A Comprehensive Scheduling Method for PCB Assembly Lines Considering Space Compensation and Energy Constraints

Abstract

This invention relates to the field of intelligent manufacturing scheduling optimization technology, proposing a comprehensive scheduling method for PCB assembly lines that considers spatial compensation and energy constraints. This method includes converting the temporal constraints of process waiting into buffer physical space occupancy constraints by calculating equivalent dwell time, and constructing a spatial compensation mechanism. A mathematical model is established with the objectives of smoothing production load and minimizing total penalty cost. At the algorithmic level, an initial population is constructed using a four-segment chromosome structure including task priority and workstation segmentation, and a generalized task time matrix is ​​generated by combining vehicle assignment offset. An improved artificial bee colony algorithm is used, leveraging the collaborative search of foraging bees, observation bees, and scout bees, along with a mask-based crossover operator, to optimize the scheme while preserving process logic. Finally, the iteration results are decoded into production instructions and energy replenishment strategies, achieving deep collaboration between task allocation and energy scheduling, effectively improving the production efficiency and stability of assembly lines under complex constraints.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing scheduling optimization technology, and in particular to a comprehensive scheduling method for PCB assembly lines that considers space compensation and energy constraints. Background Technology

[0002] In PCB assembly production within the electronics manufacturing sector, the increasing integration of products and the growing demand for flexible manufacturing have significantly increased the process density and logistics interaction frequency of assembly lines. Existing technologies often employ scheduling models based on idealized assumptions, assuming unlimited material supply and continuous availability of logistics vehicles, neglecting the rigid constraints of the physical capacity of workstation buffers and the dynamic fluctuations in the energy status of vehicles such as AGVs. This leads to frequent space overflows caused by processes waiting for materials to continuously occupy buffers, or logistics response delays due to vehicles running out of power mid-journey, resulting in production interruptions. Furthermore, existing heuristic algorithms lack effective space-based guidance mechanisms when dealing with the multi-dimensional tightly coupled constraints of task allocation, space capacity, and energy management. This makes it difficult to obtain feasible solutions that balance production cycle time and logistics stability within a reasonable timeframe, resulting in insufficient feasibility and robustness of the scheduling results. Summary of the Invention

[0003] To address the aforementioned shortcomings, the present invention aims to propose a comprehensive scheduling method for PCB assembly lines that considers both spatial compensation and energy constraints. This method transforms process waiting time constraints into physical space occupancy constraints by constructing a spatial compensation mechanism for equivalent dwell time, and establishes a multi-objective collaborative optimization framework that integrates energy perception. This enables deep collaboration between production task allocation, workstation space capacity, and the energy status of logistics vehicles, thereby improving the scheduling robustness and execution accuracy of PCB assembly lines under multi-constraint environments while ensuring the physical feasibility of the solution.

[0004] To achieve this objective, the present invention adopts the following technical solution: A comprehensive scheduling method for PCB assembly lines that considers space compensation and energy constraints includes: Calculate the equivalent dwell time corresponding to the processing time of the process and the forced process waiting time, transform the timing constraint of process waiting into the physical space occupation constraint of materials on the workstation buffer, and construct space compensation constraint; A mathematical model is constructed with the goal of minimizing the production load smoothing index and the total penalty cost, wherein the total penalty cost includes logistics timeliness penalty, space violation penalty based on the spatial compensation constraint, and energy safety penalty based on the vehicle's remaining power. An initial nectar source population was constructed using a four-segment chromosome structure, including task priority, workstation segmentation, task summary, and vehicle assignment offset. The task priority sequence is truncated using the workstation segmentation to construct a task assignment matrix, and a generalized task time matrix reflecting the task time node status is generated by combining the vehicle assignment offset and spatial compensation constraints. An improved artificial bee colony algorithm is used to search through the local mutation of foraging bees, the global development of observation bees, and the random reset mechanism of scout bees. The optimal scheduling scheme is updated by using a mask-based crossover operator while preserving the process logic. The iteration results that meet the termination conditions are decoded into production instructions and logistics replenishment strategies, and the PCB assembly line is controlled to complete task allocation and energy scheduling.

[0005] Preferably, calculating the equivalent dwell time corresponding to the processing time of the process and the forced process waiting time, and transforming the timing constraint of process waiting into the physical space occupancy constraint of materials on the workstation buffer includes: Define task At the workstation Equivalent dwell time of the buffer The equivalent residence time Satisfying the relation: ; in, Indicate process At the workstation Equivalent dwell time, Indicates completion of process The required standard physical processing time, Indicate process The corresponding process requires a minimum waiting time; Calculate any time interval workstation Dynamic space usage The dynamic space occupancy Satisfying the relation: ; in, Indicates time workstation Dynamic space usage, Indicates at time The set of tasks that are still in an equivalent resident state and satisfy the following conditions: ,in Indicate process The start time of processing Indicate process The number of space units occupied by the required materials within the workstation buffer zone; Through the dynamic space occupancy Spatial exclusion checks are performed on subsequent logistics loading operations to ensure the workstation The spatial compensation flexible constraint is satisfied at all times, and the spatial compensation flexible constraint satisfies the following relationship: ; in, Indicates workstation Maximum physical space limit for storing materials.

[0006] Preferably, the mathematical model is constructed with the objectives of minimizing the production load smoothing index and the total penalty cost, including: Calculate the production smoothing index to quantify the deviation between the workload and production cycle time of each workstation. The production smoothing index Satisfying the relation: ; in, Indicates the production smoothing index. This indicates the total number of workstations. This indicates the preset production cycle time. Indicates allocation to the first A set of procedures performed by a workstation. Indicate process At the workstation The equivalent dwell time; Constraints based on time, space, and energy dimensions: total penalty cost The total penalty cost Satisfying the relation: ; in, Indicates the total cost of punishment. Indicates workstation Space violation penalties Indicates logistics task Time-limited penalties Indicates automated guided vehicle Energy security penalties This indicates the total number of workstations. Indicates the total number of processes. Indicates the total number of automated guided vehicles; Among them, the space violation penalty Calculations based on the aforementioned spatial compensation constraints satisfy the following relationship: ; in, Indicates penalties for violations of space regulations. Indicates the spatial violation coefficient. Indicates time workstation Dynamic space usage, Indicates workstation Maximum physical space limit for storing materials; Logistics timeliness penalty obtained by evaluating logistics delivery time deviation The aforementioned logistics timeliness penalty Satisfying the relation: ; in, Indicates logistics task Time-limited penalties This indicates a penalty factor for early delivery. Indicates the penalty factor for delayed delivery. Indicates logistics task Expected delivery time Indicates logistics task The actual completion time; Among them, the energy safety penalty Satisfying the relation: ; in, Indicates energy security penalty, Indicates the energy loss penalty factor. Represents a symbolic function. This indicates the power threshold that triggers forced charging. This indicates the current remaining battery power of the automated guided vehicle. This indicates the time loss caused by the charging process. By adaptively balancing production smoothness and violation costs, a comprehensive optimization objective function is constructed. The comprehensive optimization objective function Satisfying the relation: ; in, This represents the comprehensive optimization objective function value. and These represent the adaptive weighting coefficients used to dynamically balance production smoothness and total penalty cost, respectively. Indicates the production smoothing index. This represents the total cost of the penalty.

