze the production process characteristics of medium and heavy plate billet design, and describe the mathematical expression of its constraints based on the defined decision variables, specifically including order constraints, production process constraints, and unique master order constraints. The order constraint is that, in the production of the medium plate, the designed blank is rolled by the rolling mill, and the volume of the medium blank changes after rolling. Due to the limitation of the production process and the production equipment, the thickness interval T min and T max , the width interval W min and W max of the medium plate change; therefore, T min ≤T j ≤T max and W min ≤W j ≤W max ; The manufacturing process constraints include: each design scheme has at most one master order, i.e. z kkj This indicates whether order k is designed as a master order onto billet j; if yes, it is 1, otherwise 0. In the design scheme for medium-thick plate billets, the width jump of sub-plates cannot exceed the upper limit. Width jump is a parameter affecting the yield of medium-thick plates, referring to the difference between the maximum and minimum width of all sub-plates in a design scheme. Therefore, |min{1,x uj }w u -min{1,x vj }w v |≤W c Where order u and order v are orders with different production order numbers, W c For the maximum jump width, w u It is the width of order u, w v It is the width of order v; the length of the design scheme cannot exceed the limit length required by the design of medium and heavy plate blanks, mathematically expressed as... Where D j d is the limit length value. i The length of order i is given, and the yield rate of the design scheme must not be less than the lower limit of yield rate Y. MIN The mathematical expression is: The unique master order constraint means that each design scheme can only have one master order. Where M is a maximal number.

3. The multi-objective-based medium-thick plate billet design method according to claim 1, characterized in that, Step 2 specifically includes the following steps: Step 2.1: Filter orders that meet the design specifications to form the original order set; The order design specifications mentioned above refer to the requirements of the medium and heavy plate billet design on the thickness range, width range, length range, steel code, number of incomplete orders, stock quantity, and width difference of the order; orders that do not meet the order design specifications cannot meet the finished product requirements of medium and heavy plate production; Step 2.2: Construct an empty order set and load orders that meet the order design specifications; Step 2.3: Select an unselected order from the original order set, and search if there exists a first-level order set in the order set with the same thickness as the selected order; if it exists, proceed to step 2.4; otherwise, proceed to step 2.

5. Step 2.4: Search within the primary order set of the same thickness for a secondary order set whose steel code is the same as the steel code of the selected order. If they are the same, proceed to step 2.6; otherwise, proceed to step 2.

7. Step 2.5: Create a new primary order set, and record the thickness of the current order as the thickness of the orders contained in this primary order set; proceed to step 2.7; Step 2.6: Add the order to the secondary order set; proceed to step 2.

8. Step 2.7: Create a new secondary order set, record the steel code of the current order as the steel code of the orders contained in this secondary order set, and add this secondary order set to the currently searched primary order set; add the currently selected order to the newly created secondary order set; proceed to step 2.

8. Step 2.8: Check if there are any orders in the original order set that have not yet been selected. If so, proceed to step 2.1; otherwise, terminate the grouping process and proceed to step 2.

9. Step 2.9: Perform sorting operations on the second-level order sets under each first-level order set; The sorting operation is based on the order length and the number of orders that are missing from the order list. All orders in each secondary order set are sorted in descending order by order length, and orders with the same order length are sorted in descending order by the number of orders that are missing from the order list. The quantity of orders that are not yet fulfilled refers to the quantity that was designed into the remaining demand of the order.

