A method for optimizing flexible job shop scheduling considering machine preventive maintenance and controllable working hours

Abstract

The application discloses a kind of flexible job shop scheduling optimization methods considering machine pre-maintenance and working hours controllable, comprising: analyzing the influence of machine tool pre-maintenance and shutdown / restart strategy on flexible job shop energy consumption and modeling;MOPSO and NSGA-Ⅱ hybrid algorithm are used to optimize and solve the model, including initialization parameter, using roulette initialization method to generate initial population, non-dominated sorting and selection operation to population, updating particle velocity and position etc.;The effectiveness of the model and method is verified by examples, including giving the workpiece to be processed and workshop configuration information and analyzing the optimization results.The application is suitable for flexible job shop scheduling with machine tool equipment failure, machine idle waiting and other problems, which can effectively improve production efficiency, reduce energy consumption, and improve the stability and reliability of workshop scheduling.

Description

TECHNICAL FIELD

[0001] The present application relates to a flexible job shop scheduling method, in particular to a flexible job shop scheduling optimization method considering machine preventive maintenance and working hours controllable. BACKGROUND

[0002] Manufacturing equipment as the main body of equipment manufacturing industry is irreplaceable equipment in various fields of manufacturing industry, and has many energy sources, complex energy consumption characteristics and large total energy consumption, and is the main energy consumption in the manufacturing process. In a flexible job shop, the running condition of manufacturing equipment such as numerical control machine tools directly affects the production efficiency and energy consumption level.

[0003] In recent years, many scholars have carried out researches in many aspects for the flexible job shop scheduling problem, but most of the existing researches only consider minimizing the maximum completion time as the target to solve the single-objective FJSP, ignoring the energy consumption in the production process. With the increasingly serious environmental problems and the development of green manufacturing, it is becoming increasingly important to consider energy consumption related targets in production scheduling, but these researches all assume that the machine is continuously available.

[0004] In fact, the scheduling model considering the machine availability constraint is more in line with the actual production environment, and is helpful to improve the stability of production plan. Under this background, the integrated optimization of production scheduling and preventive maintenance plan has gradually attracted attention. In actual processing, machine tool equipment failure will cause processing interruption, energy consumption increase and scheduling influence, and machine idle waiting will also bring negative influence, therefore, it is very important to consider the preventive maintenance of machine tool equipment and when to start and stop the machine in production scheduling, and the current research results are very limited, so that the workshop scheduling plan cannot be close to the actual production situation, it is difficult to overcome the negative influence of machine tool failure and machine tool idle waiting, resulting in too low processing efficiency and consuming more energy. SUMMARY

[0005] The purpose of the present application is to provide a flexible job shop scheduling optimization method considering machine preventive maintenance and working hours controllable, which is suitable for flexible job shop scheduling, so that the workshop scheduling plan is more in line with the actual situation, improves the productivity and reduces the energy consumption.

[0006] The technical scheme of the present application is a flexible job shop scheduling optimization method considering machine preventive maintenance and working hours controllable, comprising the following steps:

[0007] S1: analyzing the influence of machine tool preventive maintenance, machine tool shutdown and restart strategy on the energy consumption of flexible job shop, and modeling the FJSP considering the machine tool preventive maintenance, machine tool shutdown and restart strategy, the model including objective function, constraint model;

[0008] S2: the model is solved by using a hybrid algorithm of MOPSO and NSGA-II, parameters are initialized, the maximum number of iterations is set, the values of inertia weight w, learning factor c1 and c2 are determined, and the parameters of NSGA are determined;

[0009] S3: the expression of the solution of FJSP of each particle is optimized by using a double encoding mode, two-level encoding is adopted for each chromosome, and the two are combined together;

[0010] S4: the roulette initialization mode is adopted, the maximum completion time of the machine tool and the energy consumption of the workshop are used as the fitness function, the optimization target weight is 0.6 and 0.4 respectively, and then the initial population is generated;

