A multi-flow online width-adjustable continuous casting machine multi-steel grade group pouring method

Abstract

This invention relates to a multi-strand online width adjustment method for continuous casting machines with multiple steel grades, belonging to the technical field of continuous casting casting schedule development in steelmaking. The method involves clustering the heats to be produced into Lots based on steel grade and target width. All Lots are rearranged using a sorting model to generate a Lot sequence that minimizes the total number of tundishes used, the number of mixed steel grades, and the order width adjustment cost during continuous casting. Subsequently, based on a two-stage width optimization model, the target widths of different Lots are adjusted to the allowable width adjustment range of the continuous casting machine, and Lots that meet the continuous casting conditions are merged into casting segments. Finally, based on tundish life constraints, the generated casting segments are cut to generate a complete casting plan, clarifying the affiliation of each heat, ultimately producing a complete and feasible casting schedule. Since the width still meets the order requirements, the matching degree between slabs and orders can be significantly improved, reducing order cancellations caused by specification incompatibility.

Description

Technical Field

[0001] This invention relates to the technical field of continuous casting sequence planning in steelmaking, and particularly to a method for group casting of multiple steel grades in a continuous casting machine with multi-strand online width adjustment. Background Technology

[0002] When formulating casting plans, the premise is to meet order and product quality requirements. The use of furnaces, castings and tundishes should be organized in a reasonable manner so that the continuous casting machine can achieve the longest possible continuous casting while ensuring product quality, reducing the number of tundish replacements and equipment adjustments, and improving capacity utilization and production stability.

[0003] The shortcomings of existing technologies can be summarized as follows:

[0004] Insufficient width optimization: In actual production, the specifications required by orders are not fixed, but rather range-based. Most methods do not make full use of the adjustable range of slab width, and cannot fully utilize the adjustable width crystallizer to design as many specifications as possible that meet the order requirements while maintaining process feasibility.

[0005] Limited applicability: Existing methods are mostly designed for specific scenarios (such as two-flow different widths or continuous casting of a single steel grade), and their applicability is limited in complex production of multiple steel grades and specifications.

[0006] Insufficient comprehensiveness: It lacks a complete solution and is difficult to simultaneously address production continuity, equipment constraints, and cost control.

[0007] The solution is inefficient: existing methods treat the casting schedule as a whole, resulting in long solution times. This is especially problematic when dealing with large-scale projects with numerous constraints, failing to meet the needs of practical applications.

[0008] In summary, existing technologies have failed to fully utilize the adjustable range of slab width, resulting in low order fulfillment rates, insufficient casting continuity, and high production costs. Summary of the Invention

[0009] Based on the above analysis, the present invention aims to provide a multi-strand online width-adjustable continuous casting machine multi-steel grade group casting method to solve at least one of the problems of existing technology failing to fully utilize the adjustable range of slab width, resulting in low order hanging rate, insufficient casting continuity and high production cost.

[0010] On the one hand, the present invention provides a method for casting multiple steel grades in a continuous casting machine with multi-flow online width adjustment, comprising the following steps: S1: Lot generation. Based on the steel composition, process conditions and target width, the heats to be produced are clustered into a series of Lots with high internal width continuity and the same steel grade. S2: Lot sorting, based on the sorting model, rearranges all Lots, taking into account the continuous casting rules of steel grades, the range of width adjustment and the life constraints of the tundish, and generates the Lot sequence that minimizes the total number of tundishes used, the number of mixed steel grades, and the cost of order width adjustment during the continuous casting process; S3: Lot merging and width optimization. Based on a two-stage width optimization model, the target width of different Lots is adjusted to the width adjustment range allowed by the continuous casting machine, and Lots that meet the continuous casting conditions are merged into a casting segment. S4: Based on the tundish lifespan constraint, cut the casting segment generated in S3 to generate a complete casting scheme, clarify the affiliation of each furnace, and finally generate a complete and feasible casting plan.

