Chemical product production plan generation method and device
By sorting and cost-based screening of chemical product production plans, the problem of inefficiency in multi-product design for chemical production equipment was solved, and efficient and low-cost production plan generation was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-27
- Publication Date
- 2026-03-27
AI Technical Summary
In chemical production equipment, when formulating production plans for multiple product brands, the existing exhaustive method is time-consuming and difficult to effectively select the optimal solution. Especially when there are many product brands and complex switching rules, the workload is huge and errors are easy to occur.
By obtaining the product brand of the target production equipment, sorting them according to the predetermined switching rules, screening the initial candidate plans and inserting production processes, and combining two cost screenings, a chemical product production plan that meets the switching rules is generated.
It improves the efficiency of generating production plans for chemical products, ensures that the plans are low-cost and highly feasible, and reduces the computational scale and error rate.
Smart Images
Figure CN121745511A_ABST
Abstract
Description
Technical Field
[0001] This article relates to the field of chemical product manufacturing, and in particular to a method and apparatus for generating chemical product production plans. Background Technology
[0002] Chemical products are the main products and profit source for refining and chemical enterprises. A set of chemical production equipment can produce multiple grades of chemical products. For example, a certain production equipment can produce polypropylene, and the corresponding product grades of polypropylene include F401, T10F, H02M, etc. Each set of chemical production equipment has its own corresponding grade switching rules, switching costs, transition materials, shutdown conditions, etc. Currently, when formulating production plans for chemical production equipment, an exhaustive method combined with production experience is usually used to obtain the production order of all product grades. When the number of product grades is small, the final production plan can be obtained from all the exhaustive product grade sequences through production experience or manual calculation. However, when the number of product grades is large, such as needing to produce 10 product grades of polypropylene, and the product grades can be arbitrarily switched between each other in production position, feasible solutions include... It is obviously unrealistic and extremely time-consuming to exhaustively enumerate such a large number of sorting schemes. Even if all the exhaustive sorting schemes were obtained, the workload of considering the brand switching rules and switching costs among so many schemes would be enormous, making it difficult to select the final production plan from these schemes. Summary of the Invention
[0003] This application provides a method and apparatus for generating chemical product production plans, which can quickly and efficiently obtain low-cost chemical product production plans that comply with grade switching rules.
[0004] On one hand, embodiments of this application provide a method for generating a chemical product production plan, including: Obtain the product grades corresponding to multiple chemical products to be produced by the target production equipment within the target production planning cycle; According to the predetermined brand switching rules, the production sequence of the product brands is sorted to generate an initial candidate plan; An initial production plan is obtained by filtering all the initial candidate plans based on a first switching cost, wherein the first switching cost includes the sum of the switching costs when switching between any two product brands that are ranked adjacent. Based on the grade switching rules, production process grades are added to the initial production plan to obtain a complete production plan; All the complete production plans are screened based on the second switching cost to obtain the chemical product production plan for the target production plan cycle. The second switching cost includes the sum of the switching costs when switching between every two adjacent grades. The grades include the product grade and the production process grade.
[0005] On the other hand, embodiments of this application also provide a chemical product production plan generation device, including a memory and a processor; The memory is used to store the chemical product production plan generation program; The processor is used to read the chemical product production plan generation program and perform the chemical product production plan generation method as described in the above embodiments.
[0006] Compared with related technologies, the chemical product production plan generation method and apparatus of this application can quickly obtain a low-cost and highly feasible chemical product production plan by sorting the production sequence according to the grade switching rules, inserting production processes, and performing two cost screenings, which greatly improves the efficiency of obtaining chemical product production plans.
[0007] Other features and advantages of this application will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the application. Other advantages of this application can be realized and obtained by means of the solutions described in the description and the accompanying drawings. Attached Figure Description
[0008] The accompanying drawings are used to provide an understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.
[0009] Figure 1 This is a flowchart of a method for generating a chemical product production plan according to an embodiment of this application; Figure 2 This is a schematic diagram of a chemical product production plan generation device according to an embodiment of this application. Detailed Implementation
[0010] This application describes several embodiments, but these descriptions are exemplary and not restrictive, and it will be apparent to those skilled in the art that many more embodiments and implementations are possible within the scope of the embodiments described herein. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are also possible. Unless specifically limited, any feature or element of any embodiment may be used in combination with, or may replace, any feature or element of any other embodiment.
[0011] This application includes and contemplates combinations of features and elements known to those skilled in the art. The embodiments, features, and elements disclosed in this application can also be combined with any conventional features or elements to form unique inventive solutions. Any feature or element of any embodiment can also be combined with features or elements from other inventive solutions to form another unique inventive solution. Therefore, it should be understood that any feature shown and / or discussed in this application can be implemented individually or in any suitable combination. Therefore, the embodiments are not limited except by the limitations imposed by the appended claims and their equivalents. Furthermore, various modifications and changes can be made within the scope of the appended claims.