[0011] S5: the particle swarm is combined into a population, the population is subjected to non-dominated sorting, so as to divide different Pareto front layers; then, the crowding degree of the solution in each front layer is calculated, and the selection operation is carried out according to the non-dominated sorting result and the crowding degree, the excellent individuals are selected, the selected individuals are subjected to cross processing, the next generation of individuals is generated, and the next generation of individuals is subjected to mutation processing, so as to improve the diversity of the population;

[0012] S6: according to the particle motion rule, the particle velocity and position are updated, the fitness of the updated position is calculated, if the fitness of the new position is better than that of the individual optimal solution, the individual optimal solution is updated, so as to find the global optimal solution;

[0013] S7: steps S5 and S6 are repeated until the fitness reaches an acceptable range or the maximum number of iterations is reached, and finally the final Pareto front solution set is output, that is, the optimization result is obtained.

[0014] In the application, the MOPSO and NSGA-II hybrid algorithm is used to optimize and solve the flexible job shop scheduling problem considering machine pre-maintenance and shutdown / restart strategy; secondly, by analyzing the influence of machine pre-maintenance and shutdown / restart strategy on the energy consumption of the flexible job shop, a flexible job shop scheduling problem model considering energy consumption is established; thirdly, the population initialization and updating operations of the NSGA-II algorithm are used to realize the optimization of the algorithm; finally, the effectiveness of the model and method is verified through examples.

[0015] Further, in step S1, the setting of the established model includes: n kinds of to-be-processed workpieces n i , m available machine tools M m , wherein the batch of workpiece n i is q i , each workpiece n i includes s processes o ijWhen the machine tool is in pre-maintenance, shutdown and restart strategy, the machine tool is in the disabled state.

[0016] Further, in step S1, the objective function includes:

[0017] maxC m m∈[1,k];

[0018] Wherein, C m is the completion time, taking the maximum value of all process completion times of all machines processing all workpieces;

[0019]

[0020] Wherein, TE is the total energy consumption of the machine tool in the flexible job shop machining process, AE is the idle cutting energy consumption, CE is the cutting energy consumption of the machine tool, and IE is the idle energy consumption of the machine tool.

[0021] Further, in step S1, according to the workpiece to be processed and the workshop configuration information table of the flexible job shop production line, a constraint model is established based on the flexible job shop scheduling problem, process sequence and machine availability, including:

[0022] A1: Energy consumption constraint model;

[0023]

[0024] In the formula, P au is the auxiliary system power, P u is the machine tool idle power, is the tool setting time required before machining on the machine tool;

[0025]

[0026] In the formula, P cut is the cutting power, is the process o ijr machining time on the machine tool M m ;

[0027]

[0028] In the formula, P st is the machine standby power, indicates the idle waiting time of the machine M m , y m is a 0-1 decision variable, the machine M m implements shutdown / restart strategy, y m value is 1, otherwise 0;

[0029] A2: Production constraint;

[0030] Ensure that each process can only be implemented by a machine tool processing, and the workpiece processing needs to meet the same workpiece process processing sequence constraints:

[0031]

[0032] And

[0033] In the formula, x ijm Is a 0-1 variable, x ijmo =1 indicates that under the jth process route, the process o ij Select machine tool M m ;

[0034] A machine tool can only process one workpiece at the same time:

[0035]

[0036] In the formula, Indicates the start time of machine tool M m Processing the rth process, Indicates the completion time of machine tool M m Processing the rth process;

[0037] A3: Pre-maintenance constraints:

[0038] Machine tool pre-maintenance tasks and process processing cannot be carried out at the same time:

[0039]

[0040]

[0041] In the formula, Is the pre-maintenance start time of machine tool M m , Is the pre-maintenance end time of machine tool M m ;

[0042] At any time t, the machine tool equipment M m Participating in processing, its reliability R m Cannot be lower than the reliability threshold R0, otherwise the machine tool carries out pre-maintenance:

[0043] R m > R0;

[0044]

[0045] In the formula R mn (t) indicates the reliability function of machine tool M m At time t;

[0046] A4: Machine tool shutdown, restart constraints;

[0047] The shortest time required to implement the machine tool shutdown, restart strategy is not less than the no-load balance time of the machine tool:

[0048]

[0049] In the formula, M is the machine tool M m The completion time of processing the rth process.