[0011] Further, in step S1, the Lot generation step includes: Heat batch data collection and preprocessing; the data includes steel grade code, chemical composition range, metallurgical process conditions and order slab size, and then the data is cleaned and standardized. Preliminary sorting of the pieces in each heat batch; preliminary sorting of all heat batches according to their target slab width; Aggregation based on steel compatibility; based on the initial sorting, clustering is performed according to the compatibility of steel composition and metallurgical process; Width continuity check and segment splitting; after obtaining the initial grouping, further verify whether the width difference within and between segments meets the width adjustment capability of the continuous casting machine, and further split the clustered results into Lots; Clustering results are output.

[0012] Furthermore, in the polymerization step based on steel compatibility, batches meeting the following conditions need to be grouped into the same Lot: Chemical composition compatibility, metallurgical process condition compatibility, and width continuity are considered. The heat is gradually divided into several Lots, and the steel grades and process conditions within each segment are relatively similar.

[0013] Furthermore, in the Lot sorting step, the sorting of Lots is affected by multiple constraints and objectives, including metallurgical process constraints, width variation constraints, production efficiency constraints, and batch production characteristic constraints.

[0014] Furthermore, in the Lot sorting step, a heuristic rule combined with an optimization algorithm is adopted. A mathematical model is established and solved using an optimization algorithm or solver. The mathematical model is as follows: The objective function consists of the continuous casting cost between steel grades, the tundish switching cost, and the width adjustment cost: ; Sub-loops eliminate constraints and prevent sub-loops from forming between Lots:

[0015]

[0016]

[0017] Lot unique predecessor and successor constraints:

[0018]

[0019] Intermediate package constraints:

[0020]

[0021] Intermediate package usage quantity calculation:

[0022] Widening constraints:

[0023] Calculation of lenient penalties:

[0024] Variable value constraints:

[0025]

[0026]

[0027]

[0028] ; Lot With Lot Switching costs; Represents the total number of Lots; Indicates the maximum lifespan of a single intermediate package; Lot Processing time; This indicates the threshold values ​​for upward and downward sizing of the crystallizer; Lot With Lot Whether or not a leniency adjustment has occurred; Lot Lot followed closely behind The time is 1; This represents the switching cost weighting coefficient; This indicates the use of intermediate package cost coefficients; Indicates the adjustment cost coefficient; Lot The order; Lot Assigned to the The value is 1 when there is one intermediate packet; Represents Lot The maximum and minimum widths.

[0029] Furthermore, in step S3, Lot merging and width optimization include: Merge the casting segments; take out the sorted Lots one by one as the initial segments: initialize the first Lot as segment 1; starting from the second Lot, check the merge condition with the previous segment one by one: if the merge condition is passed, merge the Lot into the current segment; otherwise, start a new segment. Merging condition determination; based on the matching between the widths of adjacent slabs; Width optimization is performed using a two-stage width optimization model. Update the merge results; after completing the width optimization process, the planned width of the slab has been redefined, and the Lots with the determined widths need to be merged.

[0030] Furthermore, the first-stage width optimization model: The objective function is to maximize the number of slabs that can be continuously cast:

[0031] The main constraint is to relax the feasibility constraints:

[0032]

[0033]

[0034] Decision variables:

[0035] .

[0036] Furthermore, the second-stage width optimization model: The objective function, while ensuring continuity, maximizes the number of width differences and minimizes the difference from the target width.

[0037] Phase 1 relaxation of feasible constraints:

[0038]

[0039]

[0040] Determine whether to widen the width:

[0041]

[0042] First-stage optimal value constraint:

[0043] Newly added decision variables:

[0044]

[0045]

[0046] The optimized slab width is obtained by solving the problem using an optimization algorithm or solver.

[0047] Furthermore, in step S4, the generation of the casting schedule includes: Feasibility assessment for continuous casting: For the optimized Lot sequence, first determine whether it can be continuously cast in the same casting cycle; if there is a steel grade conflict between some Lots, it must be broken at that point and a new casting cycle must be started. Tundish life constraint: During continuous casting, the life of the tundish is usually limited by the casting time or the number of casting cycles. If the cumulative production time of the current casting exceeds the tundish life limit, the current casting must be forcibly terminated at that point and a new tundish must be replaced. Subsequent Lots will then proceed to the next casting cycle. Heat allocation; After determining the boundaries of each casting batch, the specific heats contained in the Lot need to be allocated to the corresponding casting batch one by one; The plan outputs the following: the start and end lot numbers for each casting cycle, the heat allocation, the corresponding steel grade information, the target width range, and whether the tundish needs to be replaced.