[0012] Furthermore, in describing representative embodiments, the specification may have presented methods and / or processes as a specific sequence of steps. However, the method or process should not be limited to the specific order of steps described herein, to the extent that it does not depend on such a specific order. As will be understood by those skilled in the art, other sequences of steps are also possible. Therefore, the specific order of steps set forth in the specification should not be construed as a limitation of the claims. Moreover, the claims concerning the method and / or process should not be limited to the steps performed in the written order, and those skilled in the art will readily understand that these orders can be varied and still remain within the spirit and scope of the embodiments of this application.
[0013] In refining and chemical enterprises, a single production line may need to produce multiple grades of products. Using an exhaustive approach to list all production schemes and then selecting feasible plans is extremely time-consuming and inefficient. For example, if a production line needs to produce 10 grades of polypropylene, and all grades can be switched between each other... It's clearly impractical to exhaustively list all possible solutions for such a large number of options. Even considering the brand name switching rules, finding the final production plan implementation scheme is still very difficult. For example, the brand name switching rules only specify the production sequence for four product brands, and there are no clear switching rules for the other six product brands. Even without considering special production processes such as downtime or maintenance for certain brands, it's still possible to exhaustively list all possible solutions. Selecting the final production plan implementation scheme from these 720 schemes is quite challenging; moreover, the exhaustive method is not only time-consuming, but also prone to omissions or confusion. Even if a lot of time is spent exhaustively listing all the schemes, the workload of finally determining a production plan from so many schemes is still enormous.
[0014] This application provides a method for generating a chemical product production plan, including steps S1000-S5000, such as... Figure 1 As shown: S1000: Obtain the product grade corresponding to multiple chemical products to be produced by the target production equipment within the target production plan cycle; S2000: According to the predetermined brand switching rules, sort the production sequence of the product brands to generate an initial candidate plan; S3000: Filter all the initial candidate plans according to the first switching cost to obtain an initial production plan, wherein the first switching cost includes the sum of the switching costs when switching between any two adjacent product brands; S4000: Based on the grade switching rules, add production process grades to the initial production plan to obtain a complete production plan; S5000: Filter all the complete production plans according to the second switching cost to obtain the chemical product production plan for the target production plan cycle, wherein the second switching cost includes the sum of the switching costs when switching between every two adjacent grades, and the grade includes the product grade and the production process grade.
[0015] In this embodiment, the target production equipment is capable of producing chemical products with multiple product grades of a chemical material. The chemical material can be any one of materials such as polypropylene, polyolefin, or synthetic rubber. For example, the target chemical product can produce polypropylene material. Polypropylene material is a common plastic material with multiple product grades. Different product grades represent different uses, properties, and production processes of polypropylene. Multiple polypropylene product grades include F401, T10F, H02M, etc. The target production plan cycle is a certain production cycle of the target production equipment, such as one month as a production cycle. The chemical product production plan for a certain month is generated for the target production equipment through steps S1000-S5000. The examples of chemical materials, product grades, and target production plan cycles mentioned above are all exemplary descriptions and are not intended to limit this application. They will not be elaborated further.
[0016] In this embodiment, the predetermined grade switching rule is the grade switching rule corresponding to the target production equipment. That is, the grade switching rules corresponding to different target production equipment are also different. The grade switching rule includes the order of product grade and production process grade. When the target production equipment is producing chemical products, after the product of one grade is produced, it is necessary to operate according to the grade switching rule in order to enter the production of the product corresponding to the next product grade.
[0017] In this embodiment, the first switching cost includes the switching cost between all adjacent product brands. For example, if the target production equipment produces X product brands in a total of X product brands within the target production plan cycle, after sorting the X product brands, there are X-1 groups of adjacent product brands. The first switching cost includes the switching cost of the X-1 groups of adjacent product brands.
[0018] In this embodiment, the production process designation in step S4000 may include at least one of the following: transition material process designation, shutdown process designation, inspection process designation, and maintenance process designation. One or more production process designations are inserted between two adjacent product designations in the initial production plan.
[0019] In this embodiment, the second switching cost includes the switching cost between all adjacent brands, including the switching cost between adjacent product brands and production process brands, and the switching cost between adjacent production process brands; for example, the complete production plan includes Y brands arranged in sequence, the Y brands include product brands and production process brands, including Y-1 groups of adjacent brands, and the second switching cost includes the switching cost of Y-1 groups of adjacent brands.
[0020] In this embodiment, the first switching cost and the second switching cost represent different switching costs. The first switching cost only includes the switching cost between adjacent product brands; the second switching cost includes the switching cost between adjacent brands, which includes adjacent product brands and production process brands, two adjacent production process brands, and adjacent production process brands and product brands.
[0021] In this embodiment, the first switching cost and the second switching cost can be economic costs or time costs; or they can be a combination of economic costs and time costs, with a first weight set for the economic cost and a second weight set for the time cost, and the sum of the first weight and the second weight is 1.