[0050] Further, in step S2, the parameters of the NSGA include the NSGA-II crossover probability, the mutation probability, and the tournament selection size parameter, and the maximum number of iterations; according to the experimental results, the influence of the parameters on the algorithm performance is analyzed, and the parameters are gradually adjusted to find the optimal or better parameter combination.

[0051] Further, in step S3, the double encoding mode includes: using process ordering-based encoding, each gene in this layer represents the processing order relationship between different processing processes of all workpieces to be processed; and using machine allocation vector-based encoding, each gene in this layer represents the corresponding processing machine selected, wherein in the FJSP problem, each process corresponds to multiple processing machines;

[0052] Each particle in the solution space contains two gene segments of MA and OS, which are spliced before and after to form a particle with a length of 2T0 in the encoding process, wherein represents the sum of all process numbers of the workpiece, and the particle in the solution space is represented in the form of MA+OS.

[0053] Best, in step S5, the population iteration mode includes the following steps:

[0054] B1: Merge the initialized P t population of roulette selection, and generate the offspring population Q t through non-dominated sorting, selection, crossover and mutation; t

[0055] B2: Perform fast non-dominated sorting on the population R t with a size of 2N, and calculate the crowding degree of each individual in the non-dominated layer, and select appropriate individuals to form a new parent population P t+1 according to the non-dominated relationship and the crowding degree of the individual;

[0056] B3: Generate a new offspring population Q t+1 through NSGA-II, and merge P t+1 and Q t+1 to form a new population R​t The above operation is repeated until the maximum number of iterations is reached.

[0057] Best, in step S6, the particle updated model is:

[0058] v ij (t+1) = w x v ij (t) + c1 r1 [p ij (t) - x ij (t)] + c2 r2 [p gj (t) - x ij (t)].

[0059] x ij (t+1) = x ij (t) + v ij (t+1).

[0060] In the formula, w represents the degree to which the particle retains the velocity of the particle.

[0061] In the case of a larger value of w, the method has stronger global optimization ability; in the case of a smaller value of w, the algorithm has better local optimization performance. Generally, the inertia weight decreases with the increase of the number of iterations.

[0062] Best, in step S7, the fitness is set as the threshold of the maximum completion time and total energy consumption, and if the fitness reaches the acceptable range, a better solution that meets the requirements has been found, and the iteration can be stopped; the maximum number of iterations is the maximum number of iterations set to prevent the algorithm from falling into an infinite loop.

[0063] In each iteration, first update the particle velocity and position according to the particle motion rule, and then calculate the fitness of the updated position. If the fitness of the new position is better than the fitness of the individual optimal solution, update the individual optimal solution. Then, it is judged whether the fitness reaches the acceptable range. The fitness is set as the threshold of the maximum completion time and total energy consumption. If the fitness reaches the acceptable range, it means that a better solution that meets the requirements has been found, and the iteration can be stopped. In addition, it is judged whether the maximum number of iterations set is reached, and if the maximum number of iterations is reached, the iteration is stopped to prevent the algorithm from falling into an infinite loop.

[0064] Finally, the final Pareto front solution set is output. The Pareto front solution set is a collection of non-dominated solutions found in the multiple iteration process, which are optimized and balanced between the maximum completion time and the total energy consumption, representing different optimal solutions. By outputting the Pareto front solution set, the most suitable scheduling scheme can be selected according to the actual demand, so as to achieve the optimization goal of flexible job shop scheduling.

[0065] The present application is aimed at the problems of processing interruption, energy consumption increase, workshop total energy consumption increase and machine idle waiting caused by machine tool equipment failure in flexible job shop scheduling, and mainly includes three parts: a flexible job shop scheduling problem model is established, considering machine tool preventive maintenance, shutdown / restart strategy and energy consumption constraint factors; a MOPSO and NSGA-II hybrid algorithm combines the fast search and preliminary optimization ability of particle swarm algorithm and the crossover and mutation operation of non-dominated sorting genetic algorithm to increase the diversity of solutions; double-layer coding mode, roulette wheel population initialization and particle updating operations help to improve the solution efficiency and quality of the algorithm; and instance verification ensures the effectiveness of the model and method in practical application.