[0048] Furthermore, the metallurgical process conditions include the superheat control range, the cooling intensity of the secondary cooling zone, and the allowable drawing speed window.

[0049] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects: 1. This invention fully considers order requirements and slab width matching during the Lot generation and Lot sorting stages, and expands the continuously castable segments through width optimization and merging strategies. Since the width still meets the order requirements, it can significantly improve the matching degree between slab and order and reduce the phenomenon of order dropout due to specification incompatibility.

[0050] 2. By using a two-stage width optimization model and judging the feasibility of continuous casting, this invention can extend the length of continuous casting as much as possible within the width adjustment range, reducing frequent width adjustment operations. At the same time, by introducing a tundish life constraint in the casting cycle planning stage, the number of tundish replacements can be effectively controlled, achieving "minimum use of tundish", thereby reducing consumption costs and downtime.

[0051] 3. The optimized process of this invention does not depend on specific steel grades or production scenarios. Instead, it achieves flexible casting of multiple steel grades and specifications on multi-flow casting machines through the comprehensive application of clustering, sorting, and width optimization. This feature significantly enhances the applicability of the method and can meet the needs of different production environments and diverse orders.

[0052] 4. This invention decomposes the complex casting optimization problem into sub-problems such as Lot generation, Lot sorting, Lot merging, and width optimization, gradually reducing the problem size and difficulty of solving. While ensuring optimization quality, it effectively improves operational efficiency, enabling the generated scheduling scheme to quickly respond to on-site needs and possessing good operability and real-time performance.

[0053] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained from what is particularly pointed out in the description and drawings. Attached Figure Description

[0054] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0055] Figure 1 This is the overall flowchart. Detailed Implementation

[0056] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which constitute a part of the present invention and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0057] When formulating casting plans, the premise is to meet order and product quality requirements. The use of furnaces, castings and tundishes should be organized in a reasonable manner so that the continuous casting machine can achieve the longest possible continuous casting while ensuring product quality, reducing the number of tundish replacements and equipment adjustments, and improving capacity utilization and production stability.

[0058] The shortcomings of existing technologies can be summarized as follows: Insufficient width optimization: In actual production, the specifications required by orders are not fixed, but rather range-based. Most methods do not make full use of the adjustable range of slab width, and cannot fully utilize the adjustable width crystallizer to design as many specifications as possible that meet the order requirements while maintaining process feasibility.

[0059] Limited applicability: Existing methods are mostly designed for specific scenarios and have limited applicability in complex production processes involving multiple steel types and specifications.

[0060] Insufficient comprehensiveness: It lacks a complete solution and is difficult to simultaneously address production continuity, equipment constraints, and cost control.

[0061] The solution is inefficient: existing methods treat the casting schedule as a whole, resulting in long solution times. This is especially problematic when dealing with large-scale projects with numerous constraints, failing to meet the needs of practical applications.

[0062] In summary, existing technologies have failed to fully utilize the adjustable range of slab width, resulting in low order fulfillment rates, insufficient casting continuity, and high production costs.

[0063] Therefore, the present invention provides a method for casting multiple steel grades in a continuous casting machine with multi-strand online width adjustment, comprising the following steps: S1: Lot generation. Based on the steel composition, process conditions and target width, the heats to be produced are clustered into a series of Lots with high internal width continuity and the same steel grade. S2: Lot sorting, based on the sorting model, rearranges all Lots, taking into account the continuous casting rules of steel grades, the range of width adjustment and the life constraints of the tundish, and generates the Lot sequence that minimizes the total number of tundishes used, the number of mixed steel grades, and the cost of order width adjustment during the continuous casting process; S3: Lot merging and width optimization. Based on a two-stage width optimization model, the target width of different Lots is adjusted to the width adjustment range allowed by the continuous casting machine, and Lots that meet the continuous casting conditions are merged into a casting segment. S4: Based on the tundish lifespan constraint, cut the casting segment generated in S3 to generate a complete casting scheme, clarify the affiliation of each furnace, and finally generate a complete and feasible casting plan.

