A hybrid flow shop scheduling method based on historical information ant colony algorithm

Abstract

The application discloses a mixed flow shop scheduling method based on a historical information ant colony algorithm, and relates to the technical field of glass workpiece processing. The application discloses a kind of based on historical information ant colony algorithm's mixed flow shop scheduling method, comprising:1 constructs the total time of glass workpiece processing and energy consumption model;2 set corresponding constraint condition, construct double objective mixed flow shop scheduling model;3 according to glass workpiece and machine relevant information, and using historical information ant colony algorithm to the double objective mixed flow shop scheduling model is solved, obtains production processing scheme, to carry out production processing to glass workpiece.The application can obtain the optimal production processing scheme combined with time consumption and energy consumption, so as to improve the production processing efficiency of glass workpiece.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of mixed flow shop scheduling optimization control and intelligent decision-making, in particular to a mixed flow shop scheduling method based on historical information ant colony algorithm. BACKGROUND

[0002] The mixed flow shop scheduling problem is an important research topic in the field of production scheduling, and is widely used in the semiconductor, glass production and steelmaking industries. Unlike traditional flow shops, each process in a mixed flow shop can have multiple machines, which can be of different types. The number of machines and processing time for different processes can also be different. The goal of mixed flow shop scheduling is usually to optimize production efficiency and reduce production cycle time by reasonably arranging the processing order of each task.

[0003] In recent years, global energy consumption problems have become increasingly prominent, and many countries and regions have formulated strict environmental protection regulations. Among them, the energy consumption of the manufacturing industry accounts for about 50% of the total. For energy-intensive factories, if energy consumption cannot be effectively reduced, they will face elimination, so managers need to consider energy-saving and efficient production methods. In view of this, how to balance production efficiency and energy consumption has become a problem worth studying.

[0004] Tempered glass is a type of glass product that has been specially treated to enhance its strength and impact resistance, and is widely used in the construction, automotive, electronic equipment, and home appliance industries. Tempered glass production is a typical manufacturing scenario, and the academic and industrial communities have conducted extensive research on single-objective problems such as reducing production time. However, the problem of considering both completion time and energy consumption has not been fully explored. SUMMARY

[0005] The present application is to solve the above-mentioned deficiencies in the prior art, and proposes a mixed flow shop intelligent scheduling method based on historical information ant colony algorithm and its application, in order to obtain an optimal production and processing scheme that combines time and energy consumption, thereby improving the efficiency and accuracy of shop scheduling.

[0006] To achieve the above-mentioned application purposes, the present application adopts the following technical solutions:

[0007] The present invention discloses an intelligent scheduling optimization method for a hybrid production line based on historical information ant colony algorithm. This method is characterized by its application in a production line where each glass workpiece, after being processed by a cutting parallel machine, enters a first buffer to wait. From the first buffer, each glass workpiece is then transported to a spraying parallel machine for processing, and then enters a second buffer to wait. From the second buffer, each glass workpiece is then transported to a batch processing machine to form batches. Each batch is then tempered on the batch processing machine until all glass workpieces have been processed. Either the cutting parallel machine or the spraying parallel machine is designated as the [number missing]th parallel machine. The intelligent scheduling optimization method for hybrid flow workshops, which is similar to a parallel machine, is performed according to the following steps:

[0008] Step 1: Establish the first objective function with the goal of maximizing the completion time. And a second objective function with the goal of minimizing energy consumption. Let the two objective functions be defined in the hybrid flow shop scheduling model, and let any o-th objective function be... , ;

[0009] Step 2: Establish the constraints for the hybrid flow workshop scheduling model;

[0010] Step 3: Use the historical information ant colony algorithm to solve the hybrid assembly line workshop scheduling model to obtain the production and processing plan.

[0011] The intelligent scheduling optimization method for hybrid assembly line workshops based on historical information ant colony algorithm described in this invention is characterized in that step 1 uses equations (1) and (2) to establish two objective functions for the hybrid assembly line workshop scheduling model:

[0012] (1)

[0013] (2)

[0014] In equation (1), To maximize the completion time, This indicates that the b-th batch is in the th batch. Completion time on the batch processing machine;

[0015] In equation (2), TEC represents energy consumption. The total number of time periods. This represents the total number of parallel machines. Let f(t) represent the total number of batch processors, and f(t) represent the amount of electricity consumed in the t-th time period. Indicates the first Tai Di Determine whether the parallel machine is in a working state during the t-th time period. If so, then let... =1; otherwise, let =0; Indicates the first Tai Di Is the parallel machine in standby state during the t-th time period? If so, then let... =1; otherwise, let =0; Indicates the first Tai Di Is the parallel machine in the off state during the t-th time period? If so, then let... =1; otherwise, let =0; Indicates the first Is the batch processing machine in operation during the t-th time period? If so, then... =1; otherwise, let =0; Indicates the first Is the batch processing machine in standby mode during the t-th time period? If so, then... =1; otherwise, let =0; Indicates the first Is the batch processing machine in a stopped state during the t-th time period? If so, then... =1; otherwise, let =0; , and They represent the first The energy consumption of a parallel machine in working, standby, and off states; , and These represent the energy consumption of the batch processor when it is in working, standby, and off states, respectively.

[0016] Furthermore, in step 2, the constraints of the hybrid flow shop scheduling model are established using equations (3) to (45):

[0017] (3)

[0018] (4)

[0019] (5)

[0020] (6)

[0021] (7)

[0022] (8)

[0023] (9)

[0024] (10)

[0025] (11)

[0026] (12)

[0027] (13) (14)

[0028] (15)

[0029] (16)

[0030] (17)

[0031] (18)

[0032] (19)

[0033] (20)

[0034] (21)

[0035] (22)

[0036] (23)

[0037] (24)

[0038] (25)

[0039] (26)

[0040] (27)

[0041] (28)

[0042] (29)

[0043] (30)

[0044] (31)

[0045] (32)

[0046] (33)

[0047] (34)

[0048] (35)

[0049] (36)

[0050] (37)

[0051] (38)

[0052] (39)

[0053] (40)

[0054] (41)

[0055] (42)

[0056] (43)

[0057] (44)

[0058] (45)

[0059] in formulas (3) - (45), denotes the processing time of the th glass workpiece on the th parallel machine, denotes the arrival time of the th glass workpiece, denotes the completion time of the th glass workpiece on the th parallel machine, denotes the size of the th glass workpiece, denotes the capacity of the th batcher, denotes the processing time of the th batch on the th batcher, denotes the start processing time of the th batch on the th batcher, denotes the start processing time of the The glass workpiece in the first Processing time on a batch processing machine; Indicates the first Is the glass workpiece in the [number]th ... Tai Di Processing on a parallel machine; if so, then let =1; otherwise, let =0; Indicates the first Is the glass workpiece in the [number]th ... Tai Di Processing on a parallel machine; if so, then let Otherwise, let ; This indicates whether the j-th glass workpiece is waiting in the first buffer during the t-th time period. If so, then let... Otherwise, let ; Indicates the first The qualification constraints for the first glass workpiece, i.e., whether it can be used in the first... Tai Di Process on a parallel machine; if possible, then let... Otherwise, let ; Indicates the first Is the glass workpiece in the [number]th ... The first batch processing machine If so, then let Otherwise, let ; This indicates the t-th time period. Is the batch in the [number]th [period]? Processed on a batch processing machine in Taiwan; if so, then... Otherwise, let ; Indicates the first Does each batch contain at least one glass workpiece? If so, then let... Otherwise, let ; For the i-th stage in the t-th time period Auxiliary binary variables for parallel machines This indicates the t-th time period. Auxiliary binary variables for batch processing; and These are the capacities of the first buffer and the second buffer, respectively. Indicates high battery level; Indicates low battery level; Indicates the end time period of high battery level; Indicates the end time period of low battery; Indicates the first Taiwan batch processing machine in the first If the system is in operation for a given time period, then... =1; otherwise, let =0; Indicates the first Tai Di Parallel machines in the first If the system is in operation for a given time period, then... =1; otherwise, let =0.