[0007] Preferably, the initial nectar source population is constructed using a four-segment chromosome structure, including task priority, workstation segmentation, task summary, and vehicle assignment offset. Construct a chromosome coding structure consisting of four gene loci, wherein the chromosome coding structure satisfies the following relation: ; in, Indicates the individual nectar source. This represents a task priority segment arranged by task process index. This represents a workstation segmentation control segment, composed of binary or integer components, used to identify workstation cutting points. This section represents the workstation task summary segment, which records the number of tasks assigned to each workstation. This indicates the vehicle assignment and logistics offset segment, which includes the vehicle index and logistics response time correction value. Initialize the task priority segment By random generation The process steps are arranged and their validity is verified using topological sorting logic to ensure the priority of the tasks. Satisfying the preceding process constraints between processes, among which Indicates the total number of processes; The workstation segmentation control section is initialized using randomly generated cutting points. And according to the workstation segmented control section For the task priority segment Divide the tasks and update the workstation task summary section synchronously. To clarify the task attribution relationships of each workstation; For each logistics interaction task, a corresponding vehicle index and its logistics offset relative to the production completion time are randomly assigned. And fill the vehicle assignment and logistics offset segment. This completes the construction of the initial nectar source population.

[0008] Preferably, the process of using the workstation segmentation to truncate the task priority sequence to construct a task assignment matrix includes: Based on the workstation segmented control section The determined cut point information will be used to prioritize the task. The process index sequence is sequentially assigned to the corresponding workstations; Establish a production task assignment matrix The production task assignment matrix Satisfying the relation: ; in, This represents the production task assignment matrix. This indicates the decision variable for process assignment, when the process... Cut off and assigned to workstation hour, The value is 1 if it is not 0 otherwise. Indicates the total number of workstations; Verify the production task assignment matrix Check if the task capacity limit of each workstation is met. If not, adjust the workstation segment control section. The gene values ​​are re-trunculated and reassigned.

[0009] Preferably, generating a generalized task time matrix reflecting the state of task time nodes by combining the vehicle assignment offset and spatial compensation constraints includes: Based on the production task assignment matrix The established task hierarchy, combined with the vehicle assignment and logistics offset segment. Logistics offset in and the equivalent dwell time of each process. Construct a generalized task time matrix The generalized task time matrix Satisfying the relation: ; in, Indicate process The generalized task time matrix, Indicates the actual completed processes of the logistics vehicle. The delivery time, Indicate process At the workstation The equivalent dwell time; Wherein, the actual completion time of the logistics task Satisfying the relation: ; in, Indicates the actual completed processes of the logistics vehicle. The delivery time, Indicate process The start time of processing Indicate process At the workstation Equivalent dwell time, This indicates the vehicle assignment and logistics offset segment. The logistics offset recorded in the middle, Indicates the assigned automated guided vehicle The estimated transport time required to execute this route segment.

[0010] Preferably, the improved artificial bee colony algorithm is used to search through the local mutation of foraging bees, the global development of observation bees, and the random reset mechanism of scout bees, and the optimal scheduling scheme is updated using a mask-based crossover operator while preserving the process logic, including the following steps: S10: Foraging bees are at the current nectar source Within the neighborhood, for task priority segments Perform genetic operations based on the crossover operator and simultaneously adjust the segmented control sections of the workstation. With task summary section ; S11: Using a greedy selection mechanism to compare the comprehensive objective function of offspring nectar sources and original nectar sources. If the offspring is better, replace it; otherwise, retain the original nectar source and accumulate the stagnation count of that nectar source. S12: The observation bee determines the nectar source to be mined based on the fitness value of the nectar source through a probabilistic selection mechanism. The selected high-quality nectar source is subjected to the same crossover, mutation and greedy selection strategies as the foraging bee stage, so as to achieve in-depth mining of the excellent area in the search space and update the global optimal solution. S13: Real-time monitoring of the stagnation count of each nectar source; when the stagnation count of a certain nectar source reaches a preset threshold... If the nectar source is found to be in a local optimum, the scout bee will discard it and randomly generate a new nectar source in the search space according to the process topology constraints to replace it. Repeat steps S10-S13 until the maximum number of iterations is reached. Or the objective function converges.

[0011] Preferably, the crossover operator is a PPX crossover operator, and the execution of the PPX crossover operator includes the following steps: S20: Select parental nectar source and And create a random binary mask vector of the same length as the task sequence. ; S21: Check bit by bit The One element: If Then from Extract the leftmost unselected task from the current generation sequence and add it to the child sequence. Delete the task; S22: If Then from Extract the leftmost unselected task from the current generation sequence and add it to the child sequence. Delete the task; Repeat steps S20-S22 until the child sequence is filled. Use a mask to guide the child to automatically inherit the order of the process logic from the parent to avoid generating illegal solutions.

[0012] One of the above technical solutions has the following advantages or beneficial effects: This invention transforms the previously difficult-to-handle timing wait constraints into continuous occupancy constraints of the workstation buffer's physical space by calculating the equivalent dwell time corresponding to the processing time of each process and the forced process waiting time. This allows for the prediction and prevention of buffer overflow risks during the scheduling construction phase. Furthermore, it constructs a collaborative optimization objective using a production load smoothing index and multi-dimensional penalty costs encompassing logistics timeliness, space violations, and energy security. This enables the algorithm to proactively avoid space and energy violations while pursuing production line load balance. A four-segment chromosome structure encoding—task priority, workstation segmentation, task aggregation, and vehicle assignment offset—is used to achieve an integrated expression of task sequences, workstation allocation, and logistics scheduling, ensuring that the solution space covers the three-dimensional decision variables of task, space, and energy. Task priority sequences are further segmented through workstations. The task assignment matrix is ​​truncated and constructed, and a generalized task time matrix is ​​generated by combining vehicle assignment offset and spatial compensation constraints. This enables the decoding process to have physical perception capabilities, allowing for the verification of space occupancy and energy status during the matrix transformation stage. An improved artificial bee colony algorithm is used, which coordinates the search through the local mutation of foraging bees, the global development of observation bees, and the random reset mechanism of scout bees. A mask-based crossover operator is used to efficiently update the solution set while preserving the process logic, effectively avoiding the local optimum trap caused by the tight coupling of multi-dimensional constraints. Finally, the optimization results are decoded into directly executable production instructions and logistics energy replenishment strategies, realizing the autonomous coordination of task allocation and energy scheduling in PCB assembly lines under given cycle time constraints. This significantly improves the executability and system robustness of the scheduling scheme under complex physical constraints. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0014] Figure 1 This is a flowchart of a PCB assembly line integrated scheduling method that considers space compensation and energy constraints, provided in an embodiment of the present invention. Detailed Implementation