[0064] Compared with existing technologies, this invention fully considers order requirements and slab width matching in the Lot generation and Lot sorting stages. It expands the continuously castable segments through width optimization and merging strategies. Since the width still meets order requirements, it significantly improves the matching degree between slabs and orders, reducing order cancellations due to specification incompatibility. This invention decomposes the complex casting optimization problem into sub-problems such as Lot generation, Lot sorting, Lot merging, and width optimization, gradually reducing the problem scale and solution difficulty. While ensuring optimization quality, it effectively improves operational efficiency, enabling the generated scheduling scheme to quickly respond to on-site needs, and possessing good operability and real-time performance.

[0065] Specifically, in step S1, the Lot generation step includes: Heat batch data collection and preprocessing; the data includes steel grade code, chemical composition range, metallurgical process conditions and order slab size, and then the data is cleaned and standardized. Preliminary sorting of the pieces in each heat batch; preliminary sorting of all heat batches according to their target slab width; Aggregation based on steel compatibility; based on the initial sorting, clustering is performed according to steel composition and metallurgical process conditions; Width continuity check and segment splitting; after obtaining the initial grouping, further verify whether the width difference within and between segments meets the width adjustment capability of the continuous casting machine, and further split the clustered results into Lots; Clustering results output.

[0066] It should be noted that Lot generation is the first crucial step. Its goal is to scientifically organize and divide heat batches so that subsequent sorting, merging, and width optimization can be carried out with greater compatibility and lower adjustment costs. This section comprehensively considers multiple dimensions of factors, such as heat batch composition compatibility, metallurgical process conditions, and width continuity, to create more opportunities for continuous casting of heat batches with different steel grades and specifications, while ensuring process feasibility.

[0067] It should be noted that in this invention, the metallurgical process conditions include the superheat control range, the cooling intensity of the secondary cooling zone, and the allowable drawing speed window; while the dimensions of the order slab are mainly the width and thickness, with the thickness generally fixed and the width adjustable within a certain range. These data are mainly derived from the order plan.

[0068] It should be noted that in the initial sorting step of each heat, all heats are sorted according to their target slab width. Although this step does not directly generate clustering results, it makes the distribution of the width parameter more orderly, facilitating subsequent clustering and merging. When heats are arranged by width, the distance between adjacent heats is often within tens of millimeters, thus increasing the likelihood that they will be assigned to the same cluster or segment.

[0069] Specifically, in the polymerization step based on steel compatibility, the following conditions need to be met to divide the batches into the same Lot: it is necessary to combine chemical composition compatibility, metallurgical process condition compatibility and width continuity to gradually divide the heat batch into several Lots, with the steel grades and process conditions within each segment being similar.

[0070] It should be noted that chemical composition compatibility refers to significant differences in metallurgical properties; direct continuous casting of such steel grades can easily lead to quality fluctuations, therefore, adjacent steel grades should be avoided. Metallurgical process condition compatibility refers to the significant differences in requirements for secondary cooling water distribution and casting speed between different steel grades; combining heats with similar process parameters can reduce the number of adjustments. Width continuity requires that the width difference of slabs within the same group does not exceed the adjustable range of the continuous casting machine; for example, the width difference between any two heats within the group should not exceed 20mm.

[0071] In the width continuity check and segment splitting steps, after obtaining the initial grouping, it is necessary to further verify whether the width difference within and between segments meets the width adjustment capability of the continuous casting machine, and further split the clustered results into Lots. For example, if the adjustable width range of a continuous casting machine is: +50mm upward and -30mm downward, then the width difference between two adjacent heats must not exceed this range. If the heat span within a segment is too large (for example, there is a heat of 1100mm and a heat of 1200mm), then segment splitting must be performed, dividing the large-span segment into smaller sub-segments. The following principles should be followed when splitting: ensure that the steel grade within each sub-Lot still meets the composition and process compatibility; prioritize maintaining the continuity of adjacent heats to avoid drastically disrupting the order; if the conditions cannot be met by a single split, iterative splitting should be used, that is, gradually reducing the width span within the Lot until the width adjustment capability of the casting machine is met.