[0022] In this embodiment, step S3000 performs a first cost screening based on the first switching cost, and step S5000 performs a second cost screening based on the second switching cost. That is, the chemical product production plan generation method in this embodiment performs two cost screenings. The object of the first cost screening is the initial candidate plan, which only includes the sorting of product brands and does not include the production process brands. This avoids the increase in computational scale caused by the introduction of production process brands. Cost screening based on a large number of brands would lead to an exponential increase in computational scale, thus improving the efficiency of obtaining the initial production plan. Furthermore, after the first cost screening, the production process brands are added by executing step S4000, which reduces the computational amount of the second cost screening when executing step S5000, further improving the efficiency of obtaining the chemical product production plan.
[0023] The chemical product production plan generation method in this embodiment, by sorting the production sequence according to the grade switching rules, inserting production processes, and performing two cost screenings, can quickly obtain a low-cost chemical product production plan that conforms to the grade switching rules, greatly improving the efficiency of obtaining chemical product production plans.
[0024] In one exemplary embodiment, step S2000 may include steps S2100-S2300: S2100: Determine the total number M of the product grades and the switching difficulty of the grade switching rules; S2200: When M is greater than the quantity threshold and the switching difficulty is less than the switching difficulty threshold, the production order of all product brands is sorted according to the heuristic algorithm to generate an initial candidate plan; S2300: When M is not greater than the quantity threshold, or the switching difficulty is not less than the switching difficulty threshold, sort the production order of all product brands according to the precise method and the brand switching rules to generate an initial candidate plan.
[0025] In this embodiment, when executing step S2100, the switching difficulty of the brand name switching rule is determined according to the number of brands included in the brand name switching rule and all adjacent brand name switching rules. Adjacent brands include an adjacent product brand and a production process brand, or two adjacent production process brands. For example, if the number of brands included in the brand name switching rule is 7, and the order of the brands included in the brand name switching rule is: L1, M1, L2, L3, M2, L4, M3, where L1, L2, L3, and L4 are production process brands, and M1, M2, and M3 are product brands, then the brand name switching rule includes 7-1=6 adjacent brand name switching rules, namely, the switching rule between L1 and M1, the switching rule between M1 and L2, the switching rule between L2 and L3, the switching rule between L3 and M2, the switching rule between M2 and L4, and the switching rule between L4 and M3. The switching difficulty of the brand name switching rule is the sum of the switching difficulties corresponding to these 6 switching rules.
[0026] In this embodiment, when the total number of product brands M is large and the switching difficulty of the brand switching rules is small, it is impossible to generate all feasible initial candidate plans within a limited time. Therefore, a heuristic algorithm is used to generate initial candidate plans. The product brand switching process is optimized through the heuristic algorithm to reduce switching time and raw material consumption. The heuristic algorithm can be any one of the following: simulated annealing algorithm, genetic algorithm, particle swarm optimization algorithm, ant colony algorithm, and artificial fish swarm algorithm. The above heuristic algorithms are described exemplarily and are not intended to limit this application. They will not be described in detail hereafter.
[0027] In this embodiment, when the total number of product brands M is small, or when the brand switching rules are difficult to switch, if a feasible initial candidate plan can be generated within a limited time, then an accurate method is used to generate the initial candidate plan. During the generation of the initial candidate plan, the remaining product brands are traversed sequentially according to the brand switching rules until all product brands are in the initial candidate plan.
[0028] In one exemplary embodiment, step S2200, "sorting the production order of all product brands according to a heuristic algorithm to generate an initial candidate plan," may include steps S2210-S2250: S2210: Among the product brands, select a pre-set first product brand or randomly select a product brand as the first product brand in the sorting order; S2220: Continue sorting the M-1 product brands other than the first-ranked product brand according to a predetermined strategy to generate one or more first sequences. The predetermined strategy includes a greedy strategy or a random addition strategy. S2230: Calculate the initial first switching cost C1 corresponding to each of the first sequences. If C1 is less than the switching cost threshold C... gIf C1 is not less than C, then retain the first sequence; g If so, then the first sequence is removed; S2240: For each retained first sequence, perform the following operation: Iterate the first sequence according to the heuristic algorithm until a second sequence is obtained, or until the second sequence is still not obtained by the Nth iteration, then discard the first sequence; where N is the maximum threshold for the number of iterations, and the first switching cost C2 of the second sequence is less than C1; S2250: If one or more of the second sequences exist, then all of the second sequences are used as the initial candidate plans; if no second sequence is obtained, then the production order of all product brands is sorted according to the precise method and the brand switching rules to generate an initial candidate plan.