4. The multi-objective-based medium-thick plate billet design method according to claim 1, characterized in that, Step 3 specifically includes the following steps: Step 3.1: Construct a heuristic set of target priorities; Based on the three objectives of minimizing the number of outgoing billets, maximizing the yield, and maximizing the weight of the sub-bills, the medium and heavy plate billet design constructs six priority rankings of objectives, namely: minimum outgoing billet quantity - maximum yield - maximum sub-bills weight, minimum outgoing billet quantity - maximum sub-bills weight - maximum yield, maximum yield - minimum outgoing billet quantity - maximum sub-bills weight, maximum yield - maximum sub-bills weight - minimum outgoing billet quantity, maximum sub-bills weight - maximum yield - minimum outgoing billet quantity, maximum sub-bills weight - minimum outgoing billet quantity - maximum yield; Step 3.2: Execute a heuristic algorithm with the objective priority of "minimum quantity of spot goods - maximum weight of billets - highest yield", which is simply referred to as spot goods-bills-yield. Step 3.2.1: Select a billet from the alternative billet set as the billet to be designed; wherein the alternative billet set includes one or more billets selected by the operator; Step 3.2.2: Select a primary order set from the order set. The number of secondary order sets in the primary order set cannot be zero. Step 3.2.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-patch quantity; Step 3.2.4: Select an order from the secondary order set and designate this order as the master order; the master order confirms the thickness, width, and length range of the design scheme; the lower limit of the length range is determined by the lower limit of the billet length; the upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value; Step 3.2.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order; Step 3.2.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order; compare the maximum design quantity with the under-packing quantity of the order, and take the minimum of the two as the actual limit of the design quantity of the order; Step 3.2.7: Select the actual upper limit of the design quantity for the design order; Step 3.2.8: Design the order according to the design quantity of the design order; then check whether the design requirements are met based on the design success conditions of spot-sub-billet-yield rate; if the design success requirements of spot-sub-billet-yield rate are met, update the attributes of the order, record the design scheme, and execute step 3.2.4; if the design failure conditions of spot-sub-billet-yield rate are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of spot-sub-billet-yield rate are met, and execute step 3.2.4; if the design feasibility conditions of spot-sub-billet-yield rate are met, maintain the current design status, record each design attribute, and execute step 3.2.5; Step 3.2.9: Execute step 3.2.3 until there are no orders left to be traversed in the second-level order set within the first-level order set; Step 3.2.10: Execute step 3.2.2 until there are no more first-level order sets that have not been traversed; Step 3.2.11: Execute step 3.2.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0; The design success conditions for spot-billet-yield are: the design thickness with tolerance, the design width with shearing variable, and the design length with cutting head and tail are all between the minimum and maximum values ​​required by the design. The design failure conditions for spot-billet-yield rate are: the design width value with cutting variables is less than 0 or greater than the upper limit of the width constraint, and the design length with cutting head and tail amounts is less than 0 or greater than the upper limit of the length constraint. The design feasibility conditions for the spot-billet-yield ratio are as follows: the design thickness with tolerance and the design width with shearing variable are both between the minimum and maximum values ​​required by the design, and the design length with shearing head and tail is greater than 0 and less than its lower limit. Step 3.3: Execute a heuristic algorithm with the objective priority of "minimum quantity of spot goods - highest yield - largest weight of billet", which is simply referred to as spot goods-yield-billet; Step 3.3.1: Select a blank from the blank set as the blank to be designed; Step 3.3.2: Select a primary order set from the order set. The number of secondary order sets in the primary order set cannot be zero. Step 3.3.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-patch quantity; Step 3.3.4: Select an order from the secondary order set and designate this order as the master order. The master order confirms the thickness, width, and length range of the design scheme. The lower limit of the length range is determined by the lower limit of the billet length. The upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value. Step 3.3.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order; Step 3.3.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order; compare the maximum design quantity with the under-packing quantity of the order, and take the minimum of the two as the actual limit of the design quantity of the order; Step 3.3.7: Select the actual upper limit of the design quantity for the design order; Step 3.3.8: Design the order according to the design quantity in the design order; Step 3.3.9: Then, check whether the design requirements are met based on the design success conditions of spot goods-yield rate-sub-blank; if the design success conditions of spot goods-yield rate-sub-blank are met, update the attributes of the order, record the design scheme, and execute step 3.3.4; if the design failure conditions of spot goods-yield rate-sub-blank are met, execute step 3.3.10; if the design feasibility conditions of spot goods-yield rate-sub-blank are met, maintain the current design state, record each design attribute, and execute step 3.3.5; if all orders in the secondary order set have been traversed, execute step 3.3.