[0060] Furthermore, step 3 includes:

[0061] Step 3.1: Define the maximum number of iterations as... The total number of ants is Randomly generate the first preference vector With the second preference vector Initialize the heuristic information matrix; Define the current generation as k; Initialize k=1; Initialize the pheromone matrix of the kth generation. ;

[0062] Step 3.2: Set the ant's index a=1;

[0063] Step 3.3: Initialize the parameters of the glass workpiece and the machine, set the first taboo table DE1 in n dimensions, and assign an initial value to DE1 according to the sequence number of the glass workpiece. Let there be an n-dimensional second tabu list DE2 and an n-dimensional third tabu list DE3, and assign initial values ​​to both. n represents the total number of glass workpieces;

[0064] Step 3.4: When l=1 and DE1 is not (0,0,...,0), execute the... Scheduling of the first type of parallel machine, thereby achieving the first... Tai Di On a parallel machine, a glass workpiece is processed, and the processed glass workpiece is stored in buffer 1. The value of the index position of the processed glass workpiece in DE1 is set to 0, and the value of the index position of the processed glass workpiece in DE2 is modified to the index number of the processed glass workpiece itself; thus updating DE1 and DE2; then... The Middle The generation Tai Di Parallel Machine The oth target of the glass workpiece The pheromones are updated to obtain the first... The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece ;

[0065] Step 3.5: When l=2 and DE2 is not (0,0,...,0), execute the... Tai Di Scheduling of parallel machines, thus in the first... Tai Di On a parallel machine, a glass workpiece is processed, and the processed glass workpiece is stored in buffer 2. The index value of the processed glass workpiece in DE2 is set to 0, and the index value of the processed glass workpiece in DE3 is modified to the index number of the processed glass workpiece itself; thus updating DE2 and DE3; then... The Middle The generation Tai Di Parallel Machine The oth target of the glass workpiece The pheromones are updated to obtain the first... The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece ;

[0066] Step 3.6: When DE3 is not (0,0,...,0), execute the scheduling of the m-th batch processor, thereby performing tempering processing on the b-th batch on the m-th batch processor, and setting the index position of the tempered glass workpiece in DE3 to 0, thus updating DE3; then... The Middle The generation Tai Di Parallel Machine The oth target of the glass workpiece The pheromones are updated to obtain the first... The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece ;

[0067] Step 3.7, if and and If yes, proceed to step 3.4; otherwise, it means that a workshop scheduling scheme corresponding to the a-th ant in the k-th generation has been obtained.

[0068] Step 3.8, When At that time, Assign to Proceed to step 3.3 and execute sequentially; otherwise, it indicates that the k-th generation has been obtained. A workshop scheduling scheme for individual ants, and compare it with a set of historical workshop scheduling schemes. Merge the populations to obtain the k-th generation merged population, and select the one with the highest fitness. The workshop scheduling scheme serves as a new set of historical workshop scheduling schemes. When k=1, the historical workshop scheduling scheme set That is, the kth generation A workshop scheduling scheme for ants;

[0069] Step 3.9, If If the latest historical workshop scheduling plan is found, then production will proceed; otherwise, proceed to step 3.10.

[0070] Step 3.10: Use equation (61) to obtain the first... The generation Tai Di Parallel Machine The pheromone of the oth target on the glass workpiece :

[0071] (61)

[0072] In equation (61), This represents the volatility coefficient of the global pheromone. express Workshop scheduling scheme for any ant Indicates the first The generation Class 1 On the first parallel machine The change in pheromone level of the o-th target on a glass workpiece; and we have:

[0073] (62)

[0074] In equation (62), Indicates the first The generation Class 1 On the first parallel machine The 0th target value corresponding to each glass workpiece;

[0075] Step 3.11: Use equation (63) to obtain the first... +1st generation The glass workpiece and the first When a glass workpiece is simultaneously assigned to the pheromone of the 0th target in the bth batch:

[0076] (63)

[0077] In formula (63), represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time,

[0078] (64)

[0079] In formula (64), represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time, represents the otheglass workpiece of the bthbatch of the a thgeneration and the b thglass workpiece of the b thbatch of the a thgeneration are simultaneously assigned to the othtarget value of the b thbatch of time,

[0080] After step 3.12, the value of is assigned to, and then go to step 3.2 for sequential execution. Further, the step 3.4 includes:

[0081] Step 3.4.1, the a thant selects a unscheduled glass workpiece from DE1 randomly, and then according to the qualification constraints of the glass workpiece, selects the b thglass workpiece of the a thgeneration from the b thbatch of time by using formula (46),

[0082] Step 3.4.1, the a thant selects a unscheduled glass workpiece from DE1 randomly, and then according to the qualification constraints of the glass workpiece, selects the b thglass workpiece of the a thgeneration from the b thbatch of time by using formula (46), Step 3.4.1, the a thant selects a unscheduled glass workpiece from DE1 randomly, and then according to the qualification constraints of the glass workpiece, selects the b thglass workpiece of the a thgeneration from the b thbatch of time by using formula (46), Step 3.4.1, the a thant selects a unscheduled glass workpiece from DE1 randomly, and then according to the qualification constraints of the glass workpiece, selects the b thglass workpiece of the a thgeneration from the b thbatch of time by using formula (46), Step 3.4.1, the a thant selects a unscheduled glass workpiece from DE1 randomly, and then according to the qualification constraints of the glass workpiece, selects the b thglass workpiece of the a thgeneration from the b thbatch of time by using formula (46),

[0083] (46)

[0084] (47)

[0085] (48)

[0086] In formula (46), represents the preference weight of the othtarget function; represents the preference weight of the othtarget function; represents the preference weight of the othtarget function; represents the preference weight of the othtarget function; ​​​​the othet objective value of the jth glass workpiece in the parallel machine; formula (47) and (48), the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the

[0087] the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the

[0088] (49)

[0089] (50)

[0090] Step 3.4.3, the ath ant utilizes formula (51) to calculate the selection probability of the jth glass workpiece in the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the

[0091] (51)

[0092] in formula (51), and are two weight values respectively; the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the the jth glass workpiece reaches the The glass workpiece arrived at the first Tai Di Second candidate list of parallel machines Heuristic information, Indicates the first The generation Tai Di Parallel Machine The pheromone of the oth target in a glass workpiece;

[0093] Step 3.4.4: The a-th ant uses formula (52) to... Update to obtain the first The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece :

[0094] (52)

[0095] In equation (52), Indicates the volatility coefficient of local pheromones. This is the initial pheromone.