[0029] In this embodiment, if there is a pre-set first product brand, then the first product brand is used as the first product brand in the ranking. For example, in a certain monthly production plan cycle, the product at the beginning of the month (the product brand that needs to be produced first in that month) is already fixed, then the product brand corresponding to the product at the beginning of the month is used as the first product brand, and the first product brand is set as the first in the initial candidate plan; if there is no pre-set first product brand, then a product brand is randomly selected from M product brands as the first product brand in the ranking.
[0030] In one embodiment of this example, when executing step S2220, for product brands ranked from the second to the Mth position, a greedy strategy can be used to determine the production order. First, according to the order of A product brands included in the brand switching rules, the positions of these A product brands are determined. Then, the remaining MA-1 product brands are randomly added to the remaining positions to generate the first sequence.
[0031] In this embodiment, for example, according to the brand name switching rules, if the production order of product brand A is adjacent to each other and is second only to the first product brand, then product brand A is placed in the second to the A+1th position, and the remaining MA-1 product brands are randomly added and placed in the A+2th to the Mth position.
[0032] In this embodiment, for example, according to the brand name switching rules, the production order of product brands A is adjacent to each other, but not necessarily adjacent to the first product brand. Then, product brands A can be placed in any A consecutive positions other than the first position. For example, product brands A can be placed in the second to the A+1th position, or in the third to the A+2th position, or in the fourth to the A+3th position, ..., or in the M-A+1th to the Mth position. The remaining MA-1 product brands are randomly added to the remaining MA-1 positions.
[0033] In this embodiment, for example, according to the brand name switching rule, product brand A is divided into two sequence segments. The two sequence segments can be discontinuous. The product brand names within the sequence segments are in a fixed production order of being adjacent to each other: product brand name sequence segment A1 and product brand name sequence segment A2, where A = A1 + A2. Then, product brand name sequence segment A1 is placed in any consecutive A1 positions except the first position, product brand name sequence segment A2 is placed in any other consecutive A2 positions except the first position and A1 position, and the remaining MA-1 product brand names are randomly added to the remaining MA-1 positions.
[0034] In another embodiment of this example, when performing step S2220, for product brands ranked from the second to the Mth position, a random addition strategy can be used to determine the production order, and the product brands except the first ranked brand can be randomly added to the second to the Mth position to generate the first sequence.
[0035] In this embodiment, during step S2230, for each first sequence, the initial first switching cost C1 is calculated, and C1 is not less than the switching cost threshold C. g The first sequence is discarded as an infeasible sequence, and C1 can be less than the switching cost threshold C. g The first sequence is retained, and a feasible sequence set is generated using all retained first sequences. Step S2240 is executed for each first sequence in the feasible sequence set; if the first switching cost C1 corresponding to all first sequences is not less than the switching cost threshold C g If the feasible sequence set is empty, the method is converted to generate an initial candidate plan using the exact method. This involves sorting the production order of all product brands according to the exact method and the brand switching rules to generate an initial candidate plan.
[0036] In this embodiment, the purpose of step S2240 is to obtain a second sequence with a lower initial first switching cost C1 for the retained first sequence using a heuristic algorithm, that is, to obtain a second sequence with a better first switching cost than the first sequence. If at least one second sequence is obtained, the initial candidate plan is successfully generated; if no second sequence is obtained, it means that the heuristic algorithm has failed to generate the initial candidate plan, and the method is switched to an exact method to generate the initial candidate plan.
[0037] In one exemplary embodiment, step S2240, "iterating over the first sequence according to the heuristic algorithm until a second sequence is obtained, or until the second sequence is still not obtained by the Nth iteration, then discarding the first sequence," may include steps S2241-S2243: S2241: In the first iteration, take the first sequence as the current sequence and set the current iteration count to 0; Each iteration may include steps S2242-S2243: S2242: Perform a neighborhood operation on the current sequence to obtain the current operation sequence, wherein the neighborhood operation includes random swap, random flip, or greedy swap; S2243: Calculate the first switching cost C of the current operation sequence. n If C n If C is less than C1, then the current operation sequence is taken as the second sequence; if C n If the current iteration number is not less than C1, increment it by 1. If the current iteration number is less than N, use the current operation sequence as the current sequence for the next iteration. If the current iteration number is equal to N, discard the first sequence.
[0038] In this embodiment, if the first product brand is a pre-set first product brand, the position of the first product brand is fixed and it does not participate in the neighborhood operation. For example, if the first product brand in the current sequence is a product that has been pre-set at the beginning of the month in the production plan cycle of a certain month, the position of the first product brand is fixed and it does not participate in the neighborhood operation. Only other product brands other than the first product brand are subject to the neighborhood operation.
[0039] In this embodiment, before performing neighborhood operations on the current sequence, the current sequence can be divided into two columns in the following way: the product brand sequence in the first half of the current sequence is taken as the first column, and the product brand sequence in the second half of the current sequence is taken as the second column. For example, if the current sequence has 20 product brands, the first 10 product brands in the sequence are taken as the first column, and the last 10 product brands in the sequence are taken as the second column.