[0066] Advantages: Compared with the prior art, the present application has the advantages of:

[0067] The present application can effectively solve the problems of processing time extension and workshop total energy consumption increase caused by equipment failure and machine tool idling in workshop production scheduling. By considering machine tool preventive maintenance, shutdown / restart strategy and energy consumption constraint, a flexible job shop energy consumption model is established, and a MOPSO and NSGA-II hybrid algorithm is used for solving, which can balance the solution speed and global search ability to some extent, effectively reduce energy consumption and improve production efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0068] Figure 1 It is a schematic diagram of the framework of the present application;

[0069] Figure 2 It is a whole flow chart of the MOPSO and NSGA-II hybrid algorithm of the present application;

[0070] Figure 3 It is an initial scheduling scheme of the present application without considering various strategies;

[0071] Figure 4 It is a scheduling scheme Gantt chart of the present application considering three different strategy combination schemes;

[0072] Figure 5 It is an energy consumption composition analysis diagram of each scheme of the present application;

[0073] Figure 6 It is a time composition analysis diagram of each scheme of the present application. DETAILED DESCRIPTION

[0074] The present application will be further illustrated in conjunction with the drawings and specific embodiments, and it should be understood that these embodiments are only used to illustrate the present application and not to limit the scope of the present application.

[0075] A flexible job shop scheduling optimization method considering machine preventive maintenance and controllable working hours, as shown in Figures 1-6 includes the following steps:

[0076] S1: Analyze the impact of machine pre-maintenance and shutdown / restart strategy on the energy consumption of flexible job shop scheduling problem (FJSP), model the FJSP considering machine pre-maintenance and shutdown / restart strategy, including objective function and constraint function.

[0077] S2: Use the MOPSO and NSGA-II hybrid algorithm to optimize and solve the model, initialize the parameters, set the maximum number of iterations, determine the values of inertia weight w, learning factor c1 and c2, and the parameters related to non-dominated genetic algorithm.

[0078] S3: Use double encoding method to make each particle better express the solution of FJSP, use two-level encoding for each chromosome and combine them together.

[0079] S4: Use roulette initialization method, take the maximum completion time of machine and energy consumption of workshop as fitness function, and optimize the objective weight as 0.6 and 0.4 respectively, and then generate the initial population.

[0080] S5: Combine the particle swarm into a population, perform non-dominated sorting on the population, and then divide it into different Pareto frontiers; then, calculate the crowding degree of the solutions in each Pareto frontier, and select the excellent individuals according to the non-dominated sorting result and crowding degree, and perform crossover processing on the selected individuals to generate the next generation of individuals, and then perform mutation processing on the next generation of individuals to improve the diversity of the population.

[0081] S6: Find the global optimal solution. According to the particle motion rule, update the particle velocity and position. Calculate the fitness of the updated position. If the fitness of the new position is better than the fitness of the individual optimal solution, update the individual optimal solution.

[0082] S7: Repeat steps S5 and S6 until the fitness reaches an acceptable range or the maximum number of iterations is reached. Finally, output the final Pareto frontier solution set, which is the optimization result.

[0083] Based on the above scheme, the flexible job shop scheduling optimization method is applied in this embodiment, as follows:

[0084] The experimental data of this example comes from a flexible job shop with 8 machines and 3 types of workpieces to be processed. The machine power information is shown in Table 2, the processing batch, arrival time and delivery period are shown in Table 1, and the machine information that can be selected under different process routes for each workpiece in the workshop is shown in Table 3.

[0085] Table 1 Workpiece processing information

[0086]

[0087] Table 2 Machine tool power information

[0088]

[0089] Table 3 Workpiece processing procedure and optional machine tool information

[0090]

[0091] In the experiment, a flexible job shop is taken as the object for case analysis. The shop has 8 machine tools and 3 types of workpieces to be processed, and the experiment is carried out through relevant information setting. In the experiment, machine tool power information, processing batch, arrival time, delivery period, and optional machine tool information under different process routes of each workpiece are obtained.