11. Step 3.3.10: Replace the billet with a smaller billet index in the current design until there are no billets with smaller indices; if the design failure condition of spot-yield-sub-billet is met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success condition of spot-yield-sub-billet is met, then proceed to step 3.3.4; Step 3.3.11: Execute step 3.3.3 until there are no orders left to be traversed in the second-level order set within the first-level order set; Step 3.3.12: Execute step 3.3.2 until there are no more first-level order sets that have not been traversed; Step 3.3.13: Execute step 3.3.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0; The successful design conditions for spot-yield-sub-billet are as follows: the design thickness with tolerance, the design width with shearing variable, and the design length with shearing head and tail are all between the minimum and maximum values ​​required by the design; the width jump of the design scheme is less than the width jump limit value, and the yield is greater than the lower limit value of the yield; The design failure conditions for spot-yield-sub-billet are the same as those for spot-sub-billet-yield. The design feasibility conditions for spot-yield-sub-billet are the same as those for spot-sub-billet-yield. Step 3.4: Execute a heuristic algorithm with the objective priority of "highest yield - largest billet weight - minimum number of spot items", which is simply referred to as yield-billet-spot; Step 3.4.1: Select a blank from the blank set as the blank to be designed; Step 3.4.2: Select a primary order set from the order set. The number of secondary order sets in the primary order set cannot be zero. Step 3.4.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-patch quantity; Step 3.4.4: Select an order from the secondary order set and designate this order as the master order; the master order confirms the thickness, width, and length range of the design scheme; the lower limit of the length range is determined by the lower limit of the billet length; the upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value. Step 3.4.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order; Step 3.4.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order; compare the maximum design quantity with the under-packing quantity of the order, and take the minimum of the two as the actual limit of the design quantity of the order; Step 3.4.7: Select the actual upper limit of the design quantity for the design order; Step 3.4.8: Design the order according to the design quantity in the design order; Step 3.4.9: Then, check whether the design requirements are met based on the design success conditions of yield rate-sub-billet-spot; if the design success conditions of yield rate-sub-billet-spot are met, update the attributes of the order, record the design scheme, and execute step 3.4.4; if the design failure conditions of yield rate-sub-billet-spot are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of yield rate-sub-billet-spot are met; if the design feasibility conditions of yield rate-sub-billet-spot are met, maintain the current design status, record each design attribute, and execute step 3.4.5; if all orders in the secondary order set have been traversed, execute step 3.4.11; Step 3.4.10: If the design scheme always meets the design feasibility conditions of yield-sub-bill-spot, but cannot meet the design success conditions of yield-sub-bill-spot, then execute the "maximum width spot carry-out strategy" to add spot products; Step 3.4.10.1: Construct the optional order set brought out by the spot goods; the orders brought out by the spot goods meet two conditions: first, the order exists in the order set already designed in the solution; second, the order's under-combination quantity is less than 1; then, sort the orders in the optional order set by width in descending order, and sort those with the same width by length in descending order. Step 3.4.10.2: Record the total length of the current design scheme, select the first order from the set of available orders, calculate the remaining length based on the maximum limit length of the current design scheme and the total length of the current design scheme; calculate the upper limit of the design quantity of the order based on the remaining length and the length of the current order. Step 3.4.10.3: Bring out the maximum quantity of goods, then check the design conditions of the current solution. If successful, terminate the loop; otherwise, erase the current operation, decrement the quantity of goods brought out by 1, and continue to try designing until the design is successful or the quantity of goods brought out is 0. Step 3.4.10.4: Update the current solution attributes and select the next order to continue trying to bring out the spot goods, until all orders in the order candidate set have been traversed. Step 3.4.11: Execute step 3.4.3 until there are no orders left to be traversed in the second-level order set within the first-level order set; Step 3.4.12: Execute step 3.4.2 until there are no more first-level order sets that have not been traversed; Step 3.4.13: Execute step 3.4.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0; The design success conditions for the yield-billet-spot product are the same as those for the design success conditions for the spot-billet-yield product; The design failure conditions for the yield-billet-spot are the same as those for the spot-billet-yield design. The design feasibility conditions for the yield-billet-spot are the same as those for the spot-billet-yield design. Step 3.5: Execute a heuristic algorithm with the objective priority of "highest yield - minimum number of spot items - maximum billet weight", which is simply referred to as yield-spot-billet; Step 3.5.1: Select a blank from the blank set as the blank to be designed; Step 3.5.2: Select a primary order set from the order set. The number of secondary order sets in the primary order set cannot be zero. Step 3.5.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-patch quantity; Step 3.5.4: Select an order from the secondary order pool and designate it as the master order; the master order confirms the thickness, width, and length range of the design scheme; the lower limit of the length range is determined by the lower limit of the billet length; the upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value. Step 3.5.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order; Step 3.5.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order; compare the maximum design quantity with the under-packing quantity of the order, and take the minimum of the two as the actual limit of the design quantity of the order; Step 3.5.7: Select the actual upper limit of the design quantity for the design order; Step 3.5.8: Design the order according to the design quantity in the design order; Step 3.5.9: Then, check whether the design requirements are met based on the design success conditions of yield rate - spot goods - sub-blank; if the design success requirements are met, update the attributes of the order, record the design scheme, and execute step 3.5.4; if the design failure conditions of yield rate - spot goods - sub-blank are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of yield rate - spot goods - sub-blank are met; if the design feasibility conditions of yield rate - spot goods - sub-blank are met, maintain the current design status, record each design attribute, and execute step 3.5.5; if all orders in the secondary order set have been traversed, execute step 3.5.11; Step 3.5.10: If the design scheme always meets the design feasibility conditions of yield rate-spot goods-sub-bills, but cannot meet the design success conditions of yield rate-spot goods-sub-bills, then execute the "minimum spot quantity spot carry-out strategy" and add spot products; The minimum spot quantity spot take-out strategy is as follows: if the current scheme does not meet the design success conditions of yield-spot-sub-billet, but meets the design feasibility conditions of yield-spot-sub-billet, then try to take out spot; Step 3.5.10.1: Construct the optional order set brought out by the spot goods; the orders brought out by the spot goods meet two conditions: first, the order exists in the order set already designed in the solution; second, the order's shortfall is less than 1; then, sort the orders in the optional order set by unit weight in descending order, and sort those with the same unit weight by length in descending order. Step 3.5.10.2: Record the total length of the current design scheme, select the first order from the set of available orders, calculate the remaining length based on the maximum limit length of the current design scheme and the total length of the current design scheme; calculate the upper limit of the design quantity of the order based on the