A multi-objective factory scheduling method considering machine constraints

Abstract

The present invention discloses a multi-objective factory scheduling method considering machine restrictions, which relates to the technical field of production scheduling. The method includes: establishing a machine work database; connecting the monitoring sensor of the machine, establishing the real-time state of the machine based on the monitoring sensor, extracting restriction features, and establishing static restrictions of the machine; reading the order requirements of the processing control system, and configuring the node work coordination of the machine node, performing scheduling fitting optimization, and establishing scheduling fitting optimization results; performing machine task allocation based on the scheduling fitting optimization results, and establishing real-time machine operation data, performing machine dynamic restriction trigger analysis to generate iterative scheduling, and performing machine scheduling management based on iterative scheduling. The method solves the technical problem that it is difficult to flexibly adjust the multi-objective scheduling according to the machine operation status in the prior art, which leads to poor scheduling adaptability, low production efficiency and resource utilization, and achieves the technical effect of improving production efficiency and resource utilization.

Description

Technical Field

[0001] The present application relates to the technical field related to production scheduling, and in particular to a multi-objective factory scheduling method considering machine limitations. Background Art

[0002] In modern manufacturing, factory scheduling is a vital task. Factory production scheduling optimization is a key link to achieve efficient production, reduce costs and improve resource utilization. With the expansion of production scale and the complexity of machine equipment, the traditional single-objective scheduling method can no longer meet the needs of modern factories. Especially considering the limitations of machines, how to reasonably and efficiently arrange production tasks has become an urgent problem to be solved. Existing factory scheduling methods usually only allocate tasks based on static order requirements and equipment capacity, lack of sufficient consideration of the dynamic state of the machine and real-time operation data. In actual applications, it is easy to encounter problems such as unbalanced production resource utilization, excessive or low machine load, and increased equipment failure rate, resulting in low production efficiency and even the risk of delayed delivery. In addition, since different orders have different priorities, delivery times and batch requirements, a single scheduling strategy is difficult to flexibly respond to the coordination needs of multiple production goals.

[0003] In the current multi-objective factory scheduling related technologies, there is a technical problem that it is difficult to flexibly adjust the multi-objective scheduling according to the machine operation status, which leads to poor scheduling adaptability, low production efficiency and resource utilization. Summary of the invention

[0004] This application solves the technical problem in the prior art that it is difficult to flexibly adjust the multi-objective scheduling according to the machine operating status, which leads to poor scheduling adaptability, low production efficiency and resource utilization, by providing a multi-objective factory scheduling method that takes machine limitations into consideration, thereby achieving the technical effect of improving production efficiency and resource utilization.

[0005] The present application provides a multi-objective factory scheduling method that takes machine limitations into consideration, including: establishing a machine work database, wherein the machine work database is constructed by collecting historical operation data of the machine, wherein the historical operation data includes processing task data, production capacity data, energy consumption data, and processing monitoring data; connecting the monitoring sensor of the machine, establishing the real-time status of the machine based on the monitoring sensor, extracting restriction features based on the real-time status of the machine and the machine work database, and establishing static restrictions on the machine, wherein the static restrictions on the machine include production capacity restrictions, continuous operation restrictions, energy consumption restrictions, and load fluctuation restrictions; reading the order requirements of the processing control system, and configuring the node work coordination of the machine nodes, performing scheduling fitting optimization based on the node work coordination, order requirements, and machine static restrictions, and establishing a scheduling fitting optimization result; performing machine task allocation based on the scheduling fitting optimization result, and establishing real-time machine operation data, performing machine dynamic restriction trigger analysis based on the real-time machine operation data, generating an iterative schedule based on the trigger analysis result, and performing machine scheduling management based on the iterative schedule.