[0015] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0016] In this invention, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0017] A comprehensive scheduling method for PCB assembly lines that considers space compensation and energy constraints, such as Figure 1 As shown, a preferred embodiment of the present invention includes the following steps: S1: Calculate the equivalent dwell time corresponding to the processing time of the process and the forced process waiting time, transform the timing constraint of process waiting into the physical space occupation constraint of materials on the workstation buffer, and construct space compensation constraint. It should be noted that process processing time refers to the standard physical operation time required to complete a specific PCB assembly process, such as the pure operation time required to solder components to the circuit board in the surface mount process; forced process waiting time refers to the resting time interval that must be observed during PCB production due to physical or chemical properties, such as the curing wait after dispensing or the cooling wait after soldering; equivalent dwell time refers to the total time that materials occupy in the workstation buffer after the above two times are combined and calculated; space compensation constraint refers to the constraint mechanism that transforms the timing wait into space occupation, so that the time delay that is originally difficult to model directly can be represented and controlled in the form of physical space occupation.

[0018] Understandably, by calculating the equivalent dwell time, the process waiting constraints that originally belonged to the time dimension are transformed into the occupancy constraints of the physical space dimension. This avoids the introduction of virtual processes that would cause model dimension explosion, and thus the buffer occupancy status can be predicted during the scheduling construction stage. This prevents the risk of space overflow caused by the continuous accumulation of process waiting materials, and realizes the integrated modeling of temporal logic and physical space occupancy.

[0019] S2: Construct a mathematical model with the goal of minimizing the production load smoothing index and the total penalty cost, wherein the total penalty cost includes logistics timeliness penalty, space violation penalty based on the spatial compensation constraint, and energy safety penalty based on the vehicle's remaining power. It should be noted that the production load smoothing index is a statistical measure that quantifies the deviation between the operating load of each workstation and the preset production cycle time, and is used to assess the degree of load balance on the production line; the total penalty cost is a system evaluation function that comprehensively considers violations of multi-dimensional constraints of time, space and energy; the logistics timeliness penalty refers to the violation cost incurred due to the logistics delivery time deviating from the expected delivery time, including both early delivery and late delivery; the space violation penalty refers to the penalty triggered when the actual volume occupied by the workstation buffer exceeds the maximum physical limit; and the energy safety penalty refers to the risk warning cost incurred when the remaining power of the AGV vehicle is lower than the safety threshold.

[0020] Understandably, by constructing a dual-objective optimization model that integrates production smoothness and multi-dimensional penalty costs, while pursuing a balanced distribution of operation time for each workstation, the algorithm actively avoids violations such as logistics time deviation, spatial overflow, and energy depletion. This allows the algorithm's search direction to balance efficiency and feasibility, guiding the solution to converge toward a feasible region that simultaneously satisfies cycle balance and physical constraints.

[0021] S3: The initial nectar source population is constructed using a four-segment chromosome structure that includes task priority, workstation segmentation, task summary, and vehicle assignment offset. It should be noted that the task priority segment refers to the gene sequence of the execution order of storage processes, and its arrangement must meet the process topology constraints of PCB assembly; the workstation segment control segment refers to the set of cutting points that identify the boundaries of the task sequence between workstations; the workstation task summary segment refers to the statistical vector that records the number of tasks assigned to each workstation; the vehicle assignment offset segment refers to the decision variable set containing the AGV vehicle index and the logistics response time correction value; the four-segment chromosome structure refers to the complete coding scheme formed by concatenating the above four gene segments, which is used to simultaneously characterize task sorting, workstation allocation and logistics scheduling decisions.

[0022] Understandably, by constructing a four-segment chromosome structure, the decision variables of four dimensions—task sequencing, workstation allocation, task counting, and logistics assignment—are integrated into a unified code, enabling a single honey source individual to fully express all decision information of the PCB assembly line's comprehensive scheduling. This ensures that the coding scheme covers the three-dimensional optimization requirements of task-space-energy, providing a structured data foundation for subsequent decoding and matrix transformation.

[0023] S4: The task priority sequence is truncated using the workstation segmentation to construct a task assignment matrix, and a generalized task time matrix reflecting the task time node status is generated by combining the vehicle assignment offset and spatial compensation constraints. It should be noted that the task assignment matrix refers to a two-dimensional decision matrix describing the relationship between process and workstation allocation, and its elements indicate whether a specific process is assigned to a specific workstation; the generalized task time matrix refers to a time state matrix that records the actual completion time and equivalent dwell time of a process; the truncation operation refers to the process of dividing a continuous sequence of task priorities into several workstation subsets based on the cutting points of the workstation segment control segment; and the time node state refers to a set of states including key time points such as process start, processing completion, process waiting end, and actual logistics delivery.

[0024] It is understandable that by segmenting the task priority sequence by workstations, the clear allocation of tasks among physical workstations can be achieved, and a task assignment matrix can be constructed. The actual delivery time of logistics is calculated by combining the vehicle assignment offset and the equivalent dwell time under spatial compensation constraints to generate a generalized task time matrix. This enables the decoding process to have physical perception capabilities and can simultaneously represent the workstation allocation status and the time and space occupancy status at the matrix level.

[0025] S5: An improved artificial bee colony algorithm is used to search through the local mutation of foraging bees, the global development of observation bees, and the random reset mechanism of scout bees. The optimal scheduling scheme is updated by using a mask-based crossover operator while preserving the process logic. It should be noted that the improved artificial bee colony algorithm refers to a hybrid optimization algorithm that introduces genetic operators and greedy selection mechanisms on the basis of the traditional artificial bee colony algorithm; the foraging bee refers to the bee individual responsible for local search within the neighborhood of the current nectar source, exploring the neighboring solution space by performing crossover and mutation operations; the observation bee refers to the bee individual that conducts global development based on the fitness value of the nectar source through a probabilistic selection mechanism, realizing in-depth mining of high-quality areas; the scout bee refers to the bee individual responsible for discarding the nectar source and randomly generating a new nectar source in the search space when the stagnation count of a nectar source reaches a threshold; the mask-based crossover operator refers to the genetic operation that guides the recombination of parent genes through random binary mask vectors, ensuring that the offspring inherit the parent genes in the order of process logic; the stagnation count refers to the counter that records the number of consecutive iterations in which the nectar source has not been improved.