remaining length and the length of the current order. Step 3.5.10.3: Design the order onto the solution with the maximum design quantity, update the current solution length, check if the design success condition is met. If it is met, terminate the design. If it is not met, remove the current design operation, decrement the design quantity of the order by 1, design it onto the solution again, and re-check the design success condition until the design quantity of the order drops to 0. Step 3.5.10.4: Return to step 3.5.2 and select another order from the order set until all orders in the order set have been traversed. Step 3.5.10.5: Determine the success conditions for the design. If successful, record the attributes of this design; otherwise, release all orders related to this design. Step 3.5.11: Execute step 3.5.3 until there are no orders left to be traversed in the second-level order set within the first-level order set; Step 3.5.12: Execute step 3.5.2 until there are no more first-level order sets that have not been traversed; Step 3.5.13: Execute step 3.5.1 until all blanks have been traversed or the number of missing sub-boards in all orders is equal to 0; Step 3.6: Execute a heuristic algorithm with the objective priority of "maximum billet weight - highest yield - minimum quantity of spot goods", which is simply referred to as billet-yield-spot goods; Step 3.6.1: Select the first billet in the billet set as the design billet; Step 3.6.2: Select a primary order set from the order set. The number of secondary order sets in the primary order set cannot be zero. Step 3.6.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-patch quantity; Step 3.6.4: Select an order from the secondary order set and designate this order as the master order; the master order confirms the thickness, width and length range of the design scheme; the lower limit of the length range is determined by the lower limit of the billet length; the upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length and the upper limit of the rolling length, i.e. the limit length value; Step 3.6.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order; Step 3.6.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order; compare the maximum design quantity with the under-packing quantity of the order, and take the minimum of the two as the actual limit of the design quantity of the order; Step 3.6.7: Select the actual upper limit of the design quantity for the design order; Step 3.6.8: Design the order according to the design quantity in the design order; Step 3.6.9: Then, check whether the design requirements are met based on the design success conditions of billet-yield-spot goods; if the design success requirements are met, update the attributes of the order, record the design scheme, and execute step 3.6.4; if the design failure conditions of billet-yield-spot goods are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of billet-yield-spot goods are met; if the design feasibility conditions of billet-yield-spot goods are met, maintain the current design status, record each design attribute, and execute step 3.6.5; if all orders in the secondary order set have been traversed, execute step 3.6.11; Step 3.6.10: If the design scheme consistently meets the design feasibility conditions of billet-yield-spot availability but fails to meet the design success conditions of billet-yield-spot availability, then execute the "bill replacement strategy". The billet replacement strategy is as follows: the index values ​​of all billets in the candidate billet set are greater than the index of the billet corresponding to the current scheme; starting from the billet with the smallest index value in the candidate billet set, the billets in the candidate billet set are successively subtracted from the current design scheme for design; if the design success condition is met, the design state of the design scheme is set and the billet replacement process is terminated; otherwise, the billet traversal continues until all billets in the billet candidate set have been traversed, and the billet replacement process is terminated. Step 3.6.11: If the design success conditions of sub-blank-yield-spot stock cannot be met in step 3.6.10, then execute the "maximum width spot stock carry-out strategy" to add spot sub-boards; Step 3.6.11: Execute step 3.6.3 until there are no orders left to be traversed in the second-level order set within the first-level order set; Step 3.6.12: Execute step 3.6.2 until there are no more first-level order sets that have not been traversed; Step 3.7: Execute a heuristic algorithm with the objective priority of "maximum billet weight - minimum quantity of spot goods - highest yield", which is simply referred to as billet-spot goods-yield. Step 3.7.1: Select the first billet in the billet set as the design billet; Step 3.7.2: Select a primary order set from the order set. The number of secondary order sets in the primary order set cannot be zero. Step 3.7.3: Select a secondary order set from the primary order set, sort the orders in the secondary order set in descending order of width, sort the orders with the same width in descending order of length, and sort the orders with the same width and length in descending order of order under-patch quantity; Step 3.7.4: Select an order from the secondary order set and designate this order as the master order; the master order confirms the thickness, width, and length range of the design scheme; the lower limit of the length range is determined by the lower limit of the billet length; the upper limit of the length range is the minimum value of the upper limit of the billet length, the upper limit of the manually given length, and the upper limit of the rolling length, i.e., the limit length value; Step 3.7.5: If there are still orders in the secondary order set with a non-zero number of missing sub-boards, select the order that is sorted after the main order from the secondary order set as the design order; Step 3.7.6: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order; compare the maximum design quantity with the under-packing quantity of the order, and take the minimum of the two as the actual limit of the design quantity of the order; Step 3.7.7: Select the actual upper limit of the design quantity for the design order; Step 3.7.8: Design the order according to the design quantity in the design order; Step 3.7.9: Then, check whether the design requirements are met based on the design success conditions of billet-spot goods-yield rate; if the design success requirements are met, update the attributes of the order, record the design scheme, and execute step 3.7.4; if the design failure conditions of billet-yield rate-spot goods are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions of billet-yield rate-spot goods are met; if the design feasibility conditions of billet-yield rate-spot goods are met, maintain the current design status, record each design attribute, and execute step 3.7.5; if all orders in the secondary order set have been traversed, execute step 3.7.11; Step 3.7.10: If the design scheme always meets the design feasibility conditions of billet-yield-spot, but cannot meet the design success conditions of billet-yield-spot, then execute the "maximum width spot carry-out strategy". Step 3.7.11: Execute step 3.7.3 until there are no orders left to be traversed in the second-level order set within the first-level order set; Step 3.7.12: Execute step 3.7.2 until there are no more first-level order sets that have not been traversed; Step 3.7.13: Implement the strategy to eliminate spot inventory; The strategy for eliminating spot inventory is as follows: if the current solution has spot inventory, then attempt to eliminate the spot inventory; Step 3.7.13.1: Construct a "spot optimization candidate solution set" based on the original optimization solution set; the original optimization solution set refers to the design solution set generated in steps 3.7.1-3.7.12; the "spot optimization candidate solution set" includes all design solutions with spot availability; Step 3.7.13.2: Remove the design schemes from the "Spot Optimization Candidate Scheme Set" from the original design scheme set; Step 3.7.13.3: Divide the candidate solutions in the candidate spot optimization solution set into multiple subsets according to their thickness; Step 3.7.13.4: Select one subset from multiple subsets and release all orders contained in the schemes therein; delete the spot orders from the released orders, and the remaining released orders constitute the "redesigned order set"; Step 3.7.13.5: Use the heuristic method of "maximum billet weight - highest yield - minimum number of outgoing items" to redesign the orders in the "redesign order set" to obtain a new set of design schemes; Step 3.7.13.6: Merge the set of new design schemes into the set of original design schemes.