[0006] In a possible implementation, the scheduling fitting optimization is performed based on the node work coordination, order requirements and machine static restrictions, and the scheduling fitting optimization result is established. The following processing is also performed: the optimization evaluation function is configured as follows:

[0007] ,

[0008] Among them, S represents the comprehensive score, represents the node work collaboration score, n is the total number of collaborative tasks, i represents any collaborative task node, represents the waiting time of the i-th collaborative task node before connecting to the next node, represents the total completion time of the i-th collaborative task node, represents the order demand adaptation, m represents the total number of orders, j represents any order, , , are the weights of priority, delivery time, and batch size, respectively. represents the priority adaptation score of the jth order, Characterizes the delivery time adaptation score of the jth order, represents the batch adaptation score of the j-th order, Characterize the machine static limit satisfaction score, , , The weight factors of node work coordination, order demand adaptation and machine static limitation are respectively characterized; the optimization evaluation function is used to perform scheduling fitting optimization and establish the scheduling fitting optimization result.

[0009] In a possible implementation, the restriction feature extraction is performed based on the real-time status of the machine and the machine working database, and the static restriction of the machine is established. The following processing is also performed: data analysis is performed on the machine working database to establish a capacity data set, and the capacity data set is provided with a load mapping identifier; dynamic factors are established using the machine working database, and step load capacity restrictions are established based on the denoised capacity mean and dynamic factors of the capacity data set; applicable matching of the step load capacity restrictions is performed based on the real-time status of the machine, and capacity restrictions are established based on the applicable matching results, and the static restriction of the machine is constructed using the capacity restrictions.

[0010] In a possible implementation, before establishing the step load capacity limit based on the denoised capacity mean and dynamic factors of the capacity data set, the following processing is also performed: the load is divided into steps to establish dynamic step load weights; when calculating the denoised capacity mean, step denoising is performed by matching the dynamic step load weights with the step levels of the denoised capacity mean; and the denoised capacity mean mapped to the step levels is established based on the step denoising processing results.

[0011] In a possible implementation, the method performs a machine dynamic limit trigger analysis based on the real-time machine operation data, generates an iterative schedule based on the trigger analysis results, and also performs the following processing: establish machine dynamic limits, the machine dynamic limits include production capacity trigger limits, continuous operation trigger limits, load fluctuation trigger limits, and energy consumption trigger limits; utilize machine dynamic limits to perform a trigger analysis on the real-time machine operation data; perform task retention authentication based on the trigger analysis results, and generate an iterative schedule based on the task retention authentication results.

[0012] In a possible implementation, the task retention authentication based on the trigger analysis result also performs the following processing: determining whether the trigger analysis result is a trigger for forced machine replacement; if the trigger analysis result is not a trigger for forced machine replacement, calling the machine's backtracking work data to establish a first retention factor; performing retention adjustment fitting based on the trigger analysis result to establish a second retention factor; performing machine replacement fitting using the trigger analysis result to establish a third retention factor; and completing the task retention authentication based on the first retention factor, the second retention factor, and the third retention factor.

[0013] In a possible implementation, the trigger analysis result is used to perform machine replacement fitting, establish a third holding factor, and further perform the following processing: obtain the number of start-stop machines under the machine replacement state, and establish a start-stop impact feature; obtain the task execution impact under the machine replacement fitting, and establish a task impact feature; perform a balance analysis of machine tasks based on the machine replacement fitting, and establish a balance impact feature; and establish a third holding factor based on the start-stop impact feature, the task impact feature, and the balance impact feature.

[0014] In a possible implementation, the multi-objective factory scheduling method considering machine limitations also performs the following processing: recording the factory production score corresponding to the iterative scheduling; establishing scheduling feedback based on the factory production score, reconstructing the machine static limitations with the scheduling feedback, and performing scheduling management based on the reconstructed machine static limitations.

[0015] It is proposed to establish a machine work database through the multi-objective factory scheduling method considering machine restrictions proposed in this application; connect the monitoring sensors of the machine, establish the real-time status of the machine based on the monitoring sensors, extract restriction features, and establish static restrictions on the machine; read the order requirements of the processing control system, and configure the node work coordination of the machine nodes, perform scheduling fitting optimization, and establish scheduling fitting optimization results; allocate machine tasks based on the scheduling fitting optimization results, and establish real-time machine operation data, perform machine dynamic restriction trigger analysis to generate iterative scheduling, and perform machine scheduling management based on iterative scheduling. It solves the technical problem that it is difficult to flexibly adjust multi-objective scheduling according to the machine operation status in the prior art, which leads to poor scheduling adaptability, low production efficiency and resource utilization, and achieves the technical effect of improving production efficiency and resource utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0017] Figure 1 A flowchart of a multi-objective factory scheduling method considering machine constraints provided in an embodiment of the present application;

[0018] Figure 2 A schematic diagram of a flow chart of establishing static constraints on machines in a multi-objective factory scheduling method considering machine constraints provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0022] The present application embodiment provides a multi-objective factory scheduling method that takes into account machine limitations, such as Figure 1 As shown, the method includes:

[0023] Step S100, establishing a machine work database, wherein the machine work database is constructed by collecting historical operation data of the machine, wherein the historical operation data includes processing task data, production capacity data, energy consumption data, and processing monitoring data.

