Method and device for optimizing blanking scheduling of MOEA-D steel plate under double-process constraints

Abstract

The application provides a duplex process constrained MOEA-D steel plate blanking scheduling optimization method and device, and relates to the technical field of steel plate blanking production scheduling.In the application, material workpiece groups are divided according to material grades and plate thickness specifications, the same material workpiece group is continuously arranged in a process time sequence coding layer, the material workpiece group is used as a cross operation unit, and device distribution variation is performed in the group, so that the possibility of material aggregation violation solution caused by evolution operation splitting of the material workpiece group can be reduced; the processes are clustered according to the device distribution results, and the process time sequence in the same cutting machine is reconstructed according to the cutting completion time of the previous workpiece and the fixed delay of the transverse moving vehicle, so that the cross workpiece time sequence conflict can be reduced. Therefore, the method can reduce the process violation individuals generated in the initialization and MOEA-D iteration process, and improve the process executability of the output process sequence and process time sequence.

Description

Technical Field

[0001] This invention relates to the field of steel plate blanking production scheduling technology, specifically to a MOEA-D steel plate blanking scheduling optimization method and device with dual process constraints. Background Technology

[0002] Steel plate cutting is a preliminary processing step in the manufacturing of construction machinery, automotive sheet metal, and steel structure components. Under flexible production conditions, workshops typically handle multiple orders and steel plate workpieces of various materials and specifications simultaneously, with multiple cutting machines completing the processing in parallel. The processing sequence of the cutting machines, the connection of lateral sorting, and the switching of material specifications all affect the feasibility of the production scheduling plan.

[0003] For the same cutting machine, after the preceding workpiece is cut, a traverse carriage is needed to pick up, transfer, and sort the workpieces before the subsequent workpieces can be loaded. Different cutting machines form independent serial processing units. At the same time, workpieces with the same material grade and plate thickness should be processed continuously to reduce material changes, machine adjustments, and process parameter switching.

[0004] Traditional multi-objective evolutionary algorithms based on decomposition (MOEA-D) and other conventional shop floor scheduling algorithms typically employ general workpiece encoding and random position crossover and mutation operations. These methods struggle to simultaneously maintain both types of constraints in the initial population, and iterative operations can easily split material workpiece groups or disrupt the sequential relationships of workpieces within the same cutting machine. Relying on post-penalty functions or manual verification can generate numerous invalid individuals, requiring secondary adjustments to the output scheme. Therefore, a steel plate blanking scheduling optimization method that can maintain dual process constraints during encoding, initialization, evolution, and decoding is needed. Summary of the Invention

[0005] (a) Technical problems to be solved To address the shortcomings of existing technologies, this invention provides a MOEA-D steel plate blanking scheduling optimization method and apparatus with dual process constraints. It solves the problem of how to maintain the fixed delay timing relationship across workpieces of the same cutting machine and the continuous processing relationship of workpieces of the same material and specifications during the initialization and MOEA-D iteration of the scheduling individual in parallel steel plate blanking production of multiple cutting machines, and outputs the equipment processing sequence and process timing that conform to the aforementioned process relationship.

[0006] (II) Technical Solution To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for optimizing the blanking schedule of MOEA-D steel plates under dual process constraints, comprising: Obtain the material grade, plate thickness, processing time of three processes, and fixed delay of the optional cutting machine and transverse carriage for the steel plate workpiece to be processed. Establish scheduling constraints that allow each workpiece to complete loading, cutting and unloading in sequence, and bind the three processes of the same workpiece to the same cutting machine. A two-layer chromosome is constructed, comprising a machine coding layer and a process timing coding layer; wherein, the machine coding layer records the cutting machine allocation for each process, and the process timing coding layer records the workpiece processing sequence; Workpieces are divided into material groups according to material grade and plate thickness specifications, so that workpieces in the same group are arranged continuously in the process timing coding layer, and an initial population is generated based on the material workpiece groups. The equipment is clustered according to the machine coding layer. The sequence of workpieces is determined in each cutting machine according to the process timing coding layer. The loading start time of the subsequent workpiece is not earlier than the sum of the cutting completion time of the previous workpiece and the fixed delay of the transverse car, so as to reconstruct the timing of each process. Using material workpiece groups as cross-operation units, equipment allocation variation is performed within the same group, and the process timing is reconstructed on the mutated chromosome. The total completion time and workpiece sequence deviation are iteratively optimized using MOEA-D. After the termination condition is met, the target scheduling solution is determined from the Pareto optimal solution obtained by iteration based on the total completion time and the preset preference weight of the workpiece sequence deviation. The target scheduling solution is then decoded, and the workpiece processing sequence of each cutting machine and the start and completion times of each process are output.

[0007] Secondly, the present invention provides a MOEA-D steel plate blanking scheduling optimization device with dual process constraints, comprising: The data acquisition and constraint modeling module is used to acquire the material grade, plate thickness, processing time of three processes, and fixed delay of the optional cutting machine and transverse carriage of the steel plate workpiece to be processed, and to establish scheduling constraints that allow each workpiece to complete loading, cutting and unloading in sequence, and bind the three processes of the same workpiece to the same cutting machine. A dual-layer coding module is used to construct a dual-layer chromosome including a machine coding layer and a process timing coding layer; wherein, the machine coding layer records the cutting machine allocation for each process, and the process timing coding layer records the workpiece processing sequence; The population initialization module is used to divide the workpieces into material groups according to the material grade and plate thickness specifications, so that the workpieces in the same group are arranged continuously in the process timing coding layer, and to generate an initial population based on the material workpiece groups. The timing reconstruction module is used to cluster equipment according to the machine coding layer, determine the workpiece sequence relationship in each cutting machine according to the timing coding layer of the process, and ensure that the loading start time of the subsequent workpiece is not earlier than the sum of the cutting completion time of the previous workpiece and the fixed delay of the transverse car, so as to reconstruct the timing of each process. The iterative optimization module is used to perform equipment allocation variation within the same group, with the material workpiece group as the cross operation unit, and to reconstruct the process timing of the mutated chromosome. The total completion time and workpiece sequence deviation are iteratively optimized using MOEA-D. The decoding output module is used to determine the target scheduling solution from the Pareto optimal solution obtained through iteration based on the preset preference weight of the total completion time and the workpiece sequence deviation after the termination condition is met, and to decode the target scheduling solution to output the workpiece processing sequence of each cutting machine and the start and completion times of each process.