[0092] The overall technical framework is as shown in Figure 1 The specific scheduling process is as follows:

[0093] 1) The assumptions of FJSP modeling considering machine pre-maintenance and shutdown / restart strategy are as follows:

[0094] An initial flexible job shop has n types of workpieces n i , m available machine tools M m , where the batch of workpiece n i is q i , and each workpiece n i includes s process o ij . When the machine implements pre-maintenance plan and shutdown / restart strategy, the machine is in a disabled state.

[0095] 2) The optimization objective function of FJSP considering machine pre-maintenance and shutdown / restart strategy is as follows:

[0096] max C m m∈[1,k]

[0097] In the formula, the completion time C m is the maximum value of the completion time of all machine tools processing all workpieces.

[0098]

[0099] In the formula, the total energy consumption TE of the flexible job shop processing process machine tool is mainly composed of five parts, including workpiece clamping and dismounting energy consumption, air cutting energy consumption AE, machine tool cutting processing period energy consumption CE, tool changing energy consumption, and machine tool idle energy consumption IE. The workpiece clamping and dismounting energy consumption and the tool changing energy consumption have less influence on the total energy consumption of the shop, and are not considered.

[0100] 3) The constraint model is as follows:

[0101] According to the flexible job-shop production line's workpiece and workshop configuration information table, determine its based on flexible job-shop scheduling problem and based on process sequence, machine availability constraint condition.

[0102] A1: Energy consumption constraint model;

[0103]

[0104] In the formula, P au is the auxiliary system power, P u is the machine tool idle power, is the tool setting time required before machine tool processing;

[0105]

[0106] In the formula, P cut is the cutting processing power, is the process o ijr The cutting time on the machine tool M m ;

[0107]

[0108] In the formula, P st is the machine tool standby power, represents the idle waiting time of the machine tool M m , y m is a 0-1 decision variable, the machine tool M m implements the shutdown / restart strategy, y m value is 1, otherwise 0;

[0109] A2: Production constraint;

[0110] Ensure that each process can only be processed by one machine, and the workpiece processing needs to meet the process sequence constraint of the same workpiece:

[0111]

[0112] And

[0113] In the formula, x ijm is a 0-1 variable, x ijmo =1 indicates that under the jth process route, the process o ij selects the machine tool M m .

[0114] A machine tool can only process one process of a workpiece at the same time:

[0115]

[0116] wherein indicates the machine tool M m the start time of the rth machining process, indicates the machine tool M m the end time of the rth machining process.

[0117] A3: Pre-maintenance constraint:

[0118] Machine tool pre-maintenance tasks and process machining cannot be performed simultaneously:

[0119]

[0120] wherein, is the pre-maintenance start time of the machine tool M m , is the pre-maintenance end time of the machine tool M m ;

[0121] At any time t, the machine tool equipment M m participating in machining m cannot be lower than the reliability threshold R0, otherwise the machine tool performs pre-maintenance:

[0122] R m > R0

[0123]

[0124] wherein R mn (t) represents the reliability function of the machine tool M m at time t;

[0125] A4: Machine tool shutdown / restart constraint

[0126] The shortest time required to implement the machine tool shutdown / restart strategy is not less than the no-load balance time of the machine tool:

[0127]

[0128] wherein is the end time of the rth machining process of the machine tool M m .

[0129] 4) The algorithm parameter settings are:

[0130] Determine the related parameters of the inertia weight w, learning factors c1 and c2 of MOPSO, and the related parameters of the crossover probability, mutation probability and tournament selection size of NSGA-II, and finally determine the maximum number of iterations.

[0131] According to the experimental results, analyze the influence of parameters on the performance of the algorithm, gradually adjust the parameters, and find the optimal or better parameter combination.

[0132] 5) The double-layer encoding method is as follows:

[0133] Operation Sequence (OS) based encoding: Each gene in this layer represents the processing sequence relationship between different processing procedures of all workpieces to be processed. Machine Assignment (MA) based encoding: Each gene in this layer represents the selected corresponding processing machine.