[0026] Understandably, by having foraging bees perform local mutations to conduct a fine search in the neighborhood of high-quality solutions, by having observation bees perform global development to expand the search range, and by having scout bees perform random resets to escape local optimum traps, the three types of bee colonies work together to achieve full coverage of the search space. By using a mask-based crossover operator to automatically retain the process logic constraints between processes during the recombination process, illegal solutions are avoided, ensuring that the search process is always within the feasible region and gradually converges to the optimal scheduling scheme that satisfies space compensation and energy constraints.

[0027] S6: Decode the iteration results that meet the termination conditions into production instructions and logistics energy replenishment strategies, and control the PCB assembly line to complete task allocation and energy scheduling.

[0028] It should be noted that the termination condition refers to the criteria for determining when the algorithm stops iterating, including reaching the maximum number of iterations or the convergence of the objective function; the iteration result refers to the optimal honey source individual obtained when the algorithm terminates, which includes the optimal gene sequence of the four-segment encoding; the production instruction refers to the control command that guides each workstation on the PCB assembly line to execute a specific process sequence; the logistics energy replenishment strategy refers to the charging timing and path planning scheme formulated based on the power status of the AGV vehicle and the task requirements; and decoding refers to the process of mapping the four-segment chromosome into an executable scheduling scheme.

[0029] Understandably, when the algorithm meets the termination condition, it transforms the four-segment code of the optimal honey source individual into an actual executable production control instruction, decodes the task assignment matrix into the process execution sequence of each workstation, decodes the generalized task time matrix into a logistics scheduling timetable, and decodes the vehicle assignment offset segment into AGV task allocation and charging strategy. This realizes the control mapping from the mathematical optimization model to the physical production system, and completes the closed-loop control of task allocation and energy scheduling of the PCB assembly line.

[0030] Preferably, calculating the equivalent dwell time corresponding to the processing time of the process and the forced process waiting time, and transforming the timing constraint of process waiting into the physical space occupancy constraint of materials on the workstation buffer includes: Define task At the workstation Equivalent dwell time of the buffer The equivalent residence time Satisfying the relation: ; in, Indicate process At the workstation Equivalent dwell time, Indicates completion of process The required standard physical processing time, Indicate process The corresponding process requires a minimum waiting time; Calculate any time interval workstation Dynamic space usage The dynamic space occupancy Satisfying the relation: ; in, Indicates time workstation Dynamic space usage, Indicates at time The set of tasks that are still in an equivalent resident state and satisfy the following conditions: ,in Indicate process The start time of processing Indicate process The number of space units occupied by the required materials within the workstation buffer zone; Through the dynamic space occupancy Spatial exclusion checks are performed on subsequent logistics loading operations to ensure the workstation The spatial compensation flexible constraint is satisfied at all times, and the spatial compensation flexible constraint satisfies the following relationship: ; in, Indicates workstation Maximum physical space limit for storing materials.

[0031] It should be noted that the equivalent length of stay It is the process At the workstation Standard physical processing time Minimum waiting time required by the process The integrated time parameters obtained by fusion through addition operations, among which Obtained through process databases or time studies, characterizing pure operation time. Determined through process specifications or physicochemical experiments, the forced settling time, including curing and cooling, is used to characterize the time. The sum of these two times yields the total time from when the material enters the workstation until it can leave; dynamic space occupancy. At any time workstation The total volume of all materials still in an equivalent residence state is determined by traversing all processes currently assigned to this workstation to determine the time. Is it at the start of each process? and the end time The accumulation object is determined between these parameters; the processing start time is then determined. Determined through production scheduling systems or by recursively calculating the completion times of preceding processes; task set. Indicates at time The set of all process indexes that satisfy the above interval conditions; number of spatial units. Obtained through Bill of Materials (BOM) analysis and packaging specification standardization, it characterizes the physical volume occupied by materials in a single process within the buffer zone; maximum physical space limit. Based on a comprehensive assessment of the physical dimensions of the workstation buffer zone, shelf capacity, and safe operating space; space exclusion verification refers to the process of determining the space exclusion of the workstation within the logistics system. Before sending new materials, pre-calculate the current dynamic space occupancy. With the volume of new materials The sum of the terms, if the sum is greater than 1 / 2. If this occurs, the interception mechanism will be triggered, preventing the material from entering and feeding back to the scheduling system to reschedule the delivery or adjust the process sequence.

[0032] It is understandable that by defining equivalent dwell time This transforms the process waiting constraints, which were originally time-based, into continuous occupancy constraints in the physical space dimension, allowing the scheduling system to dynamically manage space occupancy during the decoding phase. The system calculates and predicts the buffer accumulation status in real time. When a potential overflow risk is detected, it blocks subsequent feeding through spatial exclusion check, thereby avoiding the buffer overflow risk caused by the traditional process scheduling ignoring the continuous occupation of materials during the process waiting period. This achieves integrated modeling and real-time control of timing logic and physical space occupation.

[0033] Preferably, the mathematical model is constructed with the objectives of minimizing the production load smoothing index and the total penalty cost, including: Calculate the production smoothing index to quantify the deviation between the workload and production cycle time of each workstation. The production smoothing index Satisfying the relation: ; in, Indicates the production smoothing index. This indicates the total number of workstations. This indicates the preset production cycle time. Indicates allocation to the first A set of procedures performed by a workstation. Indicate process At the workstation The equivalent dwell time; Constraints based on time, space, and energy dimensions: total penalty cost The total penalty cost Satisfying the relation: ; in, Indicates the total cost of punishment. Indicates workstation Space violation penalties Indicates logistics task Time-limited penalties Indicates automated guided vehicle Energy security penalties This indicates the total number of workstations. Indicates the total number of processes. Indicates the total number of automated guided vehicles; Among them, the space violation penalty Calculations based on the aforementioned spatial compensation constraints satisfy the following relationship: ; in, Indicates penalties for violations of space regulations. Indicates the spatial violation coefficient. Indicates time workstation Dynamic space usage, Indicates workstation Maximum physical space limit for storing materials; Logistics timeliness penalty obtained by evaluating logistics delivery time deviation The aforementioned logistics timeliness penalty Satisfying the relation: ; in, Indicates logistics task Time-limited penalties This indicates a penalty factor for early delivery. Indicates the penalty factor for delayed delivery. Indicates logistics task Expected delivery time Indicates logistics task The actual completion time; Among them, the energy safety penalty Satisfying the relation: ; in, Indicates energy security penalty, Indicates the energy loss penalty factor. Represents a symbolic function. This indicates the power threshold that triggers forced charging. This indicates the current remaining battery power of the automated guided vehicle. This indicates the time loss caused by the charging process. By adaptively balancing production smoothness and violation costs, a comprehensive optimization objective function is constructed. The comprehensive optimization objective function Satisfying the relation: ; in, This represents the comprehensive optimization objective function value. and These represent the adaptive weighting coefficients used to dynamically balance production smoothness and total penalty cost, respectively. Indicates the production smoothing index. This represents the total cost of the penalty.