[0008] As a further improvement to the MOEA-D steel plate blanking scheduling optimization method or device with dual process constraints, the scheduling constraints include workpiece internal process timing constraints, equipment exclusive constraints, cross-workpiece timing coupling constraints, and material aggregation constraints. The internal process timing constraints of the workpiece require that the same workpiece complete the loading process, cutting process and unloading process in sequence. The device exclusivity constraint limits the same cutting machine to performing at most one process at any given time. The cross-workpiece temporal coupling constraint stipulates that the loading start time of the subsequent workpiece on the same cutting machine is not earlier than the sum of the cutting completion time of the preceding workpiece and the fixed delay of the transverse carriage; The material aggregation constraint limits the number of workpieces of the same material to occupy a continuous processing sequence interval.

[0009] As a further improvement to the MOEA-D steel plate blanking scheduling optimization method or device with dual process constraints, the lengths of the machine coding layer and the process timing coding layer are both the product of the number of workpieces to be processed and the number of processes for a single workpiece; the three processes of the same workpiece in the machine coding layer correspond to the same cutting machine number and appear three times consecutively in the process timing coding layer.

[0010] As a further improvement to the MOEA-D steel plate blanking scheduling optimization method or device under the aforementioned dual-process constraints, the step of dividing the workpieces into material groups according to material grade and plate thickness specifications, so that the workpieces in the same group are continuously arranged in the process timing coding layer, and generating an initial population based on the material workpiece groups, includes: Randomly arrange the workpiece groups of each material, and then randomly arrange the workpieces within each material group; according to the order of loading, cutting and unloading processes, continuously write the workpiece number of each workpiece into the process sequence coding layer. The corresponding machine code is determined from the optional cutting machines for each process to form the initial chromosome that complies with material aggregation constraints; The initial chromosome is generated repeatedly until the number of chromosomes reaches the preset population size, thus forming the initial population.

[0011] As a further improvement to the MOEA-D steel plate blanking scheduling optimization method or device under the aforementioned dual-process constraints, the reconstruction process of each process sequence includes: Based on the cutting machine allocation recorded in the machine coding layer, the workpieces and their processes allocated to the same cutting machine are clustered to form a serial processing sequence corresponding to each cutting machine; The sequential relationship of workpieces in each serial processing sequence is determined based on the process timing coding layer. Starting from the first workpiece in each serial processing sequence, the start time and completion time of each process are calculated recursively according to the order of loading, cutting and unloading processes; Specifically, for adjacent preceding and subsequent workpieces on the same cutting machine, the fixed delay starts when the cutting process of the preceding workpiece is completed and covers the process of receiving, transferring, and sorting materials by the traverse car.

[0012] As a further improvement to the MOEA-D steel plate blanking scheduling optimization method or device with the aforementioned dual-process constraints, the iterative optimization process of MOEA-D includes: Using a complete material workpiece group as a cross-operation unit, the process timing code segment and machine code segment corresponding to the material workpiece group are synchronously exchanged between parent chromosomes. Select at least one workpiece from the selected material workpiece group, change the cutting machine assignment of the workpiece, and simultaneously update the machine code corresponding to the three processes of the workpiece. The process timing is reconstructed for the offspring chromosomes generated after crossover and mutation. The total completion time and workpiece order deviation of the offspring chromosomes are calculated, and the population is updated based on the weight vector, neighborhood, and Chebyshev aggregation value.

[0013] As a further improvement to the MOEA-D steel plate blanking scheduling optimization method or device with the aforementioned dual-process constraints, the process of determining the target scheduling solution includes: The iteration terminates when any of the following conditions are met: (i) the number of iterations reaches the preset maximum number of iterations; (ii) the improvement of the Pareto front within a preset number of consecutive algebras does not reach the preset improvement threshold. After the iteration is terminated, the total completion time and work order deviation of each Pareto optimal solution obtained by the iteration are normalized. The normalized total completion time and work order deviation are weighted according to the preset preference weight, and the Pareto optimal solution with the smallest weighted evaluation value is determined as the target scheduling solution.

[0014] Thirdly, the present invention provides a storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the MOEA-D steel plate blanking scheduling optimization method as described in the first aspect.

[0015] Fourthly, the present invention provides an electronic device, the electronic device comprising: The processor, the memory, and the program instructions stored in the memory and executable by the processor, wherein when the processor executes the program instructions, it implements the MOEA-D steel plate blanking scheduling optimization method as described in the first aspect.

[0016] (III) Beneficial Effects This invention provides a method and apparatus for optimizing the blanking schedule of MOEA-D steel plates under dual process constraints. Compared with the prior art, it has the following advantages: In this invention, by dividing workpieces into material groups according to material grade and plate thickness specifications, workpiece groups of the same material are arranged continuously in the process timing coding layer. Using workpiece groups as cross-operation units and performing equipment allocation mutations within the group, the possibility of material aggregation violations caused by splitting workpiece groups during evolutionary operations can be reduced. By clustering processes according to equipment allocation results and reconstructing the process timing within the same cutting machine based on the previous workpiece cutting completion time and the fixed delay of the traverse carriage, cross-workpiece timing conflicts can be reduced. Therefore, this method can reduce individual process violations generated during initialization and MOEA-D iteration, improving the process executability of the output processing sequence and process timing. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A detailed flowchart of the output of a MOEA-D iterative optimization and scheduling scheme provided in an embodiment of the present invention; Figure 2 A flowchart of a MOEA-D steel plate blanking scheduling optimization method with dual process constraints provided in an embodiment of the present invention; Figure 3 A conceptual diagram of cross-workpiece temporal coupling constraints provided in an embodiment of the present invention; Figure 4 This is a conceptual diagram illustrating material aggregation constraints during the steel plate cutting process, provided as an embodiment of the present invention. Figure 5This is a comparison chart of feasible solution rates for different workpiece sizes provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0019] Component labeling explanation: 100 - electronic device, 101 - memory, 102 - processor, 103 - display. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] To solve the above-mentioned technical problems, the overall technical concept of the embodiments of the present invention is as follows: Figure 1 As shown: First, the steel plate workpiece, equipment, and process parameters are obtained, basic scheduling constraints and dual process constraints are established, and a two-layer chromosome is constructed. Second, an initial solution satisfying the material aggregation constraint is generated through continuous encoding of the material workpiece group. By recursively calculating the process time according to the fixed delay of the traversing car, the initial solution satisfies the cross-workpiece temporal coupling constraint. Subsequently, the MOEA-D iterative optimization is performed using a dual-constraint adaptive evolutionary operator, maintaining the integrity of the material workpiece group during the iteration process and reconstructing the process timing after equipment allocation changes. After the termination condition is met, the target scheduling solution is finally determined from the Pareto optimal solution and decoded for output. Understandably, Figure 1 The optimal scheduling scheme is the target scheduling solution determined based on the preset preference weights.