[0072] In the clustering result output step, a series of appropriately sized Lots are ultimately obtained, each Lot containing several heats with high consistency in steel grade characteristics and process conditions. The output clustering results will serve as input for subsequent Lot sorting, width optimization, and casting schedule generation, laying the foundation for achieving high order fulfillment rates, minimal intermediate reels, and lowest production costs.

[0073] Specifically, in the Lot sorting step, the sorting of Lots is affected by multiple constraints and objectives, including metallurgical process constraints, width variation constraints, production efficiency constraints, and batch production characteristic constraints.

[0074] It should be noted that after generating the Lot, the production units in the clustering results—that is, the Lots (usually composed of several heats)—need to be sorted appropriately. The purpose of Lot sorting is to generate the Lot sequence that minimizes the total number of tundishes used, the amount of mixed steel grades poured, and the cost of order width adjustments during continuous casting, while ensuring process constraints. This reduces energy consumption and efficiency losses caused by frequent changes in steel grades or widths, and lays the foundation for subsequent Lot merging and heat plan generation.

[0075] In actual production, lot sequencing is influenced by various constraints and objectives. These constraints mainly include: Process continuity requirements: The steel grades of adjacent lots need to be as compatible as possible to avoid production interruptions or quality issues caused by significant differences in the physicochemical properties of the steel grades. Width variation constraints: While casting machines have a certain range of width adjustment capabilities, this range is limited, and large-scale width adjustments can reduce machine lifespan; therefore, width spans should be minimized during sequencing. Production efficiency requirements: Reasonable sequencing can reduce unnecessary downtime and tundish changes, improving the overall efficiency of continuous casting. Batch production characteristics: Some orders or batches require continuous production in groups to ensure product quality consistency and seamless transitions to subsequent rolling processes.

[0076] Specifically, in the Lot sorting step, heuristic rules are combined with optimization algorithms. A mathematical model is established and solved using optimization algorithms or solvers. The mathematical model is as follows: The objective function consists of the continuous casting cost between steel grades, the tundish switching cost, and the width adjustment cost: ; Sub-loops eliminate constraints and prevent sub-loops from forming between Lots:

[0077]

[0078]

[0079] Lot unique predecessor and successor constraints:

[0080]

[0081] Intermediate package constraints:

[0082]

[0083] Intermediate package usage quantity calculation:

[0084] Widening constraints:

[0085] Calculation of lenient penalties:

[0086] Variable value constraints:

[0087]

[0088]

[0089]

[0090] ; Lot With Lot Switching costs; Represents the total number of Lots; Indicates the maximum lifespan of a single intermediate package; Lot Processing time; This indicates the threshold values ​​for upward and downward sizing of the crystallizer; Lot With Lot Whether or not a leniency adjustment has occurred; Lot Lot followed closely behind The time is 1; This represents the switching cost weighting coefficient; This indicates the use of intermediate package cost coefficients; Indicates the adjustment cost coefficient; Lot The order; Lot Assigned to the The value is 1 when there is one intermediate packet; Represents Lot The maximum and minimum widths.

[0091] It should be noted that the heuristic rules can be summarized as follows: First, place multiple lots within the same steel grade category as adjacent as possible to reduce the frequency of steel grade switching. Second, within lots of the same or similar steel grades, perform secondary sorting based on the width range of the slab. Lots with similar widths are grouped together as much as possible to reduce the magnitude of width adjustments. Third, considering production constraints (such as the number of heat runs and planned production time), make special arrangements for certain lots that must be separated or produced ahead of schedule. Finally, use the obtained lot sorting as the initial solution input into the subsequent model.