[0040] In one embodiment of this example, the neighborhood operation in step S2242 can be a random swap, in which a product brand is randomly selected from the first column and the second column respectively, and the positions of the two product brands are swapped. Then, the newly generated first column and the second column are connected to generate the current operation sequence.
[0041] In another embodiment of this example, the neighborhood operation in step S2242 can be a random flip, whereby a product brand is randomly selected from the first column and the second column respectively, and the product brands located between these two product brands are flipped sequentially to generate the current operation sequence; for example, the product brand order of the current sequence is 1,2,3,4,5,6,7,8,9,10, the first column is 1,2,3,4,5, and the second column is 6,7,8,9,10. Product brand 3 is randomly selected from the first column, and product brand 8 is randomly selected from the second column. Product brands 4-7 are flipped sequentially to generate the current operation sequence 1,2,3,7,6,5,4,8,9,10.
[0042] In another implementation of this embodiment, the neighborhood operation in step S2242 can be a greedy swap. A product brand P is randomly selected from the first column, and each product brand Q following product brand P in the current sequence is traversed one by one, and P and Q are swapped. If the sequence after the swap satisfies the brand switching rule during a certain traversal, the sequence after the swap is taken as the current operation sequence. If the sequence after the swap does not satisfy the brand switching rule, the next traversal is performed.
[0043] In an exemplary embodiment, the step S2300 and step S2250, "sorting the production order of all product brands according to the precise method and the brand switching rules to generate an initial candidate plan," may include steps S2310-S2340: S2310: Among the product brands, a pre-set first product brand is selected as the first product brand in the ranking, or each product brand is selected as the first product brand in the ranking; and the other product brands besides the first product brand in the ranking are selected as the first set of remaining product brands; S2320: Execute the first-level precise method, including: setting the product brand number ranked first in the first position, and generating a first-level sequence by combining M-1 empty positions; S2330: For each of the first layer sequences, according to the precise method and the brand name switching rule, sort the product brands in the first remaining product brand name set in the M-1 empty positions to obtain the third sequence; S2340: Use all of the third sequences as the initial candidate plans.
[0044] In this embodiment, if there is a pre-set first product brand, then the first product brand is used as the first product brand in the ranking. For example, in a certain month's production plan cycle, the product at the beginning of the month (the product brand that needs to be produced first in that month) has been fixed, then the product brand corresponding to the product at the beginning of the month is used as the first product brand, and the first product brand is set as the first position in the initial candidate plan. In this embodiment, if there is no pre-set first product brand, then M product brands are used as the first product brand, and combined with the M-1 empty positions after the first position to generate a first-level sequence. For each first-level sequence, feasible product brands located at the second to the Mth position can be generated sequentially according to the brand switching rules; for example, if there are 3 brands 1, 2, 3, and no pre-set first product brand, when executing step S2310, three first-level sequences can be generated, namely {1,0,0}, {2,0,0}, and {3,0,0}, where 0 represents an empty position; for the first-level sequence {1,0,0}, {2,3} is the first remaining In step S2330, the remaining product brand set is sorted in two empty positions according to the brand switching rule. For the first-level sequence {2,0,0}, {1,3} is the first remaining product brand set. In step S2330, {1,3} is sorted in two empty positions according to the brand switching rule. For the first-level sequence {3,0,0}, {1,2} is the first remaining product brand set. In step S2330, {1,2} is sorted in two empty positions according to the brand switching rule. The obtained third sequence is a feasible production plan that conforms to the brand switching rule. All of the third sequences are used as the initial candidate plans.
[0045] In this embodiment, since the switching of the grade switching rules is more difficult when using the precise method, it can output all feasible production plans more accurately, and its efficiency in obtaining feasible production plans is significantly higher than that of the exhaustive method.
[0046] In one exemplary embodiment, step S2330 may include steps S2331-S2332: For each of the first-level sequences, the second to Mth level precision methods are executed sequentially. The first-level sequence is used as the current level sequence for the second-level precision method, and the first set of remaining product brands is used as the current set of remaining product brands for the second-level precision method. Here, m = 2, 3, ..., M, and the mth level precision method includes: S2331: According to the brand name switching rule, select the feasible m-th product brand name at the m-th position in the current remaining product brand name set, and set the m-th product brand name at the m-th position to generate one or more m-th layer sequences; S2332: When m≤M-1, the remaining product brands in the current remaining product brand set, excluding the m-th product brand, are used as the current remaining product brand set for the (m+1)-th level precise method, and the (m+1)-th level precise method is performed; when m=M, one or more M-th level sequences generated are used as the third sequence.
[0047] In this embodiment, when performing step S2331, according to the brand name switching rule, there can be multiple feasible m-th product brands. The multiple feasible m-th product brands are placed at the m-th position respectively to generate multiple corresponding m-th layer sequences.