[0022] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0023] like Figure 2 As shown, this embodiment of the invention provides a method for optimizing the blanking schedule of MOEA-D steel plates under dual process constraints, including: S1. Obtain the material grade, plate thickness, processing time of three processes, and fixed delay of the optional cutting machine and transverse carriage for the steel plate workpiece to be processed. Establish scheduling constraints that allow each workpiece to complete loading, cutting and unloading in sequence, and bind the three processes of the same workpiece to the same cutting machine. S2. Construct a two-layer chromosome including a machine coding layer and a process timing coding layer; wherein, the machine coding layer records the cutting machine allocation for each process, and the process timing coding layer records the workpiece processing sequence; S3. Divide the workpieces into material groups according to material grade and plate thickness specifications, so that the workpieces in the same group are arranged continuously in the process sequence coding layer, and generate an initial population based on the material workpiece groups. S4. Cluster the processes according to the machine coding layer, determine the workpiece sequence relationship in each cutting machine according to the process timing coding layer, and ensure that the loading start time of the subsequent workpiece is not earlier than the sum of the cutting completion time of the previous workpiece and the fixed delay of the transverse car, so as to reconstruct the timing of each process. S5. Using the material workpiece group as the cross operation unit, perform equipment allocation variation within the same group, and reconstruct the process sequence of the mutated chromosome. Use MOEA-D to iteratively optimize the total completion time and workpiece sequence deviation. S6. After the termination condition is met, the target scheduling solution is determined from the Pareto optimal solution obtained by iteration based on the total completion time and the preset preference weight of the workpiece sequence deviation. The target scheduling solution is then decoded, and the workpiece processing sequence of each cutting machine and the start and completion times of each process are output.

[0024] The dual process constraints referred to in this embodiment refer to the cross-workpiece temporal coupling constraints and material aggregation constraints that exist simultaneously during the steel plate blanking scheduling process. Specifically, the cross-workpiece temporal coupling constraint means that for adjacent preceding and subsequent workpieces on the same cutting machine, the loading start time of the subsequent workpiece must not be earlier than the sum of the cutting completion time of the preceding workpiece and the fixed delay of the traverse carriage. The material aggregation constraint means that workpieces with the same material grade and plate thickness are grouped into the same material workpiece group, and the workpieces in the same material workpiece group are arranged continuously in the processing sequence. The workpiece internal process temporal constraints and equipment exclusivity constraints are basic scheduling constraints and do not belong to the above two process constraints.

[0025] The steel plate blanking workshop includes multiple cutting machine processing units. Each cutting machine processing unit includes supporting processes for loading, cutting, and transverse sorting of steel plates. Once a workpiece is assigned to a cutting machine, its loading, cutting, and unloading processes are all completed sequentially within that cutting machine processing unit.

[0026] The following establishes a unified convention for the parameter symbols used in the formulas of this embodiment: This represents the collection of steel plate workpieces to be processed.

[0027] This indicates the number of steel plate workpieces to be processed.

[0028] This represents a collection of cutting machines.

[0029] This indicates the number of cutting machines.

[0030] Indicates the workpiece index; Indicates the current or previous workpiece. This refers to subsequent workpieces on the same cutting machine.

[0031] Indicates the process index; These represent loading, cutting, and unloading, respectively.

[0032] This indicates the cutting machine index.

[0033] This indicates the index of the material workpiece group.

[0034] This represents the MOEA-D subproblem index.

[0035] Represents a time variable.

[0036] This represents the q-th process of workpiece j.

[0037] This represents the set of material types determined by both the material grade and the plate thickness specification.

[0038] Indicates the first Group of workpieces of similar material.

[0039] Indicates workpiece The The start time of each process.

[0040] Indicates workpiece The The completion time of each process.

[0041] Indicates workpiece The The cutting machine number corresponding to each process.

[0042] Represents a 0-1 state variable, indicating time. workpiece The Is the next process performed on the cutting machine? Execute above.

[0043] This indicates the fixed delay for the traverse vehicle to pick up, transfer, and sort materials.

[0044] They represent the first The minimum and maximum positions of a group of workpieces of similar material in the actual processing sequence.

[0045] Indicates workpiece Position in the actual processing sequence.

[0046] Indicates workpiece Its position in the planned processing sequence.

[0047] This represents a two-layer chromosome consisting of a machine coding layer and a process timing coding layer.

[0048] These represent the machine coding layer and the process timing coding layer, respectively.

[0049] Indicates the length of a single-layer code; when each workpiece contains three processes. .

[0050] This represents the scheduling individual that has completed the time-series reconfiguration.

[0051] This represents the target vector consisting of total completion time and workpiece sequence deviation.

[0052] These represent the total completion time target and the workpiece sequence deviation target, respectively.

[0053] They represent the first The weight vector of each sub-problem and its first Each target weight.

[0054] Represents the scheduling individual The A normalized target value.

[0055] They represent the first and second digits in the current population, respectively. The minimum and maximum values ​​of each target.

[0056] Let them represent the ideal point and its i in the normalized objective space, respectively. Each target component.

[0057] This represents a positive number whose denominator is zero and which should be avoided during normalization.

[0058] Represents the scheduling individual In the Chebyshev aggregation values ​​on individual problems.

[0059] These represent the population size and neighborhood size of MOEA-D, respectively.

[0060] This represents the parameters for dividing the weight vector.

[0061] This indicates the maximum number of iterations.

[0062] Specifically, in step S1, the material grade, plate thickness, processing time of the three processes, and fixed delay of the optional cutting machine and transverse carriage of the steel plate workpiece to be processed are obtained, and a scheduling constraint is established that each workpiece completes loading, cutting and unloading in sequence, and the three processes of the same workpiece are bound to the same cutting machine.

[0063] This step obtains the workpiece to be processed, material grade, plate thickness, processing time of each process, and selectable cutting machines recorded by the workshop production management system, and also obtains the traverse carriage's fixed delay. The workpiece set is denoted as... The set of cutting machines is denoted as Each workpiece include , and The three processes correspond to feeding, cutting, and unloading, respectively.