[0092] Specifically, in step S3, Lot merging and width optimization include: Merge the casting segments; take out the sorted Lots one by one as the initial segments: initialize the first Lot as segment 1; starting from the second Lot, check the merge condition with the previous segment one by one: if the merge condition is passed, merge the Lot into the current segment; otherwise, start a new segment. Merging condition determination; based on the matching between the widths of adjacent slabs; Width optimization is performed using a two-stage width optimization model. Update the merge results; after completing the width optimization process, the planned width of the slab has been redefined, and the Lots with the determined widths need to be merged.

[0093] It should be noted that after completing Lot generation and sorting, this invention further optimizes continuous casting and order matching between heats through a Lot merging and width optimization model. Under the constraints of process and equipment, the slab width is reasonably adjusted to allow adjacent Lots to be continuously merged, thereby extending the continuous casting segment, improving production continuity, order matching rate, and reducing the use of tundishes and overall costs.

[0094] In the merging condition determination step, the specific operation is as follows: For the end slab of the current segment and the starting slab of the next Lot, comparison samples are taken from the production sequence of each continuous casting machine flow; within the allowable width adjustment range, if the difference between the maximum width of the end slab of the previous segment and the minimum width of the starting slab of the next segment does not exceed the set threshold, it is determined that it can be merged; if the width difference exceeds the threshold, the width optimization processing module is called to make adjustments; if the merging condition is still not met after width optimization, the segmentation is maintained and a new segment is started.

[0095] Specifically, the first-stage width optimization model: The objective function is to maximize the number of slabs that can be continuously cast:

[0096] The main constraint is to relax the feasibility constraints:

[0097]

[0098]

[0099] Decision variables:

[0100] .

[0101] Specifically, the second-stage width optimization model: The objective function, while ensuring continuity, maximizes the number of width differences and minimizes the difference from the target width.

[0102] Phase 1 relaxation of feasible constraints:

[0103]

[0104]

[0105] Determine whether to widen the width:

[0106]

[0107] First-stage optimal value constraint:

[0108] The decision variables in the one-stage model will not be repeated here. The following are the newly added decision variables:

[0109]

[0110]

[0111] The optimized slab width is obtained by solving the problem using an optimization algorithm or solver.

[0112] Parameter description: in It is the width of the i-th slab, a continuous variable. These are the minimum and maximum widths of the i-th slab, respectively. This represents whether adjacent slabs meet the width adjustment requirements; it is a 0-1 variable. These are the thresholds for downward and upward sizing of the crystallizer. Representative slab Set the width. Indicates slab The deviation between the width and the target. Representative slab The target width. This indicates whether the width between adjacent slabs has changed; it is a 0-1 variable. This represents the optimal solution obtained in the first stage.

[0113] It should be noted that, in the width optimization section, to prevent excessive computational complexity of the model and the resulting difficulty in solving it, this invention introduces a two-stage width optimization model for solution. The primary objective of the first stage is to maximize the number of slabs that can be continuously cast, ensuring the continuous and stable operation of the casting machine and reducing the difficulty of solving the second-stage optimization model. The second-stage optimization model further determines the production width of the slab and addresses the width adjustment balance issue based on the first-stage optimization model.

[0114] It should be noted that in the step of updating and merging the results, the planned width of the slab has been re-determined after the width optimization process is completed, so it is necessary to merge the Lots with already determined widths. The core purpose of this step is to ensure that the final segment division of the casting is feasible in terms of process and specifications.

[0115] First, a width compatibility check is performed on each Lot sequence after width optimization. If two adjacent Lots are adjusted to the same or similar width range after optimization and both meet the casting machine's width adjustment range constraints, they are merged back into a continuous casting segment. If some Lots are still incompatible with the preceding and following Lots after width adjustment, or if the merged Lots exceed the actual width adjustment capability of the casting machine, then a new segment boundary needs to be set at that point.

[0116] After completing the above checks and processing, the system will number and update the attributes of all formed segments. Each segment needs to have clear process parameters, including: a unified target width range, the number of lots it contains, and the corresponding furnace range. This provides the basic input for subsequent generation of casting schedules. Through this mechanism, the slabs within the segments not only meet the process feasibility requirements but also achieve a high order fulfillment rate, reduce the number of tundish changes, and lower width adjustment and production costs.