[0034] It should be noted that the production smoothing index It quantifies the workload of each workstation and the preset production cycle time. The root mean square index of the deviation between them, where This represents the total number of workstations on the production line, determined by the production line layout configuration. This represents the preset production cycle time, determined through capacity planning and order demand calculations. Indicates allocation to the first The set of procedures executed by each workstation is determined by the Task Assignment Matrix (TAM). Indicate process At the workstation The equivalent dwell time, through and The total penalty cost is obtained by summing the results. Penalties for violations of space regulations Logistics timeliness penalty and energy safety penalties It consists of three superimposed parts, among which The total number of processes is determined through production order parsing. The total number of Automated Guided Vehicles (AGVs) is determined through the logistics system configuration; penalties for space violations. By monitoring dynamic space occupancy Exceeding the maximum physical space limit The spatial violation coefficient is obtained by integrating over the time dimension. The penalty is set based on the output loss rate caused by spatial overflow per unit time, and is calibrated using historical production stoppage loss data. Its value reflects the severity of the penalty for the overflow; logistics timeliness penalty. By comparing logistics tasks Expected delivery time With actual completion time The deviation calculation, where Determined by working backward from the production cycle and material requirements planning. Early delivery penalty factor is obtained through calculation using a generalized task time matrix. Based on the spatial risk coefficient caused by premature material accumulation, the workstation space occupancy rate and process priority factor are used to set the parameters. Comprehensive calculation of late delivery penalty factor Based on the loss of output per unit time caused by work stoppages and material shortages, the total profit is set. With total production time The ratio is then multiplied by the workstation criticality coefficient. Calculation; Energy security penalty Through symbolic functions Monitoring Automated Guided Vehicle Current remaining battery power Is the battery level below the threshold that triggers forced charging? ,in Data is collected in real time through the AGV on-board battery management system. The rated capacity is set according to battery characteristics and safe operation requirements. A certain proportion, energy loss penalty factor The cost is set based on the time loss caused by the charging action, through charging time. Correlation coefficient with output per unit time calculate, Determined based on actual measurements of charging pile power and battery capacity; adaptive weighting coefficient. and The optimization ratio used to dynamically balance production smoothness and total penalty cost is calculated in the current population. and Standard deviation and Obtain, among which The calculation formula is for each nectar source in the population. The square root of the average of the squared deviations of the values ​​from the mean. Indicates population size, Indicates the first The production smoothness index of each honey source It represents the population average production smoothing index.

[0035] Understandably, by constructing a dual-objective optimization model that integrates the production smoothing index and multi-dimensional penalty costs, the algorithm can proactively quantify and penalize violations such as logistics time deviations, buffer space overflows, and AGV energy depletion while pursuing a balanced distribution of operation time across workstations. This guides the search direction towards a feasible region that simultaneously satisfies cycle balance and physical constraints. Through an adaptive weighting mechanism, it dynamically responds to the emphasis on efficiency and constraints at different production stages, avoiding load imbalances or constraint violations caused by single-objective optimization, and ensuring that the scheduling scheme takes into account both production line efficiency and physical feasibility.

[0036] Preferably, the initial nectar source population is constructed using a four-segment chromosome structure, including task priority, workstation segmentation, task summary, and vehicle assignment offset. Construct a chromosome coding structure consisting of four gene loci, wherein the chromosome coding structure satisfies the following relation: ; in, Indicates the individual nectar source. This represents a task priority segment arranged by task process index. This represents a workstation segmentation control segment, composed of binary or integer components, used to identify workstation cutting points. This section represents the workstation task summary segment, which records the number of tasks assigned to each workstation. This indicates the vehicle assignment and logistics offset segment, which includes the vehicle index and logistics response time correction value. Initialize the task priority segment By random generation The process steps are arranged and their validity is verified using topological sorting logic to ensure the priority of the tasks. Satisfying the preceding process constraints between processes, among which Indicates the total number of processes; The workstation segmentation control section is initialized using randomly generated cutting points. And according to the workstation segmented control section For the task priority segment Divide the tasks and update the workstation task summary section synchronously. To clarify the task attribution relationships of each workstation; For each logistics interaction task, a corresponding vehicle index and its logistics offset relative to the production completion time are randomly assigned. And fill the vehicle assignment and logistics offset segment. This completes the construction of the initial nectar source population.

[0037] It should be noted that the nectar source individuals It is the basic search unit in the improved artificial bee colony algorithm, consisting of four gene loci linked together; task priority segment Use a length of The integer permutation encoding stores the topologically ordered sequence of PCB assembly tasks, where This section represents the total number of processes, determined by the production order, and decides the logical order in which processes enter the production line; workstation segment control section. use A set of incrementing positive integer codes, where This indicates the total number of workstations, determined by the production line layout, and is used to identify the cutting points of the task sequence between workstations; the workstation task summary section. Use a length of Integer vector encoding to record the number of tasks assigned to each workstation. ,pass The cutting point is calculated; vehicle assignment and logistics offset segment. Use a length of The dual-layer real number encoding, where the first half stores the vehicle index. By using the available AGV set Obtained by random selection from, This indicates the total number of AGVs, with the latter half storing the logistics offset. By using a preset range Obtained by random generation within. Based on production cycle time and logistics response time; Topological sorting logic is a validity verification mechanism based on a directed acyclic graph (DAG), which ensures the validity of processes by checking the precedence and succession relationships between them. The arrangement does not violate the physical process sequence of patching, soldering, and testing.

[0038] Understandably, by constructing a four-segment chromosome structure, decision-making information from four dimensions—task priority ranking, workstation boundary division, task scale statistics, and vehicle assignment offset—is integrated into a single bee colony individual. This allows a single code to fully represent all decision-making content of the PCB assembly line's integrated scheduling, ensuring that the coding scheme covers the three-dimensional optimization requirements of task-space-energy. The topology sorting validity check ensures the process feasibility of the task sequence, the linkage update of workstation segmentation and task aggregation clarifies workstation allocation, and the random initialization of vehicle assignment and offset achieves diversified logistics scheduling. Thus, a structured initial search population is provided for improving the artificial bee colony algorithm.