[0064] The scheduling constraints include basic scheduling constraints and dual-process constraints. Basic scheduling constraints include workpiece internal process timing constraints and equipment exclusivity constraints. The workpiece internal process timing constraints limit the sequential completion of the loading, cutting, and unloading processes for the same workpiece. The equipment exclusivity constraints limit the execution of at most one process by the same cutting machine at any given time. Dual-process constraints include cross-workpiece timing coupling constraints and material aggregation constraints, the relevant definitions of which are described above.

[0065] Specifically, the internal process sequence of the workpiece, that is, the three processes of the same workpiece are executed sequentially, and satisfy the following: in, Indicates workpiece The The start time of each process, Indicates workpiece The The completion time of each process.

[0066] To ensure that the three processes for the same workpiece are processed within the same cutting machine unit, the cutting machine numbers corresponding to the three processes must satisfy the following: The equipment exclusivity constraint means that for any given cutting machine and at any given time, the number of processes that the cutting machine can simultaneously perform does not exceed one. Indicates time workpiece The Is the next process performed on the cutting machine? If executed above, the device exclusivity relationship is represented as follows: like Figure 3 As shown, Figure 3 The sequential coupling process across workpieces is illustrated according to the order of loading, cutting, transverse transfer, and loading of the second steel plate onto the same cutting machine. and subsequent workpieces The fixed delay of the traverse car starts when the cutting process of the preceding workpiece is completed, covering the traverse car's receiving, transfer, and sorting / unloading processes. The loading start time of the subsequent workpiece must meet the following requirements: in, Indicates workpiece The start time of the first process, Indicates workpiece The completion time of the second process.

[0067] When no other waiting conditions exist, the earliest feasible timing is used for decoding, and the start time of loading subsequent workpieces is the sum of the cutting completion time of the preceding workpiece and the fixed delay. When there are equipment monopolies or other waiting conditions, the latest time allowed by each constraint is taken as the earliest feasible loading time of subsequent workpieces. The above fixed delay relationship is not set between workpieces from different cutting machines.

[0068] Workpieces with the same material grade and plate thickness are grouped into workpieces of the same material. .make and Let be the minimum and maximum positions of the workpieces in the actual processing sequence, respectively. Then, the material aggregation relationship can be expressed as: like Figure 4 As shown, steel plate 1 and steel plate 2 are both marked as material A, and their loading and subsequent processing are arranged adjacently in the sequence; steel plate 3 is marked as material B, and enters the processing sequence after the workpiece group of material A is continuously arranged. Figure 4 Materials A and B are used to illustrate different material types determined by different material grades and / or plate thickness specifications; in this embodiment, only workpieces with the same material grade and plate thickness specification are grouped into the same material workpiece group. Therefore, Figure 4 The adjacent arrangement relationship shown corresponds to the material aggregation constraint where a group of workpieces of the same material occupies a continuous processing sequence interval.

[0069] in, To find the minimum value operator, This represents the maximum value operator; The cardinality of a set is the number of elements contained in the set.

[0070] Since each workpiece has a unique position in the processing sequence, the above relationship ensures that the length of the sequence interval occupied by the workpiece group of the material is equal to the number of workpieces in the group, thereby preventing workpieces of different materials and specifications from being inserted into the interval.

[0071] Specifically, in step S2, a two-layer chromosome is constructed, comprising a machine coding layer and a process timing coding layer; wherein the machine coding layer records the cutting machine allocation for each process, and the process timing coding layer records the workpiece processing sequence.

[0072] Build a machine coding layer and process timing coding layer Composed of double-layered chromosomes .for In the case of one workpiece and each workpiece includes three processes, the length of both layers of coding is [missing information]. A double-layered chromosome is represented as: Each coding position in the machine coding layer records the cutting machine number of the corresponding process. The three machine codes corresponding to the three processes of the same workpiece are consistent, binding the workpiece to the same cutting machine processing unit. Each coding position in the process sequence coding layer records the workpiece number to which the corresponding process belongs. The same workpiece number appears three times consecutively, corresponding to loading, cutting, and unloading, respectively.

[0073] In step S3, the workpieces are divided into material groups according to material grade and plate thickness specifications, so that the workpieces in the same group are arranged continuously in the process timing coding layer, and an initial population is generated based on the material workpiece groups.

[0074] This step divides all workpieces into multiple material workpiece groups based on material grade and plate thickness. Each group covers all workpieces to be processed and does not overlap with others. Workpiece groups of the same material occupy a continuous coding block in the process sequence coding layer, and the overall arrangement order between workpiece groups of different materials can be adjusted.

[0075] For each initial individual, the workpiece groups of each material are first randomly arranged, and then the workpieces within each group are randomly arranged separately. For each sorted workpiece, the workpiece number is written into the process sequence coding layer three times consecutively in the order of loading, cutting, and unloading, so that the coding positions of all workpieces in the same material workpiece group remain continuous.

[0076] For each workpiece, a cutting machine is selected from the available cutting machines that can perform the three processes of loading, cutting and unloading the workpiece, and the machine code of the cutting machine is synchronously written into the machine code layer at the position corresponding to the three processes of the workpiece, thereby generating an initial chromosome that satisfies the material aggregation constraint and the equipment binding relationship.

[0077] Repeat the above process of between-group sorting, within-group sorting, and machine encoding until the initial number of chromosomes reaches the preset population size. This forms the initial population. Before calculating the target value, each initial chromosome undergoes sequential reconstruction according to the subsequent step S4 to obtain the corresponding feasible schedulable individuals.

[0078] Specifically, in step S4, the process is clustered according to the machine coding layer. The sequence of workpieces is determined in each cutting machine according to the process timing coding layer. The loading start time of the subsequent workpiece is not earlier than the sum of the cutting completion time of the previous workpiece and the fixed delay of the transverse car, so as to reconstruct the timing of each process.

[0079] Using the obtained double-layer chromosome as input, the cutting machine allocation information in the machine coding layer is read, and the workpieces and their three processes assigned to the same cutting machine are clustered to form a serial processing sequence corresponding to each cutting machine. In each serial processing sequence, the sequence relationship of the workpieces on that cutting machine is determined according to the order in which the workpieces first appear in the process timing coding layer.

[0080] Starting with the first workpiece in each sequential processing sequence, the start and finish times of each process are calculated recursively according to the order of loading, cutting, and unloading. Within the same workpiece, the start time of the subsequent process cannot be earlier than the finish time of the preceding process. For adjacent workpieces on the same cutting machine, the loading start time of the subsequent workpiece cannot be earlier than the sum of the cutting completion time and fixed delay of the preceding workpiece, while simultaneously satisfying the exclusive constraint of the cutting machine. Therefore, the earliest feasible loading time for the subsequent workpiece is the latest time allowed by the above constraints.