[0117] Specifically, in step S4, generating the pouring schedule includes: Feasibility assessment for continuous casting: For the optimized Lot sequence, first determine whether it can be continuously cast in the same casting cycle; if there is a steel grade conflict between some Lots, it must be broken at that point and a new casting cycle must be started. Tundish life constraint: During continuous casting, the life of the tundish is usually limited by the casting time or the number of casting cycles. If the cumulative production time of the current casting exceeds the tundish life limit, the current casting must be forcibly terminated at that point and a new tundish must be replaced. Subsequent Lots will then proceed to the next casting cycle. Heat allocation; After determining the boundaries of each casting batch, the specific heats contained in the Lot need to be allocated to the corresponding casting batch one by one; The plan outputs the following: the start and end lot numbers for each casting cycle, the heat allocation, the corresponding steel grade information, the target width range, and whether the tundish needs to be replaced.

[0118] It should be noted that after merging, sorting, and adjusting the width of the Lots, a final casting schedule needs to be generated. The casting schedule is one of the outputs of this method and serves as a directly executable operational guide in actual production. The main task of this step is to allocate the optimized heats to various casting batches according to a predetermined order and actual production width, while comprehensively considering key process constraints such as tundish life and continuous casting limitations for different steel grades.

[0119] It should be noted that in the heat allocation step, after determining the boundaries of each casting cycle, the specific heats contained within the Lot need to be allocated to the corresponding casting cycles one by one. The following rules should be followed during allocation: ensure that the heat sequence is consistent with the previous optimization results to avoid artificial disruption; the target width of heats within the same casting cycle should maintain continuity and compatibility and should not exceed the optimized width range; under multi-flow continuous casting conditions, it is necessary to ensure that the flow is synchronized in time and sequence to avoid production imbalance caused by the premature end of a certain flow.

[0120] It should be noted that the final casting schedule, in the output step of the planning results, includes: the start and end lot numbers for each casting, the heat allocation, the corresponding steel grade information, the target width range, and whether the tundish needs to be replaced. This information not only provides clear operational guidance for on-site operators but also provides structured data support for subsequent integration with the production scheduling system.

[0121] The casting schedule generated through the above steps can achieve efficient combination casting of multiple steel grades while ensuring the feasibility of the production process, reducing the number of width and ladle changes, improving the utilization rate and production stability of the continuous casting machine, and achieving the production goals of high order rate, minimal use of tundishes, and lowest cost.

[0122] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for casting multiple steel grades in a continuous casting machine with multi-flow online width adjustment, characterized in that, Includes the following steps: S1: Lot generation. Based on the steel composition, process conditions and target width, the heats to be produced are clustered into a series of Lots with high internal width continuity and the same steel grade. S2: Lot sorting, based on the sorting model, rearranges all Lots, taking into account the continuous casting rules of steel grades, the range of width adjustment and the life constraints of the tundish, and generates the Lot sequence that minimizes the total number of tundishes used, the number of mixed steel grades, and the cost of order width adjustment during the continuous casting process; S3: Lot merging and width optimization. Based on a two-stage width optimization model, the target width of different Lots is adjusted to the width adjustment range allowed by the continuous casting machine, and Lots that meet the continuous casting conditions are merged into a casting segment. S4: Based on the tundish lifespan constraint, cut the casting segment generated in S3 to generate a complete casting scheme, clarify the affiliation of each furnace, and finally generate a complete and feasible casting plan.

2. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 1, characterized in that, In step S1, the Lot generation step includes: Heat batch data collection and preprocessing; the data includes steel grade code, chemical composition range, metallurgical process conditions and order slab size, and then the data is cleaned and standardized. Preliminary sorting of the pieces in each heat batch; preliminary sorting of all heat batches according to their target slab width; Aggregation based on steel compatibility; based on the initial sorting, clustering is performed according to the compatibility of steel composition and metallurgical process; Width continuity check and segment splitting; after obtaining the initial grouping, further verify whether the width difference within and between segments meets the width adjustment capability of the continuous casting machine, and further split the clustered results into Lots; Clustering results are output.

3. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 2, characterized in that, In the polymerization process based on steel compatibility, batches meeting the following conditions must be grouped into the same Lot: Chemical composition compatibility, metallurgical process condition compatibility, and width continuity are considered. The heat is gradually divided into several Lots, and the steel grades and process conditions within each segment are relatively similar.

4. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 1, characterized in that, In the Lot sorting step, the sorting of Lots is affected by multiple constraints and objectives, including metallurgical process constraints, width variation constraints, production efficiency constraints, and batch production characteristic constraints.

5. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 1, characterized in that, In the Lot sorting step, heuristic rules are combined with optimization algorithms. A mathematical model is established and solved using optimization algorithms or solvers. The mathematical model is as follows: The objective function consists of the continuous casting cost between steel grades, the tundish switching cost, and the width adjustment cost: ； Sub-loops eliminate constraints and prevent sub-loops from forming between Lots: Lot unique predecessor and successor constraints: Intermediate package constraints: Intermediate package usage quantity calculation: Widening constraints: Calculation of lenient penalties: Variable value constraints: ； Lot With Lot Switching costs; Represents the total number of Lots; Indicates the maximum lifespan of a single intermediate package; Lot Processing time; This indicates the threshold values ​​for upward and downward sizing of the crystallizer; Lot With Lot Whether or not a leniency adjustment has occurred; Lot Lot followed closely behind The time is 1; This represents the switching cost weighting coefficient; This indicates the use of intermediate package cost coefficients; Indicates the adjustment cost coefficient; Lot The order; Lot Assigned to the The value is 1 when there is one intermediate packet; Represents Lot The maximum and minimum widths.

6. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 1, characterized in that, In step S3, Lot merging and width optimization include: Merge the casting segments; take out the sorted Lots one by one as the initial segments: initialize the first Lot as segment 1; starting from the second Lot, check the merge condition with the previous segment one by one: if the merge condition is passed, merge the Lot into the current segment; otherwise, start a new segment. Merging condition determination; based on the matching between the widths of adjacent slabs; Width optimization is performed using a two-stage width optimization model. Update the merge results; after completing the width optimization process, the planned width of the slab has been redefined, and the Lots with the determined widths need to be merged.

7. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 6, characterized in that, First-stage width optimization model: The objective function is to maximize the number of slabs that can be continuously cast: The main constraint is to relax the feasibility constraints: Decision variables: 。 8. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 6, characterized in that, Second-stage width optimization model: The objective function, while ensuring continuity, maximizes the number of width differences and minimizes the difference from the target width. Phase 1 relaxation of feasible constraints: Determine whether to widen the width: First-stage optimal value constraint: Newly added decision variables: The optimized slab width is obtained by solving the problem using an optimization algorithm or solver.

9. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 1, characterized in that, In step S4, generating the casting schedule includes: Feasibility assessment for continuous casting: For the optimized Lot sequence, first determine whether it can be continuously cast in the same casting cycle; if there is a steel grade conflict between some Lots, it must be broken at that point and a new casting cycle must be started. Tundish life constraint: During continuous casting, the life of the tundish is usually limited by the casting time or the number of casting cycles. If the cumulative production time of the current casting exceeds the tundish life limit, the current casting must be forcibly terminated at that point and a new tundish must be replaced. Subsequent Lots will then proceed to the next casting cycle. Heat allocation; After determining the boundaries of each casting batch, the specific heats contained in the Lot need to be allocated to the corresponding casting batch one by one; The plan outputs the following: the start and end lot numbers for each casting cycle, the heat allocation, the corresponding steel grade information, the target width range, and whether the tundish needs to be replaced.

10. The multi-strand online width adjustment continuous casting machine multi-steel grade group casting method according to claim 2, characterized in that, Metallurgical process conditions include superheat control range, secondary cooling intensity, and allowable drawing speed window.

Images