[0024] Preferably, historical operation data of the machine is collected to build a machine work database, wherein the historical operation data includes processing task data, production capacity data, energy consumption data, and processing monitoring data. Specifically, the processing task data includes task type (the type of task performed by the machine, such as cutting, welding, assembly, etc.), task duration (the actual running time of each task, recording the time taken for the task to be completed from the start to the completion), task frequency (the execution frequency of different types of tasks, used to analyze the impact of tasks on the load of the machine) and completion rate (the completion status of the task, marking whether it is completed on time and in quantity, and whether delays or interruptions occur). The production capacity data includes the capacity per unit time (the production volume of each machine per unit time), peak capacity (the maximum capacity that the machine can reach under high load), and average capacity (the long-term average capacity based on historical data). , used to plan the normal workload of the machine) and capacity change trends (changes in capacity over time, task type or load); energy consumption data include energy consumption per unit time (the amount of energy consumed by the machine per unit time, such as electricity, gas, etc.), task energy consumption (energy consumption when performing a specific task), peak energy consumption (energy consumption of the machine under maximum load) and energy efficiency (energy consumption corresponding to unit output, reflecting the energy efficiency level of the machine); processing monitoring data include temperature monitoring (temperature data of the machine during operation, monitoring temperature changes to avoid overheating), vibration monitoring (recording the vibration of the machine to help identify possible mechanical failures), operating status records (recording the start, stop, pause and other states of the machine) and fault records (the number, type and duration of machine failures, including detailed records of maintenance and repairs).

[0025] Step S200, connect the monitoring sensor of the machine, establish the real-time status of the machine based on the monitoring sensor, extract restriction features based on the real-time status of the machine and the machine working database, and establish static restrictions on the machine. The static restrictions on the machine include production capacity restrictions, continuous working restrictions, energy consumption restrictions, and load fluctuation restrictions.

[0026] Preferably, the real-time status of the machine (real-time operation data) is obtained by connecting to the monitoring sensor of the machine, and the restriction feature is extracted in combination with the historical operation data of the machine in the machine work database, so as to establish the static restriction of the machine, that is, to establish the capacity restriction, continuous operation restriction, energy consumption restriction, and load fluctuation restriction. Specifically, the capacity restriction refers to the maximum production volume or workload allowed by the machine in a unit time to prevent the machine from overloading. Based on the historical capacity data in the machine work database, including the peak capacity and the average capacity, an upper capacity limit is set. By monitoring the real-time output of the machine, it is ensured that the actual production volume does not exceed the set capacity limit, so as to avoid the machine running for too long under high load state, and to prevent equipment wear, precision reduction or failure due to excessive capacity output; the continuous operation limit refers to the maximum time that the machine can run continuously without stopping or cooling. Based on the continuous working limit of the machine in the historical data Keep a record of the duration, and set the upper limit of continuous working time in combination with the equipment instructions, monitor the working time of the machine in real time, ensure that the machine stops for rest or reduces the load after reaching the upper limit of continuous working time, reduce machine overheating, equipment loss or mechanical failure caused by long-term continuous operation, and extend the service life of the equipment; energy consumption limit refers to the maximum energy consumption allowed by the machine in a unit time to control the energy consumption in the production process. By analyzing historical energy consumption data, the average energy consumption and peak energy consumption under different tasks and loads are determined, and the energy consumption upper limit is set. The power consumption data of the machine is monitored in real time. When the real-time energy consumption approaches or exceeds the set energy consumption upper limit, task adjustments or load distribution are carried out to optimize energy use and avoid energy waste and increased production costs due to excessive energy consumption; load fluctuation limit refers to the allowable range of load changes of the machine in a short period of time. According to the fluctuations in historical load data, the allowable load fluctuation range is set, such as the average load range of ±10%. The change frequency and amplitude of the machine load are monitored in real time to ensure that the load fluctuation remains within the set range, prevent the increase of mechanical stress of equipment components due to excessive load fluctuations, reduce the decline in production efficiency caused by frequent load adjustments, and ensure a smooth and efficient production process.