[0096] 6. The intelligent scheduling optimization method for hybrid assembly line workshops based on historical information ant colony algorithm according to claim 5, characterized in that step 3.5 includes:

[0097] Step 3.5.1: The a-th ant randomly selects an unscheduled glass workpiece from DE2, and then selects the ant according to the qualification constraints of the glass workpiece using equation (46). Tai Di Parallel machines, using equations (47) and (48) respectively, yield the first... Tai Di First candidate list of parallel machines With the second candidate list ;

[0098] Step 3.5.2: Using equation (49), we obtain the result that the j-th glass workpiece has reached the first position. Tai Di First candidate list of parallel machines Heuristic information Using equation (50), we can obtain the result that the j-th glass workpiece reaches the first position. Tai Di Second candidate list of parallel machines Heuristic information ;

[0099] Step 3.5.3: The a-th ant uses formula (51) to calculate the th ant. Tai Di The selection probability of the j-th glass workpiece in a parallel machine This process obtains the selection probabilities of all glass workpieces, selects the glass workpiece with the highest selection probability, processes it, stores the processed glass workpiece in buffer 2, sets the index position of the processed glass workpiece in DE2 to 0, and modifies the index position of the processed glass workpiece in DE3 to the index number of the processed glass workpiece itself; thus updating DE2 and DE3.

[0100] Step 3.5.4, the a-th ant uses formula (52) to... Update to obtain the first The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece .

[0101] Furthermore, step 3.6 includes:

[0102] Step 3.6.1: The a-th ant uses equation (53) to select the m-th batch processing machine, which is used to add the glass workpiece that meets the eligibility constraints and arrives at the m-th batch processing machine earliest to the b-th batch. Equations (54) and (55) are used to construct the first batch that arrives at the m-th batch processing machine earliest. A glass workpiece is assigned to the first candidate list of the bth batch. With the The glass workpieces are assigned to the second candidate list of the bth batch. :

[0103] (53)

[0104] (54)

[0105] (55)

[0106] In equations (56) and (57), This indicates the m-th batch processor. The arrival time of each glass workpiece This indicates the start time of the b-th batch; Indicates the first The dimensions of a glass workpiece This represents the capacity of the m-th batch processing machine. This indicates the size of the b-th batch;

[0107] Step 3.6.2, the heuristic information of the first candidate list of the bth batch to which the jth glass workpiece is assigned is obtained by using formula (56) the heuristic information of the second candidate list of the bth batch to which the jth glass workpiece is assigned is obtained by using formula (57)

[0108] (56)

[0109] (57)

[0110] Step 3.6.3, the a th ant uses formula (58) to calculate the selection probability of the jth glass workpiece in the list of the bth batch , so as to obtain the selection probability of all glass workpieces and select the glass workpiece corresponding to the maximum selection probability for tempering treatment, and set the value of the index position of the glass workpiece after tempering treatment in DE3 to 0, thereby updating DE3;

[0111] (58)

[0112] In formula (58), and are respectively two weight values; represents the pheromone of the oth target when the jth glass workpiece is assigned to the bth batch in the oth generation, represents the heuristic information of the first candidate list of the bth batch to which the jth glass workpiece is assigned, represents the heuristic information of the second candidate list of the bth batch to which the jth glass workpiece is assigned, represents the pheromone of the oth target when the jth glass workpiece is assigned to the bth batch in the oth generation; Step 3.6.4, the pheromone of the oth target when the jth glass workpiece is assigned to the bth batch in the oth generation is calculated by using formula (59)

[0113] Step 3.6.4, the pheromone of the oth target when the jth glass workpiece is assigned to the bth batch in the oth generation is calculated by using formula (59)

[0114] (59)

[0115] In formula (61), ,​​​​​​​​​​​​​​​​ represents the oth target pheromone when the (o-1)th glass workpiece and the oth glass workpiece are simultaneously assigned to the oth target of the bth batch represents the number of glass workpieces contained in the bth batch

[0116] Step 3.6.5, obtaining the oth target pheromone when the (o-1)th glass workpiece and the oth glass workpiece are simultaneously assigned to the oth target of the bth batch

[0117] (60)。

[0118] The electronic device comprises a memory and a processor, and the memory is used for storing a program supporting the processor to execute the mixed flow shop scheduling method, and the processor is configured to execute the program stored in the memory.

[0119] The computer readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the mixed flow shop scheduling method are executed.

[0120] Compared with the prior art, the present application has the following advantages:

[0121] 1. According to the actual production situation of the mixed flow shop, multiple constraint conditions are set, a double-objective optimization model for the required completion time and energy consumption of the mixed flow shop production is established, and a historical information ant colony algorithm is used to solve the problem model to obtain an optimal double-objective solution. The intelligent method is used to replace or assist the scheduling personnel to make decisions, thereby greatly improving the production efficiency of the workshop.

[0122] 2. The historical information ant colony algorithm used in the present application has good global search ability and fast convergence speed, and is very suitable for solving large-scale mixed flow shop scheduling problems. The mixed flow shop scheduling problem of glass workpiece production has multiple complex processing steps. The standard ant colony algorithm is easily trapped in local optimum and cannot search for the optimal production scheme due to its high blindness and high randomness. Therefore, the historical workshop scheduling scheme set is used to design and improve the ant colony algorithm, so that the effect of solving the tempered glass production scheduling problem is better. BRIEF DESCRIPTION OF DRAWINGS

[0123] Figure 1 is the scheduling method flowchart of the present application.​

[0124] Figure 2 For the glass production process flowchart;

[0125] Figure 3 For the history information ant colony algorithm flowchart. DETAILED DESCRIPTION

[0126] In this embodiment, a mixed flow shop intelligent scheduling optimization method based on a history information ant colony algorithm, with reference to Figure 2 is to obtain the process information of tempered glass production. When the customer places an order, the manufacturer needs to prepare the required glass sheet within a certain time, and after cutting, spraying and tempering three stages of operation, the final tempered glass is formed. In the cutting stage, multiple parallel machines are used for processing at the same time, and each parallel machine can only process a specific type of glass sheet. Therefore, the corresponding parallel machine needs to be selected according to the customer's needs to complete the glass (glass workpiece) cutting operation. Subsequently, in the spraying stage, a suitable one will be selected from multiple parallel machines with different spraying functions to coat a special coating on the surface of the cut glass, such as an anti-reflective coating, a scratch-resistant coating. Finally, in the tempering stage, multiple furnaces with different capacities are used to temper the coated glass. As long as the total size of the glass processed at one time does not exceed the capacity of the furnace, multiple glass tempering operations can be performed simultaneously, and this process can be abstracted as a parallel batch processor scheduling problem with different machine capacities. It is worth noting that between the two stages, due to the limited storage space (i.e. limited buffer capacity), the glass of the current stage can only be processed in the next stage or placed in the storage space if and only if the device of the next stage is idle or the storage space is idle; otherwise, the current device will not be able to process the next (batch) glass. Specifically, as shown in the figure, the mixed flow shop intelligent scheduling optimization method is carried out as follows: Figure 1

[0127] Step 1, two objective functions of the mixed flow shop scheduling model are established by using formula (1) and formula (2), including: the first objective function of the maximum completion time and the second objective function of the minimum energy consumption , and let any othertarget function be , :

[0128] (1)

[0129] (2)

[0130] In formula (1), is the maximum completion time, represents the bth batch in the ​The completion time on the batch processor;