[0134] Each particle in the solution space contains MA and OS two gene segments, which are spliced together to form a particle with a length of 2T0 in the encoding process, where represents the sum of all processing procedures of the workpiece, and thus the particle in the solution space can be represented in the form of MA+OS.

[0135] 6) The population initialization method is as follows:

[0136] When solving the FJSP problem, for intelligent optimization algorithms, generating an initial population is the starting point of algorithm optimization, and thus its quality has a great influence on the quality of the final optimal solution in a certain sense.

[0137] A high-quality initial population can improve the convergence speed of the algorithm. Considering that the optimization objective is to minimize the maximum completion time and total energy consumption, the roulette selection population initialization method is used to generate the initial population. This method increases the chances of selecting excellent individuals in the algorithm, which helps the algorithm converge to the optimal solution faster.

[0138] The roulette selection population initialization method is used to generate the initial population, with the machine maximum completion time and workshop energy consumption as the fitness function, and the optimization objective weights being 0.6 and 0.4, respectively.

[0139] 7) The population iteration method is as follows:

[0140] B1: Merge the roulette selection initialized P t population, and generate the offspring population Q t through non-dominated sorting, selection, crossover, and mutation. t

[0141] B2: Perform fast non-dominated sorting on the population R t with a size of 2N, and calculate the crowding degree of each individual in the non-dominated layer. According to the non-dominated relationship and the crowding degree of the individual, select appropriate individuals to form a new parent population P t+1

[0142] ​​B3: Generate a new offspring population Q by the basic operation of NSGA-II t+1 , P t+1 is merged with Q t+1 to form a new population R t . Repeat the above operations until the maximum number of iterations is reached.

[0143] 8) Particle update is:

[0144] v ij (t+1) = w x v ij (t) + c1 r1 [p ij (t) - x ij (t)] + c2 r2 [p gj (t) - x ij (t)]

[0145] x ij (t+1) = x ij (t) + v ij (t+1)

[0146] In the formula, w represents the extent to which the particle retains the velocity of the particle; in the case of a larger value of w, the method has stronger global optimization ability; in the case of a smaller value of w, the algorithm has better local optimization performance. Generally, the inertia weight decreases with the increase of the number of iterations.

[0147] 9) Iterative output of Pareto front solution set:

[0148] In the process of solving, steps S5 and S6 need to be repeatedly executed to constantly update the position and velocity of the particle and find the optimal solution. Specifically, in each iteration, first update the particle velocity and position according to the particle motion rule, then calculate the fitness of the updated position. If the fitness of the new position is better than the fitness of the individual optimal solution, update the individual optimal solution. Then, judge whether the fitness is within the acceptable range. The fitness is set with a maximum completion time and total energy consumption threshold. If the fitness is within the acceptable range, it means that a better solution that meets the requirements has been found, and the iteration can be stopped. In addition, it is judged whether the maximum number of iterations is reached. If the maximum number of iterations is reached, the iteration is stopped to prevent the algorithm from falling into an infinite loop.

[0149] Finally, the final Pareto front solution set is output. The Pareto front solution set is a collection of non-dominated solutions found in multiple iterations, which are optimized and balanced between maximum completion time and total energy consumption, representing different optimal solutions. By outputting the Pareto front solution set, the most suitable scheduling scheme can be selected according to actual needs, thereby achieving the optimization goal of flexible job shop scheduling.