[0027] Further, such as Figure 2 As shown, step S200 also includes step S210, performing data analysis on the machine work database to establish a capacity data set, wherein the capacity data set is provided with a load mapping identifier; step S220, establishing a dynamic factor using the machine work database, and establishing a step load capacity limit based on the denoised capacity mean and the dynamic factor of the capacity data set; step S230, performing applicable matching of the step load capacity limit based on the real-time status of the machine, establishing a capacity limit based on the applicable matching result, and using the capacity limit to construct a static limit of the machine.

[0028] Preferably, the historical operation data in the machine work database is parsed, and the capacity-related information is extracted therefrom to establish a capacity data set. Specifically, the unit time capacity of the machine under different load states is extracted to reflect the production capacity of the machine under a specific load, as well as the historical capacity data of different tasks and different time periods, and the corresponding relationship between the task type and the capacity is recorded to establish a capacity data set. Then, according to the capacity performance of the machine under different load states, a load mapping identifier is set for the data in the capacity data set to indicate the load states corresponding to different capacity data, such as low load, medium load and high load; dynamic factors are set based on the historical operation data of the machine, taking into account capacity changes, load fluctuations and task types, to reflect the machine The dynamic factor is the coefficient used to adjust the capacity limit of the machine under different working conditions. The data in the capacity data set is denoised to eliminate outliers and volatility noise, and the denoised capacity mean representing the machine in a stable state is calculated. Then, according to the load mapping identifier of the capacity data, the denoised capacity mean is divided into multiple steps (such as low load, medium load and high load) according to the load level. Each step represents the capacity limit under different load states. The denoised capacity mean is combined with the dynamic factor to obtain the capacity limit under different load steps. For example, the low / medium / high load capacity limits are calculated according to the low / medium / high load states of the denoised mean combined with the dynamic factor, that is, the step load capacity limit.

[0029] Preferably, the real-time load status of the machine is obtained through sensors to determine the current load level (low load, medium load or high load). According to the current load level of the machine, the corresponding step load capacity limit is matched. For example, when the machine is in a medium load state, the medium load capacity limit is selected, and then the capacity limit value matched with the real-time state is established to establish the capacity limit under the current production conditions to ensure that the machine will not be overloaded. On the basis of the capacity limit, combined with other static constraints (such as continuous working time limit, energy consumption limit and load fluctuation limit), a complete static limit of the machine is formed as a constraint condition for multi-objective factory scheduling to ensure that the scheduling result is within the equipment capacity range, so that the machine can operate efficiently and extend the equipment life.

[0030] Furthermore, step S220 also includes step S221, dividing the load into steps and establishing dynamic step load weights; step S222, when calculating the denoised capacity mean, performing step denoising processing by matching the dynamic step load weights with the step levels corresponding to the denoised capacity mean; and step S223, establishing the denoised capacity mean mapped with the step levels based on the step denoising processing results.

[0031] Preferably, the load is divided into multiple steps according to different load conditions of the machine, such as low load step, medium load step and high load step, and a dynamic weight is set for each load step, that is, the dynamic step load weight, which is used to reflect the influence of the load on the denoised capacity mean. Faced with different step levels, the applicability of each step data is different, so multiple denoised capacity means are calculated to more accurately reflect the capacity performance of the machine under various load conditions, wherein the step level applicability indicates the weight of the capacity data at this level in the denoising calculation, reflects the credibility of each load step data, and calculates the denoised capacity. When averaging, the load ladder division and the corresponding dynamic weights are used to perform weighted denoising according to the level of the load ladder to obtain a more accurate denoised capacity mean. Specifically, the denoised capacity data is divided according to the load level, matched with the corresponding dynamic step load weights, and the denoised data is weighted averaged according to the weight of each load ladder to obtain a denoised capacity mean that can better reflect the actual capacity of the machine; the denoised capacity mean obtained by the step denoising process is mapped to each load ladder to form the denoised capacity mean under different load conditions, which more accurately reflects the capacity of the machine under different load conditions and is used for subsequent capacity limitation and scheduling optimization.