[0048] In one exemplary embodiment, step S3000 may include steps S3100-S3300: S3100: For each of the initial candidate plans, calculate the corresponding first switching cost; S3200: Sort all the initial candidate plans in ascending order of the first switching cost to obtain the first cost sequence; S3300: Select one or more of the initial candidate plans from the first cost sequence according to the first predetermined principle, as the initial production plan.
[0049] In this embodiment, for each initial candidate plan, the first switching cost includes the sum of the switching costs when switching between any two adjacent product brands.
[0050] In one embodiment of this invention, the first predetermined principle may be to select the first initial candidate plan ranked first in the first cost sequence, that is, to select the initial candidate plan with the lowest first switching cost as the initial production plan.
[0051] In another embodiment of this invention, a selection quantity threshold S can be preset. g Select the top S from the first cost sequence g One initial candidate plan is selected as the initial production plan.
[0052] In another embodiment of this invention, a first threshold can be preset, and an initial candidate plan with a first switching cost less than the first threshold can be selected from the first cost sequence as the initial production plan.
[0053] In one exemplary embodiment, step S4000 may include step S4100: For each of the initial production plans, perform step S4100: S4100: Based on the grade switching rules, insert at least one of the following production process grades into the initial production plan: transition material process grade, shutdown process grade, overhaul process grade, and maintenance process grade to obtain a complete production plan.
[0054] In this embodiment, all inserted production process grades must follow the grade switching rules. When an inspection or maintenance process is required within the target production plan cycle, an inspection or maintenance grade is inserted. The position of the inserted process grade must meet the stop grade requirements before inspection or maintenance, and the start grade requirements after inspection or maintenance.
[0055] In this embodiment, if the initial production plan requires the production of transitional materials between two adjacent product grades, then a transitional material process grade needs to be added between these two adjacent product grades. For example, not producing transitional materials may easily cause the target production equipment to explode, or the production condition parameters of the former product grade cannot be seamlessly switched to the production condition parameters of the latter product grade, and parameter transition adjustments must be made. During this period, transitional materials will inevitably be generated.
[0056] In this embodiment, if a shutdown is required between the production of two adjacent product grades in the initial production plan, a shutdown process grade needs to be added between these two adjacent product grades. For example, after the production of the first product grade, the target production equipment must be cleaned or the catalyst replaced before the production of the second product grade can begin.
[0057] In this embodiment, if the production plan requires maintenance of the target production equipment between the production of two adjacent product brands, then a maintenance process brand needs to be added between these two adjacent product brands. For example, based on production experience, when producing the product of the first product brand, if a certain part operates at high speed and high temperature, the lubricating oil on its surface will fail. In this case, the part needs to be maintained, the residue on its surface needs to be removed and the lubricating oil needs to be reapplied, so as to ensure the smooth production of the product of the second product brand.
[0058] In this embodiment, if the initial production plan requires maintenance of the target production equipment between the production of two adjacent product brands, then a maintenance process brand needs to be added between these two adjacent product brands. For example, based on production experience, when producing the product of the first product brand, a certain part is severely worn, and continuing production cannot guarantee the quality of the product of the second product brand, or there is a risk of safety accidents. Therefore, the part must be replaced before the production of the product of the second product brand can continue.
[0059] In one exemplary embodiment, step S5000 may include steps S5100-S5300: S5100: For each complete production plan, calculate the corresponding second switching cost; S5200: Sort all the complete production plans in ascending order of the second switching cost to obtain the second cost sequence; S5300: Select one or more of the complete production plans from the second cost sequence according to the second predetermined principle, as the production plan for the chemical product.
[0060] In this embodiment, the second switching cost includes the sum of the switching costs when switching between every two adjacent grades, and the grade includes the product grade and the production process grade.
[0061] In one embodiment of this invention, the second predetermined principle may be to select the complete production plan ranked first in the second cost sequence, that is, to select the complete production plan with the lowest second switching cost, as the chemical product production plan.
[0062] In another embodiment of this invention, a selection quantity threshold S can be preset. g Select the top S from the second cost sequence g A complete production plan, serving as a production plan for chemical products.
[0063] In another embodiment of this invention, a second threshold can be preset, and a complete production plan with a second switching cost less than the second threshold can be selected from the second cost sequence as the chemical product production plan.
[0064] To illustrate the chemical product production plan generation method of this application embodiment, a specific example is described in detail below.