[0150] In order to verify the effect of the flexible job shop scheduling method proposed in the application, the experimental data of the present example comes from a flexible job shop which has 8 machine tools and 3 kinds of workpieces to be processed. According to the machine tool and workpiece information, the initial scheduling scheme generated based on the MOPSO and NSGA-II hybrid algorithm has the maximum completion time and the total energy consumption of the workshop as the target, as shown in Figure 3 . In addition, the scheduling scheme considering the periodic maintenance strategy, the scheduling scheme considering the periodic maintenance strategy and the shutdown / restart strategy, and the scheduling scheme considering the machine tool preventive maintenance and the machine tool shutdown / restart strategy are shown in Figure 4 (a), (b) and (c) respectively. From the comparison of each figure, it can be seen that the method proposed in the application reasonably arranges the process flow under the condition of considering the preventive maintenance strategy and arranges the machine tool shutdown under the condition of considering the machine tool shutdown / restart strategy. In order to analyze the energy consumption and time composition, the energy consumption data and time composition of each scheme are analyzed and compared, and the comparison results are shown in Figure 5 , Figure 6 . It can be seen from the figure that the cutting energy consumption accounts for the largest proportion, followed by the machine tool idle energy consumption and the air cutting energy consumption. In terms of time composition, the cutting time accounts for the largest proportion, followed by the idle time, and the air cutting time accounts for the smallest proportion. The comparison of different schemes shows that the scheduling scheme considering the machine tool preventive maintenance and the machine tool shutdown / restart strategy reduces the total energy consumption by 5.8% and shortens the maximum completion time by 4.4%.

Claims

1. A flexible job shop scheduling optimization method considering machine preventive maintenance and controllable working hours, characterized in that The method comprises the following steps: S1: analyzing the influence of machine tool pre-maintenance, machine tool shutdown and restart strategy on the energy consumption of the flexible job shop, and modeling the FJSP considering the machine tool pre-maintenance, machine tool shutdown and restart strategy, the model comprising a target function and a constraint model; S2: using a hybrid algorithm of MOPSO and NSGA-II to optimize and solve the model, initializing parameters, setting the maximum number of iterations, determining the values of the inertia weight w, the learning factors c1 and c2, and the parameters of NSGA; S3: optimizing the expression of the solution of the FJSP of each particle through double encoding, using two-level encoding for each chromosome, and combining them together; S4: using the roulette initialization method, taking the maximum machine tool completion time and the energy consumption of the workshop as the fitness function, and optimizing the target weight to be 0.6 and 0.4 respectively, and then generating an initial population; S5: combining the particle swarm into a population, performing non-dominated sorting on the population, and thus dividing different Pareto front layers; then, calculating the crowding degree of the solutions in each front layer, and selecting excellent individuals according to the non-dominated sorting result and the crowding degree, performing crossover processing on the selected individuals, generating the next generation of individuals, and performing mutation processing on the next generation of individuals to improve the diversity of the population; S6: updating the particle speed and position according to the particle motion rule, calculating the fitness of the updated position, and if the fitness of the new position is better than that of the individual optimal solution, updating the individual optimal solution, so as to find the global optimal solution; S7: repeating steps S5 and S6 until the fitness reaches an acceptable range or the maximum number of iterations is reached, and finally outputting the final Pareto front solution set, i.e. the optimization result.

2. The method for flexible job shop scheduling optimization considering machine preventive maintenance and controllable working hours according to claim 1, characterized in that, In step S1, the settings of the established model include: n kinds of workpieces n to be processed commonly in the initial workshop i , m available machine tools M m , where the batch of workpieces n i is q i , each workpiece n i includes s procedures o ij ; when the machine tool carries out pre-maintenance, shutdown and restart strategy, the machine tool is in the disabled state. 3.The method of claim 1, wherein, In step S1, the target function comprises: maxC m m∈[1,k] where C m is the completion time, which is the maximum of all process completion times for all workpieces on all machines; Wherein, TE is the total energy consumption of the machine tool in the flexible job shop processing process, AE is the idle cutting energy consumption, CE is the energy consumption of the machine tool during cutting processing, and IE is the idle energy consumption of the machine tool.