[0032] Step S300, read the order requirements of the processing control system, and configure the node work coordination of the machine node, perform scheduling fitting optimization based on the node work coordination, order requirements and machine static constraints, and establish a scheduling fitting optimization result.

[0033] Preferably, detailed requirements of all current orders are extracted from the processing control system. Order requirements refer to production requirements for each order, which may specifically include priority (relative importance of the order), delivery time (order deadline completion time, used to ensure on-time delivery), batch requirements (production quantity required by the order), task flow (processing steps and process flow required for the order, which are convenient for allocation to each machine node) and product type, and then the node work coordination of the machine node is configured, wherein the machine node refers to the equipment or equipment group that performs a specific task in the factory, and each machine node has a specific processing capacity and load bearing capacity. Specifically, node work coordination refers to the work connection and task allocation relationship between each machine node, including the priority of tasks and resource sharing, for example, the task connection sequence defines the execution order of different tasks in the order, ensuring that the downstream task is started after the upstream task is completed; configuring the resource sharing and collaboration mode between adjacent nodes to ensure the smoothness of the workflow; reasonably distributing the load between multiple machine nodes to avoid overloading a single node and ensure that the load of all nodes is relatively balanced; by configuring the node work coordination relationship, ensure that multiple machines are coordinated when performing tasks, improve work efficiency, and reduce waiting time and resource waste.

[0034] Preferably, the scheduling is optimized based on the node work coordination, order requirements and machine static restrictions, that is, according to the order requirements, node work coordination relationship and machine static restrictions (such as capacity restrictions, continuous work restrictions, energy consumption restrictions, load fluctuation restrictions), the optimal task allocation and execution sequence are found to maximize production efficiency, improve resource utilization and complete all orders on time. Specifically, the static restrictions of the machine (such as capacity, energy consumption, etc.) and order requirements are used as constraints to ensure that the scheduling plan is feasible under all conditions. The objective function of the optimization is set, which usually includes multi-objective optimization, such as minimizing the completion time. In order to maximize equipment utilization, minimize energy consumption, etc., optimization algorithms (such as genetic algorithm, simulated annealing algorithm, particle swarm algorithm, etc.) are used to solve the scheduling problem, find the optimal scheduling solution that meets the order requirements and machine limitations, and adjust the optimization results according to real-time data during the scheduling process to ensure that the scheduling is still valid when production changes. Finally, a scheduling fitting optimization result is established, which includes the task allocation, execution sequence and time arrangement of all machine nodes, to ensure that all orders are completed under the premise of ensuring equipment load balance, while improving production efficiency, saving energy, and ensuring the long-term stable operation of machine equipment and on-time delivery of orders.

[0035] Furthermore, step S300 also includes step S310, configuring an optimization evaluation function as follows:

[0036] ,

[0037] Among them, S represents the comprehensive score, represents the node work collaboration score, n is the total number of collaborative tasks, i represents any collaborative task node, represents the waiting time of the i-th collaborative task node before connecting to the next node, represents the total completion time of the i-th collaborative task node, represents the order demand adaptation, m represents the total number of orders, j represents any order, , , are the weights of priority, delivery time, and batch size, respectively. represents the priority adaptation score of the jth order, Characterizes the delivery time adaptation score of the jth order, represents the batch adaptation score of the j-th order, Characterize the machine static limit satisfaction score, , , The weight factors respectively represent the node work coordination, order demand adaptation, and machine static limitation; step S320, using the optimization evaluation function to perform scheduling fitting optimization and establish the scheduling fitting optimization result.