[0065] The specific example of the chemical product production plan generation method includes steps S1-S11: S1: Obtain the product grades corresponding to multiple chemical products to be produced by the target production equipment within the target production planning cycle; S2: Determine the total number M of product brands and the difficulty of switching brand switching rules; S3: Determine the relationship between M and the quantity threshold, and the relationship between the switching difficulty and the switching difficulty threshold. If M > quantity threshold and switching difficulty < switching difficulty threshold, proceed to step S4; if M ≤ quantity threshold or switching difficulty ≥ switching difficulty threshold, proceed to step S5. S4: Select a heuristic algorithm to sort the production order of all product brands to generate an initial candidate plan; S41: Determine if there is a pre-set first product brand name. If there is, proceed to step S42. If not, proceed to step S43. S42: Use the pre-set first product brand as the first product brand in the sorting order, and proceed to step S44; S43: Randomly select a product brand as the first product brand in the sorting, and proceed to step S44; S44: Continue sorting the M-1 product brands other than the first-ranked product brand according to a greedy strategy or a random addition strategy to generate one or more first sequences; S45: Calculate the initial first switching cost C1 corresponding to each of the first sequences. If C1 is less than the switching cost threshold C... g If C1 is not less than C, then retain the first sequence; g If so, then the first sequence is removed; S46: For each retained first sequence, perform the following operation: Iterate the first sequence according to the heuristic algorithm; if a second sequence with a first switching cost less than C1 is obtained, add the second sequence to the initial candidate plan set; if no second sequence with a first switching cost less than C1 is obtained, delete the first sequence. S47: Determine if the initial candidate plan set is empty. If not, proceed to step S6; if yes, proceed to step S5. S5: Select the precise method and sort the production sequence of all product brands according to the brand switching rules to generate an initial candidate plan; S51: Determine if there is a pre-set first product brand name. If there is, proceed to step S52. If not, proceed to step S53. S52: Use the pre-set first product brand as the first product brand in the sorting order, and proceed to step S54; S53: Select each product brand as the first product brand in the sorting order and proceed to step S54; S54: Take all product brands other than the first-ranked product brand as the first set of remaining product brands; S55: Execute the first-level precise method, set the product brand number ranked first in the first position, and generate the first-level sequence by combining M-1 empty positions; S56: For each of the first layer sequences, according to the brand name switching rules, sort the product brands in the first remaining product brand name set in the M-1 empty positions to obtain the third sequence; S57: Use all of the aforementioned third sequences as the initial candidate plan set; S6: For each of the initial candidate plans, calculate the corresponding first switching cost; S7: Sort all the initial candidate plans in ascending order of the first switching cost to obtain the first cost sequence; S8: Select one or more of the initial candidate plans from the first cost sequence according to the first predetermined principle, and use them as the initial production plan.
[0066] S9: Based on the grade switching rules, insert at least one of the following production process grades into the initial production plan: transition material process grade, shutdown process grade, overhaul process grade, and maintenance process grade to obtain a complete production plan.
[0067] S10: For each complete production plan, calculate the corresponding second switching cost; S11: Sort all the complete production plans in ascending order of the second switching cost to obtain the second cost sequence; S12: Select one or more of the complete production plans from the second cost sequence according to the second predetermined principle, as the production plan for the chemical product.
[0068] This application also provides a chemical product production plan generation device, such as... Figure 2 As shown, it includes a processor and memory. The memory is used to store the chemical product production plan generation program; The processor is used to read the chemical product production plan generation program and perform the chemical product production plan generation method as described in the above embodiments.
[0069] It will be understood by those skilled in the art that all or some of the steps, systems, or apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term "computer storage medium" includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and can be accessed by a computer. Furthermore, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.
Claims
1. A method for generating a production plan for chemical products, characterized in that, include: Obtain the product grades corresponding to multiple chemical products to be produced by the target production equipment within the target production planning cycle; According to the predetermined brand switching rules, the production sequence of the product brands is sorted to generate an initial candidate plan; An initial production plan is obtained by filtering all the initial candidate plans based on a first switching cost, wherein the first switching cost includes the sum of the switching costs when switching between any two product brands that are ranked adjacent. Based on the grade switching rules, production process grades are added to the initial production plan to obtain a complete production plan; All the complete production plans are screened based on the second switching cost to obtain the chemical product production plan for the target production plan cycle. The second switching cost includes the sum of the switching costs when switching between every two adjacent grades. The grades include the product grade and the production process grade.
2. The method for generating a chemical product production plan as described in claim 1, characterized in that, The step of sorting the production sequence of the product brands according to the predetermined brand switching rules to generate an initial candidate plan includes: Determine the total number M of the product brands and the switching difficulty of the brand switching rules; When M is greater than the quantity threshold and the switching difficulty is less than the switching difficulty threshold, the production order of all product brands is sorted according to the heuristic algorithm to generate an initial candidate plan. When M is not greater than the quantity threshold, or the switching difficulty is not less than the switching difficulty threshold, the production order of all product brands is sorted according to the precise method and the brand switching rules to generate an initial candidate plan.