[0039] Preferably, the process of using the workstation segmentation to truncate the task priority sequence to construct a task assignment matrix includes: Based on the workstation segmented control section The determined cut point information will be used to prioritize the task. The process index sequence is sequentially assigned to the corresponding workstations; Establish a production task assignment matrix The production task assignment matrix Satisfying the relation: ; in, This represents the production task assignment matrix. This indicates the decision variable for process assignment, when the process... Cut off and assigned to workstation hour, The value is 1 if it is not 0 otherwise. Indicates the total number of workstations; Verify the production task assignment matrix Check if the task capacity limit of each workstation is met. If not, adjust the workstation segment control section. The gene values ​​are re-trunculated and reassigned.

[0040] It should be noted that the cleavage site information refers to the information stored in... In A strictly increasing positive integer, used to indicate in The sequence is segmented at specific points to form task subsets for each workstation; the truncation operation divides a continuous sequence of process indices into segments according to the cutting points, with each segment corresponding sequentially to a task set for a workstation; the Production Task Assignment Matrix (TAM) is a... OK A binary decision matrix of columns, where row indices correspond to processes. The column index corresponds to the workstation. ,element A value of 1 indicates a process. Assigned to workstation Execution, a value of 0 indicates unallocated space, and this matrix is ​​obtained by traversing... Sequence and according to The cutting result is obtained by assignment; the task capacity limit refers to the upper limit of each workstation based on equipment processing capacity, personnel configuration, or production cycle time. The maximum total equivalent dwell time or the maximum number of tasks allowed is determined by the production line process quota.

[0041] Understandably, by truncating the task priority sequence based on the cutting points of the workstation segmented control sections, a clear allocation of processes among physical workstations is achieved. A binary task assignment matrix is ​​constructed, transforming abstract coding into a concrete workstation allocation scheme. The allocation result is ensured not to exceed the workstation's capacity boundary by verifying the task capacity limit; if this is not met, adjustments are made. The gene values ​​are redistributed, thereby establishing a reliable mapping from the coding space to the physical workstation allocation during the decoding stage, providing a clear task assignment basis for subsequent space occupancy calculations and energy scheduling.

[0042] Preferably, generating a generalized task time matrix reflecting the state of task time nodes by combining the vehicle assignment offset and spatial compensation constraints includes: Based on the production task assignment matrix The established task hierarchy, combined with the vehicle assignment and logistics offset segment. Logistics offset in and the equivalent dwell time of each process. Construct a generalized task time matrix The generalized task time matrix Satisfying the relation: ; in, Indicate process The generalized task time matrix, Indicates the actual completed processes of the logistics vehicle. The delivery time, Indicate process At the workstation The equivalent dwell time; Wherein, the actual completion time of the logistics task Satisfying the relation: ; in, Indicates the actual completed processes of the logistics vehicle. The delivery time, Indicate process The start time of processing Indicate process At the workstation Equivalent dwell time, This indicates the vehicle assignment and logistics offset segment. The logistics offset recorded in the middle, Indicates the assigned automated guided vehicle The estimated transport time required to execute this route segment.

[0043] It should be noted that the generalized task time matrix It is A vector structure used to record processes. The time attribute status, where the first element is the actual completed process of the logistics vehicle. Delivery task time The second element is the process. At the workstation Equivalent stay time Task dependency refers to the process determined by the Production Task Assignment Matrix (TAM). With workstation The allocation correspondence between them; logistics offset It is the vehicle assignment and logistics offset section The time correction value recorded in the system is obtained by randomly generating it during the initialization phase and optimizing it during the evolution process. It is used to adjust the actual delivery time of the AGV and estimate the transportation time. Automated Guided Vehicles Drive from the current location to the target workstation. The time required to complete loading and unloading operations is calculated by the path planning module of the AGV scheduling system. This calculation is based on the production line logistics topology distance, the AGV's rated operating speed, and the fixed loading and unloading time; the start time of processing. The production scheduling system determines the start time of the production line for the first process and the start time of subsequent processes for the processes that have been completed and transferred by AGV.

[0044] Understandably, by combining the task subordination relationship determined by the task assignment matrix, the logistics offset of the vehicle assignment offset segment is extracted, and the equivalent dwell time is integrated to construct a generalized task time matrix. This enables the coordinated representation of workstation allocation, logistics scheduling, and process time, allowing the decoding process to accurately calculate the actual delivery time of materials and simultaneously record the total occupation time of the process at the workstation. This provides an accurate time reference for subsequent dynamic calculation of space occupation and energy status assessment, ensuring the time sequence alignment of production and logistics.

[0045] Preferably, the improved artificial bee colony algorithm is used to search through the local mutation of foraging bees, the global development of observation bees, and the random reset mechanism of scout bees, and the optimal scheduling scheme is updated using a mask-based crossover operator while preserving the process logic, including the following steps: S10: Foraging bees are at the current nectar source Within the neighborhood, for task priority segments Perform genetic operations based on the crossover operator and simultaneously adjust the segmented control sections of the workstation. With task summary section ; S11: Using a greedy selection mechanism to compare the comprehensive objective function of offspring nectar sources and original nectar sources. If the offspring is better, replace it; otherwise, retain the original nectar source and accumulate the stagnation count of that nectar source. S12: The observation bee determines the nectar source to be mined based on the fitness value of the nectar source through a probabilistic selection mechanism. The selected high-quality nectar source is subjected to the same crossover, mutation and greedy selection strategies as the foraging bee stage, so as to achieve in-depth mining of the excellent area in the search space and update the global optimal solution. S13: Real-time monitoring of the stagnation count of each nectar source; when the stagnation count of a certain nectar source reaches a preset threshold... If the nectar source is found to be in a local optimum, the scout bee will discard it and randomly generate a new nectar source in the search space according to the process topology constraints to replace it. Repeat steps S10-S13 until the maximum number of iterations is reached. Or the objective function converges.