[0081] The completion time of each process is determined by its start time and corresponding processing time. When equipment allocation or workpiece sequence changes, the subsequent processes in the affected cutting machine sequence starting from the changed position are recalculated, updating the start and completion times of each process. The reconstructed results are written to the scheduling individual for target value calculation and final decoding.

[0082] Specifically, in step S5, the material workpiece group is used as the cross operation unit, the equipment allocation variation is performed within the same group, and the process timing is reconstructed on the mutated chromosome. The total completion time and workpiece sequence deviation are iteratively optimized using MOEA-D.

[0083] This step corresponds to Figure 1 The iterative optimization stage of the dual-constraint adaptive evolution operator shown is implemented in some embodiments as follows: The multi-objective optimization problem of steel plate blanking is decomposed into multiple sub-problems with different weight vectors by MOEA-D, and neighborhoods are established according to the distance between the weight vectors.

[0084] Specifically, the population size is denoted as The neighborhood size is denoted as Each weight vector contains a total completion time weight and a workpiece sequence deviation weight, and the sum of the two weights is 1.

[0085] For the crossover operation, each complete material workpiece group is first treated as a unit, and a parent group sequence is formed based on the order of the material workpiece groups in the parent chromosome. At least one consecutive material workpiece group is randomly selected from the first parent group sequence, and the selected material workpiece group, along with the workpiece arrangement within it, is copied to the corresponding position in the first child group sequence. Then, the second parent group sequence is traversed according to the order in which each material workpiece group appears in the second parent group sequence, skipping material workpiece groups already included in the first child group sequence, and filling the remaining material workpiece groups into the empty positions in the first child group sequence. Using the second parent as the source of the retained segments and the first parent as the source of the remaining material workpiece groups, the second child group sequence is generated in the same manner.

[0086] After constructing the first and second generation sequences, the coding segments corresponding to each material workpiece group are spliced ​​according to their arrangement order in the corresponding generation sequence. For the material workpiece group retained by the first parent in the first generation sequence, the process timing coding segment and the corresponding machine coding segment of the material workpiece group in the first parent are copied. For the material workpiece group filled in by the second parent, the process timing coding segment and the corresponding machine coding segment of the material workpiece group in the second parent are copied. For the material workpiece group retained by the second parent and the material workpiece group filled in by the first parent in the second generation sequence, the process timing coding segment and the corresponding machine coding segment are copied from the corresponding parent in the same way. During the copying process, the corresponding positional relationship between the process timing coding layer and the machine coding layer is maintained to form the first and second generation chromosomes.

[0087] This ensures that each material workpiece group appears only once in the offspring chromosome, each workpiece corresponds to only one complete three-process coding unit, and the workpiece number of the workpiece appears once each in the loading process, cutting process, and unloading process, thereby maintaining the continuity of the workpiece coding in the same material workpiece group.

[0088] For example, the material workpiece group order of the first parent generation is G1, G2, G3, G4, and the material workpiece group order of the second parent generation is G3, G1, G4, G2. When G2 and G3 in the first parent generation are selected as the retained groups, the first child generation first retains G2 and G3 in the corresponding group positions, and then skips the existing G3 and G2 according to the group order of the second parent generation, and fills G1 and G4 into the empty positions in turn. The workpiece number and its machine code in each material workpiece group are copied synchronously with the corresponding group segment. According to the aforementioned filling method, the material workpiece group order of the first child generation is G1, G2, G3, G4.

[0089] For mutation operations, a workpiece group of a random material is selected, and at least one workpiece within that group is selected for equipment allocation mutation. A target cutting machine different from the current cutting machine is determined from the available cutting machines capable of performing the three processes for that workpiece, and the machine code corresponding to the three processes for that workpiece is synchronously updated to the target cutting machine number to maintain the equipment binding relationship for the same workpiece. If no other available cutting machine different from the current cutting machine exists for the selected workpiece, the machine code for that workpiece remains unchanged, and a workpiece with other available cutting machines within the same material workpiece group is reselected; if no other available cutting machines exist within the same material workpiece group, the current equipment allocation mutation for that material workpiece group is cancelled.

[0090] After the crossover and mutation operations are completed, the start and finish times of each process corresponding to the offspring chromosome are reconstructed according to step S4, and the constraint compliance of the offspring chromosome is verified. When the equipment allocation changes, at least the affected parts of the original cutting machine sequence and the target cutting machine sequence are recalculated; when the workpiece sequence relationship changes, the affected sequence is recalculated from the changed position. After the time sequence reconstruction is completed, the two target values ​​of the offspring are calculated.

[0091] Based on total completion time and workpiece sequence deviation The two objectives are to be minimized. The total completion time is the maximum completion time of all workpiece cutting processes, and the workpiece sequence deviation is the sum of the absolute values ​​of the differences between the planned position and the actual position of each workpiece.

[0092] Among them, superscript Indicates transpose; This represents a scheduling entity that has completed time-series reconfiguration; This indicates the position of workpiece j in the planned processing sequence. This indicates its position in the actual processed sequence after decoding.

[0093] In each iteration, offspring are generated in batches for all subproblems, and after temporal reconstruction, the two target values ​​are calculated in batches. To reduce the impact of the difference in the two target dimensions, normalization is performed based on the minimum and maximum values ​​of the corresponding targets in the current population. Then, Chebyshev aggregation values ​​are calculated using the weight vector, the normalized target values, and the ideal point, and individuals in the corresponding neighborhood are updated uniformly.

[0094] in, and They represent the first and second digits in the current population, respectively. The minimum and maximum values ​​of each objective; This indicates a positive number used to avoid a denominator of zero; Represents an ideal point in the normalized target space. express The Each component.

[0095] Specifically, in step S6, after the termination condition is met, the target scheduling solution is determined from the Pareto optimal solution obtained by iteration based on the total completion time and the preset preference weight of the workpiece sequence deviation, and the target scheduling solution is decoded to output the workpiece processing sequence of each cutting machine and the start and completion times of each process.

[0096] The iteration terminates when any of the following conditions are met: First, the number of iterations reaches the preset maximum number of iterations. Second, the improvement of the Pareto front relative to the previous generation is less than the preset improvement threshold in consecutive preset generations.

[0097] Pareto optimal solution refers to a multi-objective optimization problem in which there is no other feasible solution that makes all objective functions non-deteriorate and at least one objective function strictly better. The set of all Pareto optimal solutions is called the Pareto frontier. The above content is all existing technology and will not be repeated here.