[0131] TEC is the energy consumption in formula (2), is the total number of time periods, is the total number of parallel machines, is the total number of batch processors, f(t) is the electricity in the tth time period, is the electricity in the tth time period, is the electricity in the tth time period, is the electricity in the tth time period, if the tth parallel machine is in the working state, then let =1; otherwise, let =0; is the electricity in the tth time period, is the electricity in the tth time period, is the electricity in the tth time period, if the tth parallel machine is in the standby state, then let =1; otherwise, let =0; is the electricity in the tth time period, is the electricity in the tth time period, is the electricity in the tth time period, if the tth parallel machine is in the shutdown state, then let =1; otherwise, let =0; is the electricity in the tth time period, is the electricity in the tth time period, is the electricity in the tth time period, if the tth batch processor is in the working state, then let =1; otherwise, let =0; is the electricity in the tth time period, is the electricity in the tth time period, is the electricity in the tth time period, if the tth batch processor is in the standby state, then let =1; otherwise, let =0; is the electricity in the tth time period, is the electricity in the tth time period, , and respectively represent the energy consumption of the tth parallel machine in the working state, standby state and shutdown state; , and respectively represent the energy consumption of the batch processor in the working state, standby state and shutdown state.

[0132] Step 2, constraints of the mixed flow shop scheduling model are established by using formula (3)-(45):

[0133] (3) ​

[0134] (4)

[0135] (5)

[0136] (6)

[0137] (7)

[0138] (8)

[0139] (9)

[0140] (10)

[0141] (11)

[0142] (12)

[0143] (13) (14)

[0144] (15)

[0145] (16)

[0146] (17)

[0147] (18)

[0148] (19)

[0149] (20)

[0150] (21)

[0151] (22)

[0152] (23)

[0153] (24)

[0154] (25)

[0155] (26)

[0156] (27)

[0157] (28)

[0158] (29)

[0159] (30)

[0160] (31)

[0161] (32)

[0162] (33)

[0163] (34)

[0164] (35)

[0165] (36)

[0166] (37)

[0167] (38)

[0168] (39)

[0169] (40)

[0170] (41)

[0171] (42)

[0172] (43)

[0173] (44)

[0174] (45)

[0175] Constraint (3) indicates that each glass workpiece is processed on only one parallel machine at each stage;

[0176] Constraint (4) indicates that a glass workpiece can only be processed on a parallel machine that can process that glass workpiece;

[0177] Constraint (5) indicates that the work status of the jth parallel machine of the ith type in the kth time period is determined by the work status of the jth parallel machine of the ith type in the (k-1)th time period and the work status of the jth parallel machine of the ith type in the kth time period; ​​​

[0178] Constraint (6) indicates the calculation of glass workpieces. The processing time for each type of parallel machine;

[0179] Constraint (7) represents the glass workpiece Assigned to the Processing can only be done after parallel machines are used;

[0180] Constraint (8) indicates that if the first... The glass workpiece in the first Tai Di If processing is performed on a parallel machine, then the first... A parallel machine has at least one time period. Processing the first A glass workpiece;

[0181] Constraint (9) indicates the first The preparation time for each glass workpiece must be before the start time of the first type of parallel machine.

[0182] Constraint (10) indicates that in the first... Parallel machine, if the first The glass workpiece in the first In the time period, at the ... Processing on a parallel machine, then Greater than or equal to the The start time of processing a glass workpiece.

[0183] Constraint (11) indicates that when the first The glass workpiece in the first In the time period, at the ... Class 1 If processed on a parallel machine, then the first... The completion time for processing a glass workpiece is greater than or equal to .

[0184] Constraint (12) indicates a restriction on the first A parallel machine can only be in one state during a time period;

[0185] Constraints (13) and (14) represent the first... Tai Di Parallel machines, in order to achieve this, in the first... If a time period is in an idle state, then at least one time period has been worked before. Furthermore, it will need to be re-evaluated in subsequent time periods. Glass workpieces processed by a parallel machine.

[0186] Constraint (15) indicates a restriction on the first Tai Di Parallel machines in the first The time period and the first The device cannot be in standby mode during any given time period;

[0187] Constraints (16) and (17) indicate that once the parallel machine is shut down, it will not be restarted, i.e., if the first... The parallel machine in the first If it closes within a certain time period, then from arrive No more glass workpieces will be processed during the time period;

[0188] Constraint (18) indicates the first Only after a glass workpiece has completed processing by the first type of parallel machine can processing by the second type of parallel machine begin.

[0189] Constraint (19) indicates the computation of the first... The waiting time of a glass workpiece in the first buffer zone is called the [number]th [time]. The difference between the start time of the second type of parallel machine and the completion time of the first type of parallel machine for a glass workpiece;

[0190] Constraints (20) and (21) indicate that the total size of the glass workpieces held in the buffer must be less than or equal to the capacity of the buffer;

[0191] Constraint (22) indicates the computation of the first... The waiting time of the first glass workpiece in the buffer zone, i.e., the first... The start time of processing a glass workpiece on a batch processing machine minus the completion time on a type 2 parallel processing machine.

[0192] Constraint (23) indicates that each glass workpiece can only be processed in one batch on one batching machine;

[0193] Constraint (24) indicates that the size of the batch does not exceed the capacity of the batch processing machine that processes the batch;

[0194] Constraint (25) indicates that the processing time of a batch is determined by the maximum processing time of the glass workpieces in the batch;

[0195] Constraint (26) indicates the judgment of the first Taiwan batch processing machine in the first Is the device in operation during each time period?

[0196] Constraint (27) indicates the first The batch in Processing time on a batch processing machine;

[0197] Constraint (28) indicates that when the first The first batch processing machine When no batch has been formed, the processing time is 0.

[0198] Constraint (29) indicates batch number On a batch processing machine, only if the first Only after the first batch is generated can the second batch be formed. +1 batch;

[0199] Constraint (30) indicates that when the first The first batch processing machine When a batch contains at least one glass workpiece It equals 1;

[0200] Constraint (31) represents coordination and The value;

[0201] Constraint (32) indicates the first Only after a glass workpiece has been processed by the second type of parallel machine can the batch processing machine begin processing.

[0202] Constraint (33) indicates the first The first batch processing machine Each batch and the first +1 batch sequence;

[0203] Constraints (34) and (35) indicate that batch processing takes at least one time period and cannot be preempted;

[0204] Constraint (36) represents the calculation of the maximum completion time, which is the maximum completion time of all non-empty batches on all batch processors;

[0205] Constraint (37) indicates that all batches must be in the first batch. Processing must be completed before the specified time period;

[0206] Constraint (38) indicates the first Taiwan batch processing machine in the first Each time period can only be in one state;

[0207] Constraints (39) and (40) represent the first... Only when the batch processing machine has worked for at least one time cycle and there are still glass workpieces waiting to be processed can it proceed to the next... The system is in standby mode for one time period.

[0208] Constraint (41) indicates the first Taiwan batch processing machine in the first The time period and the first The device cannot be in standby mode during any given time period;

[0209] Constraints (42) and (43) represent setting the first The batch processing machine is currently in a stopped state, and once stopped, it will not be restarted.

[0210] Constraint (44) indicates the computation of the first... The first batch processing machine The completion time of each batch.

[0211] Constraint (45) represents the amount of electricity at different time periods.