[0046] It should be noted that the improved artificial bee colony algorithm is a hybrid intelligent optimization algorithm that introduces genetic operators and a greedy selection mechanism on the basis of the traditional artificial bee colony algorithm. The foraging bee stage is the part of the algorithm responsible for local search within the neighborhood of their respective associated nectar sources, generating new solutions by performing genetic operations based on crossover operators. The observation bee stage is the part of the algorithm that performs probabilistic selection based on the fitness value of nectar sources and performs in-depth mining on high-quality nectar sources, determining the target to be mined through a roulette wheel selection mechanism. The scout bee stage is the part of the algorithm responsible for maintaining population diversity and escaping local optima. When the stagnant count of a nectar source reaches a preset threshold, it is discarded and a new nectar source is randomly generated. Local mutation refers to a search method that generates small perturbations in the vicinity of a single nectar source through operations such as swapping and insertion. Global development refers to a strategy of systematically and deeply exploring high-quality areas within the search space. Random reset refers to an alternative mechanism that randomly generates entirely new solutions within the search space based on process topological constraints. The greedy selection mechanism refers to the comprehensive objective function that compares the offspring nectar sources with the original nectar sources. The selection strategy involves retaining the best solution; the stagnation count is a counter that records the number of times a single honey source has not improved in consecutive iterations, incrementing by 1 each time the original solution is retained and resetting to zero when it is replaced; a preset threshold is used. The maximum number of iterations is set based on the population size and problem complexity, typically 1.5 to 2.5 times the population size. The algorithm is set according to the production line scheduling response time requirements to ensure that it converges within the allowable range of industrial real-time performance. The global optimal solution refers to the honey source individual with the smallest comprehensive objective function value found during the algorithm iteration process.

[0047] It is understandable that by having foraging bees perform genetic operations based on crossover operators in the neighborhood of the current nectar source to achieve local fine-grained search, by having observation bees select high-quality nectar sources based on fitness probabilities for global development to achieve in-depth mining of excellent areas, and by having scout bees monitor stagnation and count and randomly reset nectar sources when a threshold is reached to escape the local optimum trap, the three types of bee colonies work together to form a dynamic balance between local search and global exploration. The greedy selection mechanism ensures that the quality of the solution is continuously improved, and finally, the optimal scheduling scheme that satisfies space compensation and energy constraints is obtained through iterative convergence.

[0048] Preferably, the crossover operator is a PPX crossover operator, and the execution of the PPX crossover operator includes the following steps: S20: Select parental nectar source and And create a random binary mask vector of the same length as the task sequence. ; S21: Check bit by bit The One element: If Then from Extract the leftmost unselected task from the current generation sequence and add it to the child sequence. Delete the task; S22: If Then from Extract the leftmost unselected task from the current generation sequence and add it to the child sequence. Delete the task; Repeat steps S20-S22 until the child sequence is filled. Use a mask to guide the child to automatically inherit the order of the process logic from the parent to avoid generating illegal solutions.

[0049] It should be noted that the PPX crossover operator is a priority-preserving genetic operation suitable for process-ordering encoding. Its core function is to automatically preserve the process topology constraints between processes when two parent honey sources undergo gene recombination, avoiding the generation of illegal solutions that violate the precedence-successor logic; random binary mask vector. It is a length of The vector, where Equal to the total number of tasks Each element is independently and randomly assigned a value of 1 or 2, used to determine which parent generation to extract tasks from; the leftmost unselected task refers to the first process encountered in the current parent generation sequence that has not yet been extracted into the child generation sequence; the deletion operation refers to removing the extracted tasks from the temporary copy of the parent generation sequence to ensure that subsequent extractions always target the remaining task set; illegal solution refers to the sequence arrangement that violates the process precedence constraint, such as placing the reflow soldering process that must be performed after chip placement before chip placement.

[0050] Understandably, by creating a random binary mask vector to guide the task extraction process of the two parent honey sources, the child sequence automatically inherits the order of the parent sequence that satisfies the process logic during the construction process. Since each extraction is of the leftmost legal task in the current parent sequence, and the deletion operation ensures that the task is not selected repeatedly, this mechanism naturally preserves the topological constraint relationship between processes, avoids generating illegal solutions that violate the process logic, and thus improves the search efficiency and feasibility of the solution in complex constraint environments.

[0051] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0052] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A comprehensive scheduling method for PCB assembly lines considering space compensation and energy constraints, characterized in that, include: Calculate the equivalent dwell time corresponding to the processing time of the process and the forced process waiting time, transform the timing constraint of process waiting into the physical space occupation constraint of materials on the workstation buffer, and construct space compensation constraint; A mathematical model is constructed with the goal of minimizing the production load smoothing index and the total penalty cost, wherein the total penalty cost includes logistics timeliness penalty, space violation penalty based on the spatial compensation constraint, and energy safety penalty based on the vehicle's remaining power. An initial nectar source population was constructed using a four-segment chromosome structure, including task priority, workstation segmentation, task summary, and vehicle assignment offset. The task priority sequence is truncated using the workstation segmentation to construct a task assignment matrix, and a generalized task time matrix reflecting the task time node status is generated by combining the vehicle assignment offset and spatial compensation constraints. An improved artificial bee colony algorithm is used to search through the local mutation of foraging bees, the global development of observation bees, and the random reset mechanism of scout bees. The optimal scheduling scheme is updated by using a mask-based crossover operator while preserving the process logic. The iteration results that meet the termination conditions are decoded into production instructions and logistics replenishment strategies, and the PCB assembly line is controlled to complete task allocation and energy scheduling.

2. The PCB assembly line integrated scheduling method considering space compensation and energy constraints according to claim 1, characterized in that, The equivalent dwell time corresponding to the processing time of the operation and the forced process waiting time is calculated, and the timing constraints of process waiting are transformed into physical space occupancy constraints of materials on the workstation buffer, including: Define task At the workstation Equivalent dwell time of the buffer The equivalent residence time Satisfying the relation: ； in, Indicate process At the workstation Equivalent dwell time, Indicates completion of process The required standard physical processing time, Indicate process The corresponding process requires a minimum waiting time; Calculate any time interval workstation Dynamic space usage The dynamic space occupancy Satisfying the relation: ； in, Indicates time workstation Dynamic space usage, Indicates at time The set of tasks that are still in an equivalent resident state and satisfy the following conditions: ,in Indicate process The start time of processing Indicate process The number of space units occupied by the required materials within the workstation buffer zone; Through the dynamic space occupancy Spatial exclusion checks are performed on subsequent logistics loading operations to ensure the workstation The spatial compensation flexible constraint is satisfied at all times, and the spatial compensation flexible constraint satisfies the following relationship: ； in, Indicates workstation Maximum physical space limit for storing materials.