[0098] For example, for the improvement amount, a fixed target value can be used to normalize the upper and lower bounds of the total completion time and workpiece sequence deviation of the Pareto optimal solution for each generation. A fixed reference point is set in the normalized target space, and the hypervolume corresponding to the Pareto front for each generation is calculated. The relative increase of the hypervolume of the current generation relative to the hypervolume of the previous generation is taken as the Pareto front improvement amount. When the improvement amount is less than a preset improvement threshold for a consecutive preset number of generations, it is determined that the Pareto front tends to stabilize and the iteration is terminated. For example, if the fixed reference point is (1.1, 1.1), the hypervolumes of the Pareto front for the 99th and 100th generations are 0.7420 and 0.7424, respectively, then the improvement amount of the 100th generation relative to the 99th generation is approximately 0.00054. If the preset improvement threshold is 0.001, and the improvement amount is less than 0.001 for 20 consecutive generations, then the iteration is terminated in the 100th generation.

[0099] After the iteration terminates, the total completion time and job order deviation corresponding to each Pareto optimal solution are normalized using the normalized range of the target value updated at the end of the iteration. The total completion time corresponds to the first preset preference weight, and the job order deviation corresponds to the second preset preference weight. Both preset preference weights are non-negative and their sum is 1. The first and second preset preference weights are used for the selection of the final target scheduling solution, which is different from the weight vector used for the decomposition of the MOEA-D subproblem in the previous step S5. Furthermore, based on two preset preference weights, the normalized total completion time and workpiece sequence deviation are weighted and evaluated, and the Pareto optimal solution with the smallest weighted evaluation value is determined as the target scheduling solution. Specifically, when two or more Pareto optimal solutions have the same minimum weighted evaluation value, the Pareto optimal solution with the smaller total completion time is preferentially determined as the target scheduling solution; if the total completion time is still the same, the Pareto optimal solution with the smaller workpiece sequence deviation is preferentially selected as the target scheduling solution. The machine encoding layer and process timing encoding layer of the target scheduling solution are read, and the time of each process is recalculated according to the previous step S4, and the internal process timing constraints, equipment exclusive constraints, cross-workpiece timing coupling constraints, and material aggregation constraints are verified.

[0100] After successful verification, the system outputs the workpiece processing sequence for each cutting machine, as well as the start and finish times for each process. Based on actual scheduling and management needs, it can also output the cutting machine allocation results for each workpiece, the finish time of each cutting machine, and the workpiece sequence deviation corresponding to the target scheduling solution.

[0101] To further understand the MOEA-D steel plate blanking scheduling optimization method with dual process constraints (hereinafter referred to as "this method") provided in this invention, the following specific examples are provided: Production data from a high-strength steel workshop at a materials center was used for verification. The workshop includes three industrial cutting machines. , and Three workpiece sizes are set: 9, 12, and 20 pieces. Material type is determined by a combination of material grade and plate thickness, including Q590D-10, Q355B-10, Q355B-20, Q690D-10, and Q690D-25; the traverse car has a fixed delay time. for The processing time, optional equipment, and material parameters for each workpiece's three processes are collected by the workshop production management system.

[0102] Under 20 working conditions, the length of both the machine coding layer and the process timing coding layer is 60. Population size. The value is set to 100, using 100 two-dimensional equally spaced weight vectors. The weight partitioning parameters are... Take 99, neighborhood size Set the number of iterations to 15. Take 200. Crossover is performed on a per-material-workpiece-group basis. Mutation adjusts the equipment allocation within the material-workpiece-group and performs local timing reconstruction after mutation.

[0103] The control group used random coding, standard MOEA-D, and a post-penalty function. The method in this application employs double-layer coding, material-workpiece group order-preserving operations, and temporal reconstruction. Each workpiece size is independently run 10 times, and the average value is taken. In this example, a feasible individual is a scheduled individual that simultaneously satisfies the workpiece's internal process timing constraints, equipment exclusivity constraints, cross-workpiece temporal coupling constraints, and material aggregation constraints. For each independent run, the feasible solution rate is the ratio of the number of feasible individuals in the final population to the population size at the end of the iteration. Each workpiece size is run independently 10 times, and the feasible solution rates in Table 1 are the arithmetic mean of the feasible solution rates obtained from the 10 independent runs.

[0104] The first feasible generation refers to the number of iterations in a single independent run when at least one feasible individual is found for the first time. The initial population is designated as generation 0. When all 10 independent runs yield a feasible solution, the first feasible generation in Table 1 is the arithmetic mean of the first feasible generation numbers obtained from the 10 independent runs. When only some independent runs yield a feasible solution, the first feasible generation is the arithmetic mean of the first feasible generation numbers corresponding to the independent runs that yielded a feasible solution, with the number of runs that did not yield a feasible solution recorded separately, and the results of runs that did not yield a feasible solution not included in the arithmetic mean. When none of the 10 independent runs yield a feasible solution, the first feasible generation in Table 1 is represented by ∞.

[0105] The feasible solution rate and the number of iterations required to generate a feasible solution for the first time are shown in Table 1.

[0106] Table 1 The above feasible solution rates are as follows: Figure 5 As shown. Figure 5 In the trials with 9, 12, and 20 workpieces, the feasible solution rate of the proposed method was 1.000%, while the control group's rates were 99.00%, 51.50%, and 0.00%, respectively. As the workpiece size increased, the feasible solution rate of the control group decreased, while the proposed method maintained a feasible solution rate for all trials. This indicates that encoding based on material workpiece groups and combining evolutionary operations with temporal reconstruction can reduce the proportion of individuals violating dual-process constraints.

[0107] Meanwhile, this method obtains feasible individuals in the initial population stage; this result further illustrates the role of material workpiece group coding and temporal reconstruction in constraint compliance.