5. The multi-objective-based medium-thick plate billet design method according to claim 1, characterized in that, Step 4 specifically includes the following steps: Step 4.1: Divide the set of cross-cutting schemes; divide all design schemes included in the billet design plan according to thickness, and for the same thickness, divide them according to steel code; establish a set of cross-cutting schemes for design schemes with the same thickness and the same steel code; sort the schemes in the set of cross-cutting schemes according to the billet index from largest to smallest; the schemes ranked higher are the schemes that need to be improved, and the schemes ranked lower are the basis for improvement. Step 4.2: Select the solutions that need to be used as the basis for improvement in ascending order; Step 4.3: Filter the solutions in the cross-cutting solution set that have the same sub-board as the solution to be improved, and sort these solutions down as the preferred solutions to be improved; Step 4.4: Select billets with a billet index greater than the billet index designed in the improved basis scheme to construct a reduced-order billet set for the improved basis scheme and create a candidate set for order adjustment; Step 4.5: Subtract sub-boards from the improved scheme, reducing the number of sub-boards in the scheme sequentially, and checking the design success conditions of sub-bill-yield-spot stock. If the design success conditions of sub-bill-yield-spot stock are met, update the attributes of the current scheme, record the reduced orders, and add them to the order adjustment candidate set. If the design success conditions of sub-bill-yield-spot stock are not met, replace the billet and continue checking. Step 4.6: Release all design sub-boards of the improved solution and load them into the order adjustment candidate set; Step 4.7: Query the first-level and second-level order sets to which the orders in the current order adjustment candidate set belong; query the orders in the set, and if there is an order with a missing quantity greater than 0, add the status of that order to the candidate order set; Step 4.8: Construct an upgraded billet set for the improved solution, and fill the billet candidate set with billets whose indices are less than those of the billets belonging to the improved solution. Step 4.9: Adjust the order candidates; sort the orders in the set in descending order of width and length. Step 4.10: Select the billets from the upgraded billet set of the improved scheme in sequence to prepare for the design process. Step 4.11: Select a billet from the improved billet set in ascending order; Step 4.12: Select an order from the order adjustment candidate set as the master order, and determine the width, thickness, limit length, and steel code of the solution; Step 4.13: Select orders from the order adjustment candidate set as the orders to be attempted for billet assembly; Step 4.14: Calculate the maximum allowable design quantity for the design order based on the limit length value and the order length of the design order; compare the maximum design quantity with the under-packing quantity of the order, and take the minimum of the two as the actual limit of the design quantity of the order; Step 4.15: Select the actual upper limit of the design quantity for the design order; Step 4.16: Design the order according to the design quantity in the design order; Step 4.17: Then check whether the design requirements are met according to the design success conditions; if the design success requirements are met, update the attributes of the order, record the design scheme, and execute step 4.12; if the design failure conditions are met, reduce the design quantity of the design order until the design quantity is reduced to 0 or the design success conditions are met; if the design feasibility conditions are met, maintain the current design status, record each design attribute, and execute step 4.