3. The method for generating a chemical product production plan as described in claim 2, characterized in that, The step of sorting the production order of all product brands according to a heuristic algorithm to generate an initial candidate plan includes: Among the product brands, a pre-set first product brand or a randomly selected product brand is chosen as the first product brand in the ranking. According to a predetermined strategy, the other M-1 product brands, excluding the first-ranked product brand, are sorted to generate one or more first sequences. The predetermined strategy includes a greedy strategy or a random addition strategy. Calculate the initial first handover cost C1 for each of the first sequences. If C1 is less than the handover cost threshold C... g If C1 is not less than C, then retain the first sequence; g If so, then the first sequence is removed; For each retained first sequence, perform the following operation: iterate the first sequence according to the heuristic algorithm until a second sequence is obtained, or until the second sequence is still not obtained by the Nth iteration, then discard the first sequence; where N is the maximum threshold for the number of iterations, and the first switching cost C2 of the second sequence is less than C1; If one or more of the second sequences exist, all of the second sequences are used as the initial candidate plans; if no second sequence is obtained, the production order of all product brands is sorted according to the precise method and the brand switching rules to generate the initial candidate plans.
4. The method for generating a chemical product production plan as described in claim 3, characterized in that, The step of iterating over the first sequence according to the heuristic algorithm until a second sequence is obtained, or until the second sequence is still not obtained by the Nth iteration, and then discarding the first sequence, includes: In the first iteration, the first sequence is taken as the current sequence, and the current iteration count is set to 0; Each iteration includes the following steps: Perform neighborhood operations on the current sequence to obtain the current operation sequence, wherein the neighborhood operations include random swap, random flip, or greedy swap; Calculate the first switching cost C of the current operation sequence. n If C n If C is less than C1, then the current operation sequence is taken as the second sequence; if C n If the current iteration number is not less than C1, increment it by 1. If the current iteration number is less than N, use the current operation sequence as the current sequence for the next iteration. If the current iteration number is equal to N, discard the first sequence.
5. The method for generating a chemical product production plan as described in claim 2 or 3, characterized in that, The step of sorting the production sequence of all product brands according to the precise method and the brand switching rules to generate an initial candidate plan includes: Among the product brands, a pre-set first product brand is selected as the first product brand in the ranking, or each product brand is selected as the first product brand in the ranking; the other product brands besides the first product brand in the ranking are selected as the first set of remaining product brands. Performing the first-level precise method includes: setting the product brand number ranked first in the first position, and generating a first-level sequence by combining M-1 empty positions; For each of the first-level sequences, according to the precise method and the brand name switching rule, the product brands in the first remaining product brand name set are sorted in the M-1 empty positions to obtain the third sequence; All of the third sequences are used as the initial candidate plans.
6. The method for generating a chemical product production plan as described in claim 5, characterized in that, For each of the first-level sequences, according to the precise method and the brand name switching rule, the product brands in the first remaining product brand name set are sorted in the M-1 empty positions to obtain the third sequence, including: For each of the first-level sequences, the second to Mth level precision methods are executed sequentially. The first-level sequence is used as the current level sequence for the second-level precision method, and the first set of remaining product brands is used as the current set of remaining product brands for the second-level precision method. Here, m = 2, 3, ..., M, and the mth level precision method includes: According to the brand name switching rule, select the feasible m-th product brand name at the m-th position from the current remaining product brand name set, and set the m-th product brand name at the m-th position to generate one or more m-th layer sequences; When m ≤ M-1, the remaining product brands in the current set of product brands other than the m-th product brand are used as the current set of product brands for the (m+1)-th level precise method, and the (m+1)-th level precise method is performed; when m = M, one or more M-th level sequences generated are used as the third sequence.
7. The method for generating a chemical product production plan as described in claim 1, characterized in that, The step of filtering all the initial candidate plans based on the first switching cost to obtain an initial production plan includes: For each of the initial candidate plans, calculate the corresponding first switching cost; Sort all the initial candidate plans in ascending order of the first switching cost to obtain the first cost sequence; According to the first predetermined principle, one or more of the initial candidate plans are selected from the first cost sequence as the initial production plan.
8. The method for generating a chemical product production plan as described in claim 1, characterized in that, The step of adding production process grades to the initial production plan based on the grade switching rules to obtain a complete production plan includes: For each of the initial production plans, perform the following steps: Based on the grade switching rules, at least one of the following production process grades is inserted into the initial production plan: transition material process grade, shutdown process grade, overhaul process grade, and maintenance process grade to obtain a complete production plan.
9. The method for generating a chemical product production plan as described in claim 1, characterized in that, The step of filtering all the complete production plans based on the second switching cost to obtain the chemical product production plan for the target production plan cycle includes: For each complete production plan, calculate the corresponding second switching cost; All the complete production plans are sorted in ascending order of the second switching cost to obtain the second cost sequence; In accordance with the second predetermined principle, one or more of the complete production plans are selected from the second cost sequence as the production plan for the chemical product.
10. A chemical product production plan generation device, comprising a memory and a processor, characterized in that: The memory is used to store the chemical product production plan generation program; The processor is configured to read the chemical product production plan generation program and perform the chemical product production plan generation method as described in any one of claims 1-9.