[0108] Based on the same inventive concept, this invention also provides a dual-process-constrained MOEA-D steel plate blanking scheduling optimization device, comprising: The data acquisition and constraint modeling module is used to acquire the material grade, plate thickness, processing time of three processes, and fixed delay of the optional cutting machine and transverse carriage of the steel plate workpiece to be processed, and to establish scheduling constraints that allow each workpiece to complete loading, cutting and unloading in sequence, and bind the three processes of the same workpiece to the same cutting machine. A dual-layer coding module is used to construct a dual-layer chromosome including a machine coding layer and a process timing coding layer; wherein, the machine coding layer records the cutting machine allocation for each process, and the process timing coding layer records the workpiece processing sequence; The population initialization module is used to divide the workpieces into material groups according to the material grade and plate thickness specifications, so that the workpieces in the same group are arranged continuously in the process timing coding layer, and to generate an initial population based on the material workpiece groups. The timing reconstruction module is used to cluster equipment according to the machine coding layer, determine the workpiece sequence relationship in each cutting machine according to the timing coding layer of the process, and ensure that the loading start time of the subsequent workpiece is not earlier than the sum of the cutting completion time of the previous workpiece and the fixed delay of the transverse car, so as to reconstruct the timing of each process. The iterative optimization module is used to perform equipment allocation variation within the same group, with the material workpiece group as the cross operation unit, and to reconstruct the process timing of the mutated chromosome. The total completion time and workpiece sequence deviation are iteratively optimized using MOEA-D. The decoding output module is used to determine the target scheduling solution from the Pareto optimal solution obtained through iteration based on the preset preference weight of the total completion time and the workpiece sequence deviation after the termination condition is met, and to decode the target scheduling solution to output the workpiece processing sequence of each cutting machine and the start and completion times of each process.

[0109] Each module in the above-mentioned device can be implemented by a processor executing program instructions stored in memory, or by software components, hardware circuits, or a combination of both. The module name is only used to distinguish its main function and does not require that each module correspond to a separate physical device; a processing unit can implement the functions of multiple modules, and the same module can also be distributed on multiple controlled computing nodes.

[0110] Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the MOEA-D steel plate blanking scheduling optimization method with dual process constraints as provided in any embodiment of the present invention.

[0111] In embodiments of the present invention, any combination of one or more storage media may be used. The storage medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, RAM, ROM, erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in connection with an instruction execution system, apparatus, or device.

[0112] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 6 The diagram shown is a structural schematic of the electronic device 100 provided in an embodiment of the present invention. In some embodiments, the electronic device may be a mobile phone, tablet computer, wearable device, in-vehicle device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), or other terminal device. Furthermore, the dual-process-constrained MOEA-D steel plate blanking scheduling optimization method provided in this embodiment of the present invention can also be applied to databases, servers, and service response systems based on terminal artificial intelligence. This embodiment of the present invention does not impose any limitations on the specific application scenarios of the dual-process-constrained MOEA-D steel plate blanking scheduling optimization method.

[0113] like Figure 6 As shown, the electronic device 100 provided in this embodiment of the invention includes a memory 101 and a processor 102.

[0114] The memory 101 is used to store computer programs; preferably, the memory 101 includes various media that can store program code, such as ROM, RAM, magnetic disk, USB flash drive, memory card or optical disk.

[0115] Specifically, memory 101 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. Electronic device 100 may further include other removable / non-removable, volatile / non-volatile computer system storage media. Memory 101 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0116] The processor 102 is connected to the memory 101 and is used to execute the computer program stored in the memory 101 to implement the MOEA-D steel plate blanking scheduling optimization method with dual process constraints provided in any embodiment of the present invention.

[0117] In an optional implementation, the processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates, transistor logic devices, and discrete hardware components.

[0118] In an optional embodiment, the electronic device 100 of this invention may further include a display 103. The display 103 is communicatively connected to the memory 101 and the processor 102, and is used to display the relevant GUI interactive interface of the MOEA-D steel plate blanking scheduling optimization method with dual process constraints.

[0119] It is understood that the MOEA-D steel plate blanking scheduling optimization device, storage medium and electronic device with dual process constraints provided in the embodiments of the present invention correspond to the MOEA-D steel plate blanking scheduling optimization method with dual process constraints provided in the embodiments of the present invention. The explanation, examples and beneficial effects of the relevant contents can be referred to the corresponding parts of the method, and will not be repeated here.

[0120] In summary, compared with existing technologies, it has the following beneficial effects: 1. This invention reduces process violation individuals generated during iteration, reduces invalid target value calculation, and improves the stability and optimization efficiency of obtaining executable scheduling solutions by directly embedding material aggregation constraints and cross-workpiece temporal coupling constraints during initial population generation, evolutionary operation, and temporal reconstruction.

[0121] 2. This invention incorporates the fixed delay of the transverse car from the moment the previous workpiece is cut to the moment the earliest feasible loading time of the subsequent workpiece is calculated. This allows the generated process time to reflect the transverse car's receiving, transfer, and sorting processes, reducing equipment occupancy conflicts and loading timing conflicts, and improving the matching degree between the scheduling scheme and the actual steel plate loading cycle.

[0122] 3. This invention divides workpieces into groups based on material grade and plate thickness, allowing workpiece groups of the same material to be arranged continuously and to perform evolutionary operations with complete workpiece groups of the same material. This reduces the possibility of workpieces in the same group being split up and workpieces of different materials being inserted, which helps to reduce material switching and the preparation time and energy consumption it causes, and improves the continuous processing capability of similar steel plates.

[0123] 4. This invention expresses the cutting machine allocation and workpiece processing sequence separately through a double-layer chromosome, and reconstructs the process sequence after the equipment allocation changes. It can coordinate the relationship between equipment allocation, processing sequence and process time, avoid destroying the equipment binding relationship of the three processes of the same workpiece by changing the machine code alone, and improve the accuracy of chromosome decoding results and process executability.

[0124] 5. By simultaneously optimizing total completion time and workpiece sequence deviation through MOEA-D, and determining the target scheduling solution from the Pareto optimal solution according to production preferences, it can balance shortening the overall processing cycle and maintaining the planned processing sequence, providing an adjustable scheduling decision basis for different production tasks and improving the adaptability of production planning.

[0125] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0126] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for optimizing MOEA-D steel plate blanking scheduling under dual process constraints, characterized in that, include: Obtain the material grade, plate thickness, processing time of three processes, and fixed delay of the optional cutting machine and transverse carriage for the steel plate workpiece to be processed. Establish scheduling constraints that allow each workpiece to complete loading, cutting and unloading in sequence, and bind the three processes of the same workpiece to the same cutting machine. A two-layer chromosome is constructed, comprising a machine coding layer and a process timing coding layer; wherein, the machine coding layer records the cutting machine allocation for each process, and the process timing coding layer records the workpiece processing sequence; Workpieces are divided into material groups according to material grade and plate thickness specifications, so that workpieces in the same group are arranged continuously in the process timing coding layer, and an initial population is generated based on the material workpiece groups. The equipment is clustered according to the machine coding layer. The sequence of workpieces is determined in each cutting machine according to the process timing coding layer. The loading start time of the subsequent workpiece is not earlier than the sum of the cutting completion time of the previous workpiece and the fixed delay of the transverse car, so as to reconstruct the timing of each process. Using material workpiece groups as cross-operation units, equipment allocation variation is performed within the same group, and the process timing is reconstructed on the mutated chromosome. The total completion time and workpiece sequence deviation are iteratively optimized using MOEA-D. After the termination condition is met, the target scheduling solution is determined from the Pareto optimal solution obtained by iteration based on the total completion time and the preset preference weight of the workpiece sequence deviation. The target scheduling solution is then decoded, and the workpiece processing sequence of each cutting machine and the start and completion times of each process are output.