13. Step 4.18: If the design scheme consistently meets the design feasibility conditions but fails to meet the design success conditions, then replace the billet; if replacing the billet still fails to meet the design success conditions, then execute the spot product carry-out strategy and add spot products. Step 4.19: Check if the current scheme is the same as the initial scheme. If they are the same, do not update the design flag.

6. The multi-objective-based medium-thick plate billet design method according to claim 1, characterized in that, Step 5 specifically includes the following steps: Step 5.1: Division of the variable scheme set: All design schemes included in the billet design plan are divided according to thickness, and for the same thickness, they are divided according to steel code; Step 5.2: Randomly select 3 schemes from the set of mutable schemes with the same thickness and steel code as the schemes to participate in the mutation; Step 5.3: Release all orders on the three selected schemes to form a candidate order set, and then sort the orders in the candidate order set in descending order of width and length; Step 5.4: Select one billet from the billet set that has not been selected before; Step 5.5: Select an order from the candidate order set as the master order, determine the thickness, width, limit length and steel code of the design scheme based on the attributes of the master order and the selected billet, and calculate the upper limit of the design quantity of the master order; Step 5.6: Select an order from the candidate order set, and calculate the maximum number of designs for the selected order based on the remaining length of the current design scheme and the length of the selected order; Step 5.7: Design orders according to the maximum design quantity limit, update the current design scheme, check the design success conditions. If the design is successful, record the design scheme and terminate the design. If the design failure conditions are met, release all current orders and execute step 5.