3. The PCB assembly line integrated scheduling method considering space compensation and energy constraints according to claim 1, characterized in that, The mathematical model constructed with the objectives of minimizing the production load smoothing metric and the total penalty cost includes: Calculate the production smoothing index to quantify the deviation between the workload and production cycle time of each workstation. The production smoothing index Satisfying the relation: ； in, Indicates the production smoothing index. This indicates the total number of workstations. This indicates the preset production cycle time. Indicates allocation to the first A set of procedures performed by a workstation. Indicate process At the workstation The equivalent dwell time; Constraints based on time, space, and energy dimensions: total penalty cost The total penalty cost Satisfying the relation: ； in, Indicates the total cost of punishment. Indicates workstation Space violation penalties Indicates logistics task Time-limited penalties Indicates automated guided vehicle Energy security penalties This indicates the total number of workstations. Indicates the total number of processes. Indicates the total number of automated guided vehicles; Among them, the space violation penalty Calculations based on the aforementioned spatial compensation constraints satisfy the following relationship: ； in, Indicates penalties for violations of space regulations. Indicates the spatial violation coefficient. Indicates time workstation Dynamic space usage, Indicates workstation Maximum physical space limit for storing materials; Logistics timeliness penalty obtained by evaluating logistics delivery time deviation The aforementioned logistics timeliness penalty Satisfying the relation: ； in, Indicates logistics task Time-limited penalties This indicates a penalty factor for early delivery. Indicates the penalty factor for delayed delivery. Indicates logistics task Expected delivery time Indicates logistics task The actual completion time; Among them, the energy safety penalty Satisfying the relation: ； in, Indicates energy security penalty, Indicates the energy loss penalty factor. Represents a symbolic function. This indicates the power threshold that triggers forced charging. This indicates the current remaining battery power of the automated guided vehicle. This indicates the time loss caused by the charging process. By adaptively balancing production smoothness and violation costs, a comprehensive optimization objective function is constructed. The comprehensive optimization objective function Satisfying the relation: ； in, This represents the comprehensive optimization objective function value. and These represent the adaptive weighting coefficients used to dynamically balance production smoothness and total penalty cost, respectively. Indicates the production smoothing index. This represents the total cost of the penalty.

4. The PCB assembly line integrated scheduling method considering space compensation and energy constraints according to claim 1, characterized in that, The initial nectar source population was constructed using a four-segment chromosome structure, including task priority, workstation segmentation, task summary, and vehicle assignment offset. Construct a chromosome coding structure consisting of four gene loci, wherein the chromosome coding structure satisfies the following relation: ； in, Indicates the individual nectar source. This represents a task priority segment arranged by task process index. This represents a workstation segmentation control segment, composed of binary or integer components, used to identify workstation cutting points. This section represents the workstation task summary segment, which records the number of tasks assigned to each workstation. This indicates the vehicle assignment and logistics offset segment, which includes the vehicle index and logistics response time correction value. Initialize the task priority segment By random generation The process steps are arranged and their validity is verified using topological sorting logic to ensure the priority of the tasks. Satisfying the preceding process constraints between processes, among which Indicates the total number of processes; The workstation segmentation control section is initialized using randomly generated cutting points. And according to the workstation segmented control section For the task priority segment Divide the tasks and update the workstation task summary section synchronously. To clarify the task attribution relationships of each workstation; For each logistics interaction task, a corresponding vehicle index and its logistics offset relative to the production completion time are randomly assigned. And fill the vehicle assignment and logistics offset segment. This completes the construction of the initial nectar source population.

5. The PCB assembly line integrated scheduling method considering space compensation and energy constraints according to claim 4, characterized in that, Using the workstation segmentation to truncate the task priority sequence to construct the task assignment matrix includes: Based on the workstation segmented control section The determined cut point information will be used to prioritize the task. The process index sequence is sequentially assigned to the corresponding workstations; Establish a production task assignment matrix The production task assignment matrix Satisfying the relation: ； in, This represents the production task assignment matrix. This indicates the decision variable for process assignment, when the process... Cut off and assigned to workstation hour, The value is 1 if it is not 0 otherwise. Indicates the total number of workstations; Verify the production task assignment matrix Check if the task capacity limit of each workstation is met. If not, adjust the workstation segment control section. The gene values ​​are re-trunculated and reassigned.

6. The PCB assembly line integrated scheduling method considering space compensation and energy constraints according to claim 5, characterized in that, The generalized mission time matrix, which combines the vehicle assignment offset and spatial compensation constraints to generate a state at mission time points, includes: Based on the production task assignment matrix The established task hierarchy, combined with the vehicle assignment and logistics offset segment. Logistics offset in and the equivalent dwell time of each process. Construct a generalized task time matrix The generalized task time matrix Satisfying the relation: ； in, Indicate process The generalized task time matrix, Indicates the actual completed processes of the logistics vehicle. The delivery time, Indicate process At the workstation The equivalent dwell time; Wherein, the actual completion time of the logistics task Satisfying the relation: ； in, Indicates the actual completed processes of the logistics vehicle. The delivery time, Indicate process The start time of processing Indicate process At the workstation Equivalent dwell time, This indicates the vehicle assignment and logistics offset segment. The logistics offset recorded in the middle, Indicates the assigned automated guided vehicle The estimated transport time required to execute this route segment.

7. The PCB assembly line integrated scheduling method considering space compensation and energy constraints according to claim 6, characterized in that, An improved artificial bee colony algorithm is used to search for the optimal scheduling scheme by employing local mutations in foraging bees, global expansion by observation bees, and random reset mechanisms by scout bees. The algorithm then utilizes a mask-based crossover operator to update the optimal scheduling scheme while preserving the process logic. The steps include: S10: Foraging bees are at the current nectar source Within the neighborhood, for task priority segments Perform genetic operations based on the crossover operator and simultaneously adjust the segmented control sections of the workstation. With task summary section ; S11: Using a greedy selection mechanism to compare the comprehensive objective function of offspring nectar sources and original nectar sources. If the offspring is better, replace it; otherwise, retain the original nectar source and accumulate the stagnation count of that nectar source. S12: The observation bee determines the nectar source to be mined based on the fitness value of the nectar source through a probabilistic selection mechanism. The selected high-quality nectar source is subjected to the same crossover, mutation and greedy selection strategies as the foraging bee stage, so as to achieve in-depth mining of the excellent area in the search space and update the global optimal solution. S13: Real-time monitoring of the stagnation count of each nectar source; when the stagnation count of a certain nectar source reaches a preset threshold... If the nectar source is found to be in a local optimum, the scout bee will discard it and randomly generate a new nectar source in the search space according to the process topology constraints to replace it. Repeat steps S10-S13 until the maximum number of iterations is reached. Or the objective function converges.

8. The PCB assembly line integrated scheduling method considering space compensation and energy constraints according to claim 7, characterized in that, The crossover operator is the PPX crossover operator, and the execution of the PPX crossover operator includes the following steps: S20: Select parental nectar source and And create a random binary mask vector of the same length as the task sequence. ; S21: Check bit by bit The One element: If Then from Extract the leftmost unselected task from the current generation sequence and add it to the child sequence. Delete the task; S22: If Then from Extract the leftmost unselected task from the current generation sequence and add it to the child sequence. Delete the task; Repeat steps S20-S22 until the child sequence is filled. Use a mask to guide the child to automatically inherit the order of the process logic from the parent to avoid generating illegal solutions.

Images