2. The MOEA-D steel plate blanking scheduling optimization method as described in claim 1, characterized in that, The scheduling constraints include workpiece internal process timing constraints, equipment exclusive constraints, cross-workpiece timing coupling constraints, and material aggregation constraints. The internal process timing constraints of the workpiece require that the same workpiece complete the loading process, cutting process and unloading process in sequence. The device exclusivity constraint limits the same cutting machine to performing at most one process at any given time. The cross-workpiece temporal coupling constraint stipulates that the loading start time of the subsequent workpiece on the same cutting machine is not earlier than the sum of the cutting completion time of the preceding workpiece and the fixed delay of the transverse carriage; The material aggregation constraint limits the number of workpieces of the same material to occupy a continuous processing sequence interval.

3. The MOEA-D steel plate blanking scheduling optimization method as described in claim 1, characterized in that, The lengths of the machine coding layer and the process timing coding layer are both the product of the number of workpieces to be processed and the number of processes for a single workpiece; the three processes for the same workpiece in the machine coding layer correspond to the same cutting machine number and appear three times consecutively in the process timing coding layer.

4. The MOEA-D steel plate blanking scheduling optimization method as described in claim 1, characterized in that, The process of dividing workpieces into material groups according to material grade and plate thickness specifications, arranging workpieces in the same group continuously in the process timing coding layer, and generating an initial population based on the material workpiece groups includes: Randomly arrange the workpiece groups of each material, and then randomly arrange the workpieces within each material group; according to the order of loading, cutting and unloading processes, continuously write the workpiece number of each workpiece into the process sequence coding layer. The corresponding machine code is determined from the optional cutting machines for each process to form the initial chromosome that complies with material aggregation constraints; The initial chromosome is generated repeatedly until the number of chromosomes reaches the preset population size, thus forming the initial population.

5. The MOEA-D steel plate blanking scheduling optimization method as described in claim 1, characterized in that, The reconstruction process of the timing of each process includes: Based on the cutting machine allocation recorded in the machine coding layer, the workpieces and their processes allocated to the same cutting machine are clustered to form a serial processing sequence corresponding to each cutting machine; The sequential relationship of workpieces in each serial processing sequence is determined based on the process timing coding layer. Starting from the first workpiece in each serial processing sequence, the start time and completion time of each process are calculated recursively according to the order of loading, cutting and unloading processes; Specifically, for adjacent preceding and subsequent workpieces on the same cutting machine, the fixed delay starts when the cutting process of the preceding workpiece is completed and covers the process of receiving, transferring, and sorting materials by the traverse car.

6. The MOEA-D steel plate cutting scheduling optimization method as described in claim 1, characterized in that, The iterative optimization process of MOEA-D includes: Using a complete material workpiece group as a cross-operation unit, the process timing code segment and machine code segment corresponding to the material workpiece group are synchronously exchanged between parent chromosomes. Select at least one workpiece from the selected material workpiece group, change the cutting machine assignment of the workpiece, and simultaneously update the machine code corresponding to the three processes of the workpiece. The process timing is reconstructed for the offspring chromosomes generated after crossover and mutation. The total completion time and workpiece order deviation of the offspring chromosomes are calculated, and the population is updated based on the weight vector, neighborhood, and Chebyshev aggregation value.

7. The MOEA-D steel plate cutting scheduling optimization method as described in claim 1, characterized in that, The process of determining the target scheduling solution includes: The iteration terminates when any of the following conditions are met: (i) the number of iterations reaches the preset maximum number of iterations; (ii) the improvement of the Pareto front within a preset number of consecutive algebras does not reach the preset improvement threshold. After the iteration is terminated, the total completion time and work order deviation of each Pareto optimal solution obtained by the iteration are normalized. The normalized total completion time and work order deviation are weighted according to the preset preference weight, and the Pareto optimal solution with the smallest weighted evaluation value is determined as the target scheduling solution.

8. A MOEA-D steel plate blanking scheduling optimization device with dual process constraints, characterized in that, include: The data acquisition and constraint modeling module is used to acquire the material grade, plate thickness, processing time of three processes, and fixed delay of the optional cutting machine and transverse carriage of the steel plate workpiece to be processed, and to establish scheduling constraints that allow each workpiece to complete loading, cutting and unloading in sequence, and bind the three processes of the same workpiece to the same cutting machine. A dual-layer coding module is used to construct a dual-layer chromosome including a machine coding layer and a process timing coding layer; wherein, the machine coding layer records the cutting machine allocation for each process, and the process timing coding layer records the workpiece processing sequence; The population initialization module is used to divide the workpieces into material groups according to the material grade and plate thickness specifications, so that the workpieces in the same group are arranged continuously in the process timing coding layer, and to generate an initial population based on the material workpiece groups. The timing reconstruction module is used to cluster equipment according to the machine coding layer, determine the workpiece sequence relationship in each cutting machine according to the timing coding layer of the process, and ensure that the loading start time of the subsequent workpiece is not earlier than the sum of the cutting completion time of the previous workpiece and the fixed delay of the transverse car, so as to reconstruct the timing of each process. The iterative optimization module is used to perform equipment allocation variation within the same group, with the material workpiece group as the cross operation unit, and to reconstruct the process timing of the mutated chromosome. The total completion time and workpiece sequence deviation are iteratively optimized using MOEA-D. The decoding output module is used to determine the target scheduling solution from the Pareto optimal solution obtained through iteration based on the preset preference weight of the total completion time and the workpiece sequence deviation after the termination condition is met, and to decode the target scheduling solution to output the workpiece processing sequence of each cutting machine and the start and completion times of each process.

9. A storage medium, characterized in that, It stores a computer program, wherein when the computer program is executed by a processor, it implements the MOEA-D steel plate blanking scheduling optimization method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, The electronic device includes: The processor, the memory, and the program instructions stored in the memory and executable by the processor, wherein when the processor executes the program instructions, it implements the MOEA-D steel plate blanking scheduling optimization method as described in any one of claims 1 to 7.

Images