5. If the design feasibility conditions are met, retain the current design status and execute step 5.

6. Step 5.8: If the design scheme still meets the design feasibility conditions after all orders in the candidate order set have been traversed, then keep the current design scheme unchanged, change the blank material designed for the design scheme, and check the design judgment conditions at the same time; if the design success conditions are met, then record the design scheme; otherwise, continue to change the blank material until the design success conditions are met. Step 5.9: If the design success conditions cannot be met after changing the billet, then implement the spot carry-out strategy.

7. The multi-objective-based medium-thick plate billet design method according to claim 1, characterized in that, Step 6 specifically includes the following steps: Step 6.1: Select solutions from the initial solution set, the crossover solution set, and the mutation solution set in turn to try to enter the pool, until all solutions in the three sets have been selected; Step 6.2: If the proposed solution meets the replacement criteria, replace one solution in the solution pool and then proceed to step 6.1; if the replacement criteria are not met, proceed to step 6.

3. Step 6.3: If the proposed solution meets the discard criteria, it cannot be added to the pool, and step 6.1 is executed; otherwise, step 6.4 is executed. Step 6.4: If the proposed solution meets the pooling criteria, compare whether the solution already exists in the pool. If it does not exist, add the solution directly to the pool and proceed to step 6.

1. If it exists, do not add the solution to the pool and proceed to step 6.

1.

8. The multi-objective-based medium-thick plate billet design method according to claim 1, characterized in that, Step 7 specifically includes the following steps: Step 7.1: Implement the minimum order balance strategy: Check the solutions in the solution pool. If a solution contains orders that have not been fully utilized, delete that solution. Step 7.2: Implement the minimum available stock strategy: Check the solutions in the solution pool. If a solution has available stock, delete the solution until the number of solutions in the solution pool equals the limit number of solutions. Step 7.3: Implement the strategy of minimizing the number of medium and heavy plates: Sort the schemes in the scheme pool in descending order of the number of medium and heavy plates designed in the scheme; delete the schemes in the scheme pool with a large number of medium and heavy plates designed in the scheme pool until the number of medium and heavy plates designed in the scheme pool is equal to the minimum number of medium and heavy plates or the number of schemes in the scheme pool is equal to the limit number of schemes. Step 7.4: Implement a uniform billet distribution strategy: Calculate the difference between the maximum and minimum billet numbers of the design schemes in the scheme pool, sort them in descending order according to the difference, and then delete the schemes in the scheme pool in sequence until the billet difference of all schemes in the scheme pool is equal to the minimum billet difference or the number of schemes in the scheme pool is equal to the limit number of schemes. Step 7.5: Execute the minimum billet average strategy: Calculate the average billet serial number of the design schemes in the scheme pool, and sort all schemes in descending order according to the average; then delete the schemes in the scheme pool in sequence until the average billet serial number of all schemes in the scheme pool is equal to the minimum average billet serial number or the number of schemes in the scheme pool is equal to the limit number of schemes. Step 7.6: Implement the minimum billet quantity strategy: Count the total number of billets in the scheme pool and sort them in descending order according to the total number of billets; then delete the schemes in the scheme pool in sequence until the total number of billets for all schemes in the scheme pool is equal to the minimum total number of billets or the number of schemes in the scheme pool is equal to the limit number of schemes.

9. The multi-objective-based medium-thick plate billet design method according to claim 1, characterized in that, Step 8 specifically includes the following steps: Step 8.1: Execute the sub-board optimization strategy to optimize the distribution of sub-boards in each scheme, so that sub-boards with the same order number are distributed on the same medium-thick plate. That is, query the medium-thick plates in pairs, medium-thick plate A and medium-thick plate B in turn. If there are two orders on medium-thick plate A, namely order a and order b, where order a and order b have the same specifications and the same order number as order c on medium-thick plate B, and the number of sub-boards of a is not less than the number of sub-boards of c, then swap the same number of sub-boards of order c on medium-thick plate B and order b on medium-thick plate A. Step 8.2: Implement the minimum combined rolling strategy: count the number of combined rolled slabs in the scheme pool and sort them in descending order of the number of combined rolled slabs; then delete the schemes in the scheme pool in sequence until the number of combined rolled slabs of all schemes in the scheme pool is equal to the minimum number of combined rolled slabs or the number of schemes in the scheme pool is equal to the limit number of schemes.

Images