[0212] In equations (3) to (45), Indicates the first The glass workpiece in the first Processing time on a parallel machine Indicates the first The arrival time of each glass workpiece Indicates the first The glass workpiece in the first Completion time on parallel machines Indicates the first The dimensions of a glass workpiece Indicates the first The capacity of the batch processing machine Indicates the first The batch in Processing time on a batch processing machine. Indicates the first The batch in Start processing time on the batch processing machine. Indicates the first The glass workpiece in the first Processing time on a batch processing machine; Indicates the first Is the glass workpiece in the [number]th ... Tai Di Processing on a parallel machine; if so, then let =1; otherwise, let =0; Indicates the first Is the glass workpiece in the [number]th ... Tai Di Processing on a parallel machine; if so, then let Otherwise, let ; This indicates whether the j-th glass workpiece is waiting in the first buffer during the t-th time period. If so, then let... Otherwise, let ; Indicates the first The qualification constraints for the first glass workpiece, i.e., whether it can be used in the first... Tai Di Process on a parallel machine; if possible, then let... Otherwise, let ; Indicates the first Is the glass workpiece in the [number]th ... The first batch processing machine If so, then let Otherwise, let ; This indicates the t-th time period. Is the batch in the [number]th [period]? Processed on a batch processing machine in Taiwan; if so, then... Otherwise, let ; Indicates the first Does each batch contain at least one glass workpiece? If so, then let... Otherwise, let ; For the i-th stage in the t-th time period Auxiliary binary variables for parallel machines This indicates the t-th time period. Auxiliary binary variables for batch processing; and These are the capacities of the first buffer and the second buffer, respectively. Indicates high battery level; Indicates low battery level; Indicates the end time period of high battery level; Indicates the end time period of low battery; Indicates the first Taiwan batch processing machine in the first If the system is in operation for a given time period, then... =1; otherwise, let =0; Indicates the first Tai Di Parallel machines in the first If the system is in operation for a given time period, then... =1; otherwise, let =0.

[0213] Step 3: Solve the hybrid assembly line workshop scheduling model using the historical information ant colony algorithm to obtain the production and processing plan; specifically, such as... Figure 3 As shown.

[0214] Step 3.1: Define the maximum number of iterations as... The total number of ants is Randomly generate the first preference vector With the second preference vector Initialize the heuristic information matrix; Define the current generation as k; Initialize k=1; Initialize the pheromone matrix of the kth generation. ;

[0215] Step 3.2: Set the ant's index a=1;

[0216] Step 3.3: Initialize the parameters of the glass workpiece and the machine, set the first taboo table DE1 in n dimensions, and assign an initial value to DE1 according to the sequence number of the glass workpiece. Let there be an n-dimensional second tabu list DE2 and an n-dimensional third tabu list DE3, and assign initial values ​​to both. n represents the total number of glass workpieces;

[0217] Step 3.4: When l=1 and DE1 is not (0,0,...,0), execute the scheduling policy of the l-th type of parallel machine:

[0218] Step 3.4.1: The a-th ant randomly selects an unscheduled glass workpiece from DE1, and then selects the ant according to the qualification constraints of the glass workpiece using equation (46). Tai Di Parallel machines, using equations (47) and (48) respectively, yield the first... Tai Di First candidate list of parallel machines With the second candidate list :

[0219] (46)

[0220] (47)

[0221] (48)

[0222] In equation (46), Represents the o-th objective function Preference weights; Indicates the first Tai Di The o-th objective value of the parallel machine; in equations (47) and (48), This indicates that the j-th glass workpiece has reached the first... The time of a parallel machine Indicates the first Tai Di Completion time of parallel-like machines.

[0223] Step 3.4.2: Using equation (49), we obtain the result that the j-th glass workpiece has reached the first position. Tai Di First candidate list of parallel machines Heuristic information Using equation (50), we can obtain the result that the j-th glass workpiece reaches the first position. Tai Di Second candidate list of parallel machines Heuristic information ;

[0224] (49)

[0225] (50)

[0226] This invention is constructed by determining whether the start time of the parallel machine for the glass workpiece to be processed is delayed. and Its function is to prevent parallel machines from experiencing idle waiting time delays. Glass workpieces are preferred in this way to reduce time delays.

[0227] Step 3.4.3: The a-th ant uses formula (51) to calculate the th ant. Tai Di The selection probability of the j-th glass workpiece in a parallel machine This process obtains the selection probabilities of all glass workpieces, selects the glass workpiece with the highest selection probability, processes it, stores the processed glass workpiece in buffer 1, sets the index position of the processed glass workpiece in DE1 to 0, and modifies the index position of the processed glass workpiece in DE2 to the index number of the processed glass workpiece itself; thus updating DE1 and DE2.

[0228] (51)

[0229] In equation (51), and Each has two weights; Indicates the first The generation Tai Di Parallel Machine The pheromone of the oth target on the glass workpiece Indicates the first The glass workpiece arrived at the first Tai Di First candidate list of parallel machines Heuristic information, Indicates the first The glass workpiece arrived at the first Tai Di Second candidate list of parallel machines Heuristic information, Indicates the first The generation Tai Di Parallel Machine The pheromone of the oth target on a glass workpiece.

[0230] Step 3.4.4: The a-th ant uses formula (52) to... Update to obtain the first The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece :

[0231] (52)

[0232] In equation (52), Indicates the volatility coefficient of local pheromones. This is the initial pheromone.

[0233] Step 3.5: When l=2 and DE2 is not (0,0,...,0), execute the scheduling of the second stage parallel machine:

[0234] Step 3.5.1: The a-th ant randomly selects an unscheduled glass workpiece from DE2, and then selects the ant according to the qualification constraints of the glass workpiece using equation (46). Tai Di Parallel machines, using equations (47) and (48) respectively, yield the first... Tai Di First candidate list of parallel machines With the second candidate list ;

[0235] Step 3.5.2: Using equation (49), we obtain the result that the j-th glass workpiece has reached the first position. Tai Di First candidate list of parallel machines Heuristic information Using equation (50), we can obtain the result that the j-th glass workpiece reaches the first position. Tai Di Second candidate list of parallel machines Heuristic information .

[0236] Step 3.5.3: The a-th ant uses formula (51) to calculate the th ant. Tai Di The selection probability of the j-th glass workpiece in a parallel machine This process obtains the selection probabilities of all glass workpieces, selects the glass workpiece with the highest selection probability, processes it, stores the processed glass workpiece in buffer 2, sets the index position of the processed glass workpiece in DE2 to 0, and modifies the index position of the processed glass workpiece in DE3 to the index number of the processed glass workpiece itself; thus updating DE2 and DE3.

[0237] Step 3.5.4, the a-th ant uses formula (52) to... Update to obtain the first The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece .

[0238] Step 3.6: When DE3 is not (0,0,...,0), execute the batch processing machine's scheduling:

[0239] Step 3.6.1: The a-th ant uses equation (53) to select the m-th batch processing machine, which is used to add the glass workpiece that meets the eligibility constraints and arrives at the m-th batch processing machine earliest to the b-th batch. Equations (54) and (55) are used to construct the first batch that arrives at the m-th batch processing machine earliest. A glass workpiece is assigned to the first candidate list of the bth batch. With the The glass workpieces are assigned to the second candidate list of the bth batch. :

[0240] (53)

[0241] (54)

[0242] (55)

[0243] In equations (56) and (57), This indicates the m-th batch processor. The arrival time of each glass workpiece This indicates the start time of the b-th batch; Indicates the first The dimensions of a glass workpiece This represents the capacity of the m-th batch processing machine. This indicates the size of the b-th batch.