4. The method for flexible job shop scheduling optimization considering machine preventive maintenance and controllable working hours according to claim 1, characterized in that, In step S1, according to the information table of the workpieces to be processed and the workshop configuration of the flexible job shop, a constraint model is established based on the flexible job shop scheduling problem, the process sequence, and whether the machine tool is available, comprising: A1: energy consumption constraint model; where P au is the auxiliary system power, P u is the machine tool idle power, is the time required for tool setting before machining the machine tool. In the formula, P cut is the cutting machining power, is the machining time of the process o ij In the machine tool M m cutting machining time; where P st is the machine tool standby power, represents the idle waiting time of the machine tool M m , y m is a 0-1 decision variable, the machine tool M m implements the shutdown / restart strategy, y m has the value 1, otherwise 0; A2: production constraint; Ensure that each process can only be processed by one machine tool, and that the workpiece processing needs to meet the process sequence constraint of the same workpiece: wherein x ijm is a 0-1 variable, x ijm = 1 indicates that process o ij is selected for machine tool M m ; A machine tool can only process one workpiece at the same time: In the formula, the machine tool M m start time of the rth machining process, the machine tool M m finish time of the rth machining process, A3: pre-maintenance constraint: The machine tool pre-maintenance task and the process processing cannot be performed at the same time: wherein is the start time of the pre-maintenance of the machine tool M m is the end time of the pre-maintenance of the machine tool M is the end time of the pre-maintenance of the machine tool M m is the end time of the pre-maintenance of the machine tool M At any instant t, machine tool equipment M involved in the process m its reliability R m Cannot be lower than the reliability threshold R0, otherwise the machine tool performs preventive maintenance: R m > R0; wherein R mn (t) indicates the machine M m reliability function at time t A4: machine tool shutdown and restart constraint; The shortest time required for implementing the machine tool shutdown and restart strategy is not less than the idle balance time of the machine tool: In the formula, The machine tool M m The completion time of the rth machining pass.

5. The method for flexible job shop scheduling optimization considering machine preventive maintenance and controllable working hours according to claim 1, characterized in that, In step S2, the parameters of NSGA include the NSGA-II crossover probability, the mutation probability and the tournament selection size parameter, and the maximum number of iterations; according to the experimental results, the influence of the parameters on the performance of the algorithm is analyzed, the parameters are gradually adjusted, and the optimal or better parameter combination is found.

6. The method for flexible job shop scheduling optimization considering machine preventive maintenance and controllable working hours according to claim 1, characterized in that, In step S3, the double encoding mode includes: using encoding based on process sequencing, each gene in this layer represents the processing sequence relationship between different processing procedures of all workpieces to be processed; using encoding based on machine allocation vector, each gene in this layer represents the corresponding processing machine selected, and in the FJSP problem, each processing procedure corresponds to multiple processing machines; Each particle in the solution space contains two parts of gene segments, MA and OS, which are spliced together to form a particle of length 2T0 in the encoding process, where Let S denote the sum of all the operation numbers of the workpieces, then the particle in the solution space is represented in the form of MA+OS.

7. The method for flexible job shop scheduling optimization considering machine preventive maintenance and controllable working hours according to claim 1, characterized in that, In step S5, the population iteration mode includes the following steps: B1 : roulette wheel selection is initialized P t The two populations are combined, non-dominated sorted, selected, crossed and mutated to produce a child population Q of size 2N t The two populations are combined to form a population R of size 2N t ; B2: a population R of size 2N t Fast non-dominated sorting is performed, and at the same time, the crowding distance of each individual in each non-dominated layer is calculated. According to the non-dominated relationship and the crowding distance of the individual, appropriate individuals are selected to form a new parent population P t+1 ; B3: Generate new offspring population Q by NSGA-II t+1 P t+1 is combined with Q t+1 to form a new population R t The above operations are repeated until the maximum number of iterations is reached.

8. The method for flexible job shop scheduling optimization considering machine preventive maintenance and controllable working hours according to claim 1, characterized in that, In step S6, the particle updating model is: v ij (t+1) = w x v ij (t) + c1 r1[p ij (t) - x ij (t)] + c2 r2[p gj (t) - x ij (t)]; x ij (t+1) = x ij (t) + v ij (t+1); In the formula, w represents the degree to which the particle retains the speed of the particle.

9. The method for flexible job shop scheduling optimization considering machine preventive maintenance and controllable working hours according to claim 1, characterized in that, In step S7, the fitness is the threshold value of the set maximum completion time and total energy consumption, if the fitness reaches the acceptable range, an optimal solution meeting the requirements has been found, and the iteration can be stopped; the maximum iteration number is the maximum number set to prevent the algorithm from falling into an infinite loop.

Images