[0038] Preferably, the optimization evaluation function is configured as follows:

[0039] ,

[0040] Among them, S represents the comprehensive score, represents the node work collaboration score, n is the total number of collaborative tasks, i represents any collaborative task node, represents the waiting time of the i-th collaborative task node before connecting to the next node, represents the total completion time of the i-th collaborative task node, represents the order demand adaptation, m represents the total number of orders, j represents any order, , , are the weights of priority, delivery time, and batch size, respectively. represents the priority adaptation score of the jth order, Characterizes the delivery time adaptation score of the jth order, represents the batch adaptation score of the j-th order, Characterize the machine static limit satisfaction score, , , The weight factors of node work coordination, order demand adaptation and machine static restriction are respectively characterized, and the optimization evaluation function and reverse scheduling fitting are used for optimization. Specifically, the evaluation function is used to evaluate multiple scheduling schemes to find the scheme with the highest total score, that is, the optimal task allocation and execution order, including generating multiple initial scheduling schemes as candidate schemes according to order requirements, machine node coordination relationships and static restrictions, scoring each candidate scheme using the optimization evaluation function, calculating the comprehensive performance of each scheme in terms of node coordination, order demand adaptation and static restriction satisfaction, selecting an optimization algorithm (such as genetic algorithm, simulated annealing algorithm or particle swarm algorithm) for solving, and gradually finding the scheme with the highest comprehensive score through iterative improvement of the scheme. The scheme is continuously iterated and updated during the optimization process, and the score is recalculated after each update until the optimal solution is found or the preset optimization stop condition is reached, and finally the scheduling fitting optimization result is obtained to ensure production efficiency and equipment utilization, and meet the needs of order delivery and resource balance.

[0041] Step S400, based on the scheduling fitting optimization result, machine task allocation is performed, and real-time machine operation data is established. According to the real-time machine operation data, machine dynamic limit trigger analysis is performed, and iterative scheduling is generated based on the trigger analysis result. Machine scheduling management is performed based on the iterative scheduling.

[0042] Preferably, specific processing tasks are assigned to different machine nodes based on the scheduling fitting optimization results, and the start time, end time and execution order are set for each task. Specifically, task assignment information is generated, including generating a task list for each machine node, clarifying the type of task, required time and task priority, considering the load balance of the machine when assigning tasks, avoiding too many or too few tasks for some machines, ensuring that the load of the machine is evenly distributed, setting start and stop conditions for each task, such as completing the previous task, reaching a predetermined production volume or meeting a specific operating status, and the machine automatically performs production operations according to the order and time of the assigned tasks; by connecting to the monitoring sensors on the machine, the operating data of the machine during the execution of the task is collected in real time, including real-time production capacity data, energy consumption data, load data and operating status data.

[0043] Preferably, based on the real-time machine operation data, a dynamic limit trigger analysis of the machine is performed, that is, it is determined whether the current machine has reached the preset limit conditions. If so, the corresponding adjustment mechanism is triggered. Specifically, the dynamic limit trigger conditions include capacity limit trigger (real-time output exceeds the set capacity limit), continuous work limit trigger (the continuous working time of the machine reaches the set upper limit), energy consumption limit trigger (real-time power consumption exceeds the energy consumption upper limit) and load fluctuation limit trigger (load fluctuation exceeds the allowable range). If a certain limit is triggered, it will be determined whether the current task needs to be paused, switched or adjusted according to the limit type, and then an iterative schedule is generated, that is, according to the trigger analysis results, the current scheduling plan is adjusted to ensure that production continues and meets various limit conditions. Specifically, if a machine needs to be paused or load reduced due to the triggering limit, For machines with high priority orders, resources are allocated first to ensure that their production will not be affected by adjustments to other tasks. For machines that are restricted due to continuous working time or load fluctuations, they are temporarily shut down for cooling or recovery. During the cooling period, task allocation is readjusted to continue to follow the optimization goals, such as minimizing completion time, improving equipment utilization, and reducing energy consumption, to ensure that production efficiency is still maximized after dynamic adjustment, thereby generating an iterative schedule and performing machine scheduling management, that is, real-time monitoring of the machine's execution of the new iterative schedule to ensure that the task proceeds smoothly according to the new plan. At the same time, feedback data, including task completion rate, equipment status, load changes, etc., is collected to gradually optimize the scheduling plan, improve the factory's production flexibility and resource utilization efficiency, and ensure efficient completion of production tasks.