[0244] Step 3.6.2: Using equation (56), we obtain the first... A glass workpiece is assigned to the first candidate list of the bth batch. Heuristic information Using equation (57), we obtain the first... The glass workpieces are assigned to the second candidate list of the bth batch. Heuristic information :

[0245] (56)

[0246] (57)

[0247] This invention utilizes the consideration of whether the glass workpieces to be processed will delay the start time of the current batch, while also taking into account the capacity constraints of the batch processing machine, to construct... and Its function is to prioritize glass workpieces that will not delay the start time of the current batch, thereby reducing the idle time of the batch processor.

[0248] Step 3.6.3: The a-th ant uses formula (58) to calculate the list of the b-th batch. The probability of selecting the j-th glass workpiece This allows us to obtain the selection probabilities of all glass workpieces, select the glass workpiece with the highest selection probability for processing, and set the index position of the processed glass workpiece in DE3 to 0, thereby updating DE3.

[0249] (58)

[0250] In equation (58), and Each has two weights; Indicates the first The generation When a glass workpiece is assigned to the b-th batch, the pheromone of the o-th target is... Indicates the first A glass workpiece is assigned to the first candidate list of the bth batch. Heuristic information, Indicates the first The glass workpieces are assigned to the second candidate list of the bth batch. Heuristic information, Indicates the first The generation When a glass workpiece is assigned to the b-th batch, the pheromone of the o-th target is...

[0251] Step 3.6.4: Calculate the first step using equation (59). The generation When a glass workpiece is assigned to the b-th batch, the pheromone of the o-th target... :

[0252] (59)

[0253] In equation (61), , Indicates the first The first in the bth batch The glass workpiece and the first The pheromone of the o-th target when a glass workpiece is simultaneously assigned to the b-th batch; This indicates the number of glass components contained in the b-th batch.

[0254] Step 3.6.5: Use equation (60) to obtain the first... The first in the bth batch The glass workpiece and the first When a glass workpiece is simultaneously assigned to the o-th target in the b-th batch, the updated pheromone... :

[0255] (60)

[0256] Step 3.7, if and and If yes, proceed to step 3.4; otherwise, it means that a workshop scheduling scheme corresponding to the a-th ant of the k-th generation has been obtained.

[0257] Step 3.8, When At that time, Assign to Proceed to step 3.3 and execute sequentially; otherwise, it indicates that the k-th generation has been obtained. A workshop scheduling scheme for individual ants, and compare it with a set of historical workshop scheduling schemes. Merge the populations to obtain the k-th generation merged population, and select the one with the highest fitness. The workshop scheduling scheme serves as a new set of historical workshop scheduling schemes. When k=1, the historical workshop scheduling scheme set That is, the kth generation Workshop scheduling scheme for ants.

[0258] This invention employs an iterative search using a historical workshop scheduling scheme set. Its function is that after all ants in the k-th generation have constructed their scheduling schemes, it performs a hierarchical sorting of the non-dominated scheduling schemes against the scheduling schemes in the historical workshop scheduling scheme set, thereby identifying Pareto scheduling schemes of varying quality at each level. Starting from the first level of Pareto scheduling schemes, it selects schemes based on the congestion distance of each level, continuing this process layer by layer until all schemes are completed. a scheduling scheme, reconstructs the historical set of scheduling schemes by using these scheduling schemes, so as to ensure that the optimal quality of the first scheduling scheme is selected.

[0259] Step 3.9, if , the latest historical scheduling scheme is output for production, otherwise, step 3.10 is executed.

[0260] Step 3.10, the information of the oth target of the glass workpiece on the bth parallel machine in the nth generation of the mth class is obtained by using formula (61)

[0261] (61)

[0262] In formula (61), represents the evaporation coefficient of global pheromone, represents the scheduling scheme of any ant in the colony, represents the information of the oth target of the glass workpiece on the bth parallel machine in the nth generation of the mth class,

[0263] (62)

[0264] In formula (62), represents the oth target value corresponding to the glass workpiece on the bth parallel machine in the nth generation of the mth class,

[0265] Step 3.11, the information of the oth target of the glass workpiece on the bth parallel machine in the nth generation of the mth class is obtained by using formula (63)

[0266] (63)

[0267] In formula (63), represents the oth target value corresponding to the glass workpiece on the bth parallel machine in the nth generation of the mth class, ​​​​​​​​​​​​​​​​​​​​​The generation The glass workpiece and the first The change in pheromone of the o-th target when several glass workpieces are simultaneously assigned to the b-th batch, and the following:

[0268] (64)

[0269] In equation (64), Indicates the first The generation The glass workpiece and the first When a glass workpiece is simultaneously assigned to the o-th target value in the b-th batch.

[0270] Step 3.12, Assign to Then, proceed to step 3.2 for sequential execution.

[0271] In this embodiment, an electronic device includes a memory and a processor. The memory stores a program that supports the processor in executing the above-described method, and the processor is configured to execute the program stored in the memory.

[0272] In this embodiment, a computer-readable storage medium stores a computer program, which is executed by a processor to perform the steps of the above method.

Claims

1. A hybrid flow shop intelligent scheduling optimization method based on historical information ant colony algorithm, characterized in that, is applied to the processing of each glass workpiece after cutting and parallel machine, into the first buffer zone waiting, from the first buffer zone to each glass workpiece is transported to the spray parallel machine for processing, into the second buffer zone waiting, from the second buffer zone to each glass workpiece is transported to the batch processing machine when the group of each batch, then each batch in the batch processing machine for steel processing, until all the glass workpiece processing pipeline, cutting and parallel machine and spray parallel machine in any parallel machine is marked as the first parallel machine, the mixed flow line intelligent scheduling optimization method is as follows: Step 1, establish the first objective function with the maximum completion time as the target And the second objective function with the minimum energy consumption as the target As the two objective functions of the mixed flow shop scheduling model, and let any oth objective function be , ; Step 2, establishing constraint conditions of the mixed flow shop scheduling model; Step 3, using historical information ant colony algorithm to solve the mixed flow shop scheduling model to obtain a production processing scheme; Step 3.1, define the maximum number of iterations as and the total number of ants as , randomly generate a first preference vector and a second preference vector ; initialize the heuristic information matrix; define the current generation number as k; initialize k = 1; initialize the pheromone matrix of the kth generation ; Step 3.2, setting an index a of the ant to be 1; Step 3.3, initialize parameters of the glass workpiece and the machine, set the first n-dimensional tabu table DE1, and assign the initial value of the glass workpiece sequence number to DE1 as , set the second n-dimensional tabu table DE2 and the third n-dimensional tabu table DE3, and assign the initial value as ; n represents the total number of glass workpieces; Step 3.4: When l=1 and DE1 is not (0,0,...,0), execute the... The scheduling of the first type of parallel machine, thus achieving the first... Tai Di On a parallel machine, a glass workpiece is processed, and the processed glass workpiece is stored in the first buffer. The value of the index position of the processed glass workpiece in DE1 is set to 0, and the value of the index position of the processed glass workpiece in DE2 is modified to the index number of the processed glass workpiece itself; thus updating DE1 and DE2; then... The Middle The generation Tai Di Parallel Machine The oth target of the glass workpiece The pheromones are updated to obtain the first... The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece ; Step 3.5: When l=2 and DE2 is not (0,0,...,0), execute the... Tai Di Scheduling of parallel machines, thus in the first... Tai Di On a parallel machine, a glass workpiece is processed, and the processed workpiece is stored in a second buffer. The index of the processed workpiece in DE2 is set to 0, and the index of the processed workpiece in DE3 is modified to its own index number; thus updating DE2 and DE3; then... The Middle The generation Tai Di Parallel Machine The oth target of the glass workpiece The pheromones are updated to obtain the first... The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece ; Step 3.6, when DE3 is not (0, 0,..., 0), perform the scheduling of the mth batch processor, so as to perform the tempering process on the bth batch on the mth batch processor, and set the value of the index position of the glass workpiece in DE3 to 0 after the tempering process is completed, so as to update DE3; again perform the scheduling of the mth batch processor on the (b+1)th batch th generation th generation th generation th generation th glass workpiece on the oth target of the mth batch processor of the nth parallel machine th glass workpiece on the oth target of the mth batch processor of the nth parallel machine th generation th generation th generation th glass workpiece on the oth target of the mth batch processor of the nth parallel machine th glass workpiece on the oth target of the mth batch processor of the nth parallel machine Step 3.7, if and and go to Step 3.4; otherwise, a job-shop scheduling scheme corresponding to the a-th ant of the k-th generation is obtained. Step 3.8, When At that time, Assign to Proceed to step 3.3 and execute sequentially; otherwise, it indicates that the k-th generation has been obtained. A workshop scheduling scheme for individual ants, and compare it with a set of historical workshop scheduling schemes. Merge the populations to obtain the k-th generation merged population, and select the one with the highest fitness. The workshop scheduling scheme serves as a new set of historical workshop scheduling schemes. ; When k = 1, the historical set of job shop scheduling schemes i.e. the kth generation of job shop scheduling schemes of the ith ant; Step 3.

9. If then output the latest historical shop floor schedule for production, otherwise, perform Step 3.10; Step 3.10, obtaining the first generation from the first generation from the first generation from the first generation from the first : (61) In formula (61), denotes the volatility coefficient of the global pheromone, denotes the job-shop scheduling scheme of any of the ants, denotes the generation the class the station the target of the oth glass workpiece; and has: (62) In formula (62), representing the representing the representing the representing the representing the Step 3.11, obtaining the first +1 generation of the first glass workpiece and the second glass workpiece are simultaneously assigned to the pheromone of the oth target of the bth batch glass workpiece (63) In formula (63), represents the th glass workpiece is assigned to the th target value of the bth batch at the same time, represents the th glass workpiece is assigned to the th target value of the bth batch at the same time, represents the th pheromone change amount of the oth target of the bth batch at the same time, and (64) In formula (64), represents the represents the represents the represents the othetarget value when the bthbatch is assigned to the first glass workpiece and the second glass workpiece simultaneously. Step 3.12, set to and go to Step 3.

2.

2. The hybrid flow shop intelligent scheduling optimization method based on historical information ant colony algorithm according to claim 1, characterized in that, In the step 1, two objective functions of the mixed flow shop scheduling model are established by using formula (1) and formula (2): (1) (2) In equation (1), To maximize the completion time, This indicates that the b-th batch is in the th batch. Completion time on the batch processing machine; In equation (2), TEC represents energy consumption. The total number of time periods. This represents the total number of parallel machines. Let f(t) represent the total number of batch processors, and f(t) represent the amount of electricity consumed in the t-th time period. Indicates the first Tai Di Determine whether the parallel machine is in a working state during the t-th time period. If so, then let... =1; otherwise, let =0; Indicates the first Tai Di Is the parallel machine in standby state during the t-th time period? If so, then let... =1; otherwise, let =0; Indicates the first Tai Di Is the parallel machine in the off state during the t-th time period? If so, then let... =1; otherwise, let =0; Indicates the first Is the batch processing machine in operation during the t-th time period? If so, then... =1; otherwise, let =0; Indicates the first Is the batch processing machine in standby mode during the t-th time period? If so, then... =1; otherwise, let =0; Indicates the first Is the batch processing machine in a stopped state during the t-th time period? If so, then... =1; otherwise, let =0; , and They represent the first The energy consumption of a parallel machine in working, standby, and off states; , and These represent the energy consumption of the batch processor when it is in working, standby, and off states, respectively.

3. The hybrid flow shop intelligent scheduling optimization method based on historical information ant colony algorithm according to claim 2, characterized in that, In the step 2, constraint conditions of the mixed flow shop scheduling model are established by using formula (3)-formula (45): (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41) (42) (43) (44) (45) in the formulas (3)-(45), denotes the processing time of the th glass workpiece on the th parallel machine, denotes the arrival time of the th glass workpiece, denotes the completion time of the th glass workpiece on the th parallel machine, denotes the size of the th glass workpiece, denotes the capacity of the th batcher, denotes the processing time of the th batch on the th batcher, denotes the start processing time of the th batch on the th batcher, denotes the processing time of the th glass workpiece on the th batcher; denotes whether the th glass workpiece is processed on the th th parallel machine, if yes, then let = 1; otherwise, let = 0; denotes whether the th glass workpiece is processed on the th th parallel machine, if yes, then let ; otherwise, let ; denotes whether the th glass workpiece is waiting in the first buffer at the th time period, if yes, then let ; otherwise, let ; denotes the eligibility constraint of the th glass workpiece, i.e. whether it can be processed on the th th parallel machine, if yes, then let ; otherwise, let ; denotes whether the th glass workpiece is in the th batch of the th batcher, if yes, then let ; otherwise, let Is the batch in the [number]th [period]? Processed on a batch processing machine in Taiwan; if so, then... Otherwise, let ; Indicates the first Does each batch contain at least one glass workpiece? If so, then let... Otherwise, let ; For the i-th stage in the t-th time period Auxiliary binary variables for parallel machines This indicates the t-th time period. Auxiliary binary variables for batch processing; and These are the capacities of the first buffer and the second buffer, respectively. Indicates high battery level; Indicates low battery level; Indicates the end time period of high battery level; Indicates the end time period of low battery; Indicates the first Taiwan batch processing machine in the first If the system is in operation for a given time period, then... =1; otherwise, let =0; Indicates the first Tai Di Parallel machines in the first If the system is in operation for a given time period, then... =1; otherwise, let =0.

4. The hybrid flow shop intelligent scheduling optimization method based on historical information ant colony algorithm according to claim 3, characterized in that, The step 3.4 includes: Step 3.4.1, the first ant randomly selects an unscheduled glass workpiece from DE1, and according to the qualification constraint of the glass workpiece, selects the first candidate list of the glass workpiece from DE2 using formula (46) the first candidate list of the glass workpiece from DE2 using formula (46) the first candidate list of the glass workpiece from DE2 using formula (46) the first candidate list of the glass workpiece from DE2 using formula (46) the first candidate list of the glass workpiece from DE2 using formula (46) the first candidate list of the glass workpiece from DE2 using formula (46) the first candidate list of the glass workpiece from DE2 using formula (46) (46) (47) (48) In equation (46), Represents the o-th objective function Preference weights; Indicates the first Tai Di The o-th objective value of the parallel machine; in equations (47) and (48), This indicates that the j-th glass workpiece has reached the first... The time of a parallel machine Indicates the first Tai Di Completion time of parallel-like machines; Step 3.4.2, obtaining the jth glass workpiece to reach the jth station of the first candidate list of the parallel machines using formula (49) station of the first candidate list of the parallel machines Step 3.4.3, obtaining the jth glass workpiece to reach the jth station of the second candidate list of the parallel machines using formula (50) station of the second candidate list of the parallel machines Step 3.4.3, obtaining the jth glass workpiece to reach the jth station of the second candidate list of the parallel machines using formula (50) station of the second candidate list of the parallel machines Step 3.4.3, obtaining the jth glass workpiece to reach the jth station of the second candidate list of the parallel machines using formula (50) station of the second candidate list of the parallel machines Step 3.4.3, obtaining the jth glass workpiece to reach the jth (49) (50) Step 3.4.

3. The athant uses formula (51) to calculate the table the selection probability of the jthglass workpiece in the parallel machine Thus, the selection probabilities of all glass workpieces are obtained, and the glass workpiece corresponding to the maximum selection probability is selected for processing. The processed glass workpiece is stored in the first buffer area, and the value of the index position of the processed glass workpiece in DE1 is set to 0. The value of the index position of the processed glass workpiece in DE2 is modified to the index number of the processed glass workpiece itself. Thus, DE1 and DE2 are updated. (51) In equation (51), and Each has two weights; Indicates the first The generation Tai Di Parallel Machine The pheromone of the oth target on the glass workpiece Indicates the first The glass workpiece arrived at the first Tai Di First candidate list of parallel machines Heuristic information, Indicates the first The glass workpiece arrived at the first Tai Di Second candidate list of parallel machines Heuristic information, Indicates the first The generation Tai Di Parallel Machine The pheromone of the oth target in a glass workpiece; Step 3.4.

4. The 4th ant updates the information of the 0th target glass workpiece on the 4th generation of the ath parallel machine using formula (52) :​​​​​ (52) in formula (52), denotes the volatility coefficient of the local pheromone, is the initial pheromone.

5. The hybrid flow shop intelligent scheduling optimization method based on historical information ant colony algorithm according to claim 4, characterized in that, The step 3.5 includes: Step 3.5.1, the first ant randomly selects an unscheduled glass workpiece from DE2, and according to the qualification constraint of the glass workpiece, selects the first station and the second candidate list of the parallel machine of the same type according to the qualification constraint of the glass workpiece, and according to the qualification constraint of the glass workpiece, selects the first station and the second candidate list of the parallel machine of the same type according to the qualification constraint of the glass workpiece, and according to the qualification constraint of the glass workpiece, selects the first and the second candidate list of the parallel machine of the same type according to the qualification constraint of the glass workpiece, and according to the qualification constraint of the glass workpiece, selects the first and the second candidate list of the parallel machine of the same type according to the qualification constraint of the glass workpiece, and according to the qualification constraint of the glass workpiece, selects the first Step 3.5.2, obtaining the jth glass workpiece to reach the first station of the first candidate list of parallel machines using formula (49) Step 3.5.3, obtaining the jth glass workpiece to reach the second station of the second candidate list of parallel machines using formula (50) Step 3.5.4, obtaining the jth glass workpiece to reach the third station of the third candidate list of parallel machines using formula (51) Step 3.5.5, obtaining the jth glass workpiece to reach the fourth station of the fourth candidate list of parallel machines using formula (52) Step 3.5.6, obtaining the jth glass workpiece to reach the fifth Step 3.5.3, the a-th ant calculates the j-th glass workpiece in the parallel machine using formula (51) table the the selection probability of the j-th glass workpiece in the parallel machine Thus, the selection probabilities of all glass workpieces are obtained, and the glass workpiece corresponding to the maximum selection probability is selected for processing. The processed glass workpiece is stored in the second buffer area, and the value of the index position of the processed glass workpiece in DE2 is set to 0. The value of the index position of the processed glass workpiece in DE3 is modified to the index number of the processed glass workpiece itself. Thus, DE2 and DE3 are updated. Step 3.5.4, the a-th ant uses formula (52) to... Update to obtain the first The generation Tai Di Parallel Machine The updated pheromone of the o-th target on the glass workpiece .

6. The hybrid flow shop intelligent scheduling optimization method based on historical information ant colony algorithm according to claim 5, characterized in that, The step 3.6 includes: Step 3.6.1, the a-th ant utilizes formula (53) to select the m-th batcher for adding the glass workpiece that meets the qualification constraint and arrives at the m-th batcher earliest into the b-th batch, and utilizes formula (54) and formula (55) to respectively construct the first candidate list of the glass workpieces that arrive at the m-th batcher earliest and the second candidate list of the glass workpieces that arrive at the m-th batcher earliest into the b-th batch into the b-th batch into the b-th batch : (53) (54) (55) in formulas (56) and (57), denotes the arrival time of the mth glass workpiece of the bth batch, denotes the start processing time of the bth batch; denotes the start processing time of the mth batch machine, denotes the size of the mth glass workpiece of the bth batch, denotes the size of the mth glass workpiece of the bth batch, denotes the capacity of the mth batch machine, denotes the size of the bth batch; Step 3.6.2, the heuristic information of the first candidate list of the bth batch is obtained by using formula (56) Step 3.6.2, the heuristic information of the first candidate list of the bth batch is obtained by using formula (56) Step 3.6.2, the heuristic information of the first candidate list of the bth batch is obtained by using formula (56) Step 3.6.2, the heuristic information of the first candidate list of the bth batch is obtained by using formula (56) Step 3.6.2, the heuristic information of the first candidate list of the bth batch is obtained by using formula (56) Step 3.6.2, the heuristic information of the first candidate list of the bth batch is obtained by using formula (56) Step 3. (56) (57) Step 3.6.3, the a-th ant calculates the list of the b-th batch using formula (58) the selection probability of the j-th glass workpiece Thus, the selection probabilities of all glass workpieces are obtained, and the glass workpiece corresponding to the maximum selection probability is selected for tempering treatment, and the value of the index position of the glass workpiece after tempering treatment in DE3 is set to 0, so as to update DE3; (58) in formula (58), and are each 2 weights; denotes the th generation of the heuristic information for the th glass workpiece assigned to the oth target of the bth batch, denotes the th generation of the heuristic information for the th glass workpiece assigned to the first candidate list of the bth batch denotes the th generation of the heuristic information for the th glass workpiece assigned to the second candidate list of the bth batch denotes the th generation of the heuristic information for the th glass workpiece assigned to the oth target of the bth batch; Step 3.6.

4. Calculating the information pheromone for the othetarget of the bthbatch of glass workpieces using formula (59) the bthbatch of glass workpieces to the othetarget :​ (59) In formula (61), , denotes the th target when the th glass workpiece and the th glass workpiece in the th batch are simultaneously assigned to the th target; denotes the number of glass workpieces included in the th batch; Step 3.6.5, obtaining the first glass workpiece in the bth batch from the first glass workpiece in the bth batch and the first glass workpiece in the bth batch is assigned to the oth target in the bth batch updated pheromone (60)。 7. An electronic device comprising a memory and a processor, characterized in that The memory is used for storing a program supporting the processor to execute the method in any one of claims 1-6, and the processor is configured to execute the program stored in the memory.

8. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to execute the steps of the method in any one of claims 1-6.

Images