[0044] Furthermore, step S400 also includes step S410, establishing machine dynamic restrictions, wherein the machine dynamic restrictions include production capacity trigger restrictions, continuous work trigger restrictions, load fluctuation trigger restrictions, and energy consumption trigger restrictions; step S420, using machine dynamic restrictions to perform trigger analysis on real-time machine operation data; step S430, performing task retention authentication based on the trigger analysis results, and generating an iterative schedule based on the task retention authentication results.

[0045] Preferably, dynamic restrictions on the machine are established, that is, a set of dynamic thresholds are set for the real-time status of the machine for monitoring and triggering analysis during the production process, specifically including capacity trigger restrictions, continuous work trigger restrictions, load fluctuation trigger restrictions and energy consumption trigger restrictions, wherein the capacity trigger restriction refers to a restriction condition that is triggered when the output of the machine reaches a set upper limit within a certain period of time, and is set based on historical capacity data and current scheduling requirements; the continuous work trigger restriction refers to a restriction condition that is triggered when the continuous operation time of the machine reaches a preset upper limit, and is set according to the design specifications, maintenance recommendations and historical data of the machine; the load fluctuation trigger restriction refers to a restriction condition that is triggered when the load fluctuation amplitude or frequency of the machine exceeds a set range, and a reasonable fluctuation range is set based on historical load fluctuations and equipment stability requirements (such as a load change amplitude not exceeding 10%); the energy consumption trigger restriction refers to a restriction condition that is triggered when the power consumption of the machine reaches a set upper limit, and is set based on historical energy consumption data and the design power consumption of the machine.

[0046] Preferably, the real-time operation data of the machine is obtained through sensors, including production capacity, working hours, load fluctuations and power consumption, and the dynamic limits of the machine are used to trigger analysis of the real-time machine operation data, that is, the real-time data is compared with the dynamic limits of the machine to determine whether the current state meets the trigger conditions. For example, if the current production capacity reaches or exceeds the production capacity trigger limit, the production capacity limit analysis is triggered; if the continuous working time of the machine reaches the set upper limit, the continuous working limit analysis is triggered; if the load fluctuation amplitude exceeds the limit range, the load fluctuation limit analysis is triggered; if the power consumption reaches the energy consumption trigger limit, the energy consumption limit analysis is triggered; and the trigger analysis results are classified and sorted according to the type of trigger.

[0047] Preferably, task retention authentication is performed based on the trigger analysis results. Specifically, based on the trigger analysis results, it is evaluated whether the current task can be continued or needs to be adjusted, that is, it is determined whether the current task meets the machine constraints. If the capacity trigger limit is triggered, it is evaluated whether the task intensity can be reduced (such as reducing the task frequency or output) to keep the current task running; if the continuous working time trigger limit is triggered, it is evaluated whether the machine needs to be paused or the task needs to be switched to ensure that the equipment is not overheated or overused; if the load fluctuation trigger limit is triggered, it is evaluated whether the task priority or task type can be adjusted to reduce load fluctuations and continue the task; if the energy consumption trigger limit is triggered, it is evaluated whether the power consumption can be reduced (such as reducing high-energy consumption tasks or optimizing equipment operating parameters) to keep the task running; finally, the task allocation and time arrangement are dynamically adjusted according to the task retention authentication results to meet the dynamic constraints of the machine, and then an iterative schedule is generated to achieve intelligent dynamic management of factory scheduling.

[0048] Furthermore, step S430 also includes step S431, determining whether the trigger analysis result is a trigger for forced machine replacement; step S432, if the trigger analysis result is not a trigger for forced machine replacement, calling the machine's backtracking work data to establish a first retention factor; step S433, performing retention adjustment fitting based on the trigger analysis result to establish a second retention factor; step S434, performing machine replacement fitting using the trigger analysis result to establish a third retention factor; step S435, completing task retention certification according to the first retention factor, the second retention factor, and the third retention factor.

[0049] Preferably, it is determined whether the trigger analysis result is a trigger for forced machine replacement. Specifically, when the real-time status of the machine reaches a certain irreversible restriction condition, the condition for forced machine replacement is triggered. For example, the machine cannot continue to run due to a serious failure, a sharp rise in temperature, or excessive energy consumption. In the trigger analysis, the trigger type is checked to determine whether the current trigger condition requires that the task of the current machine must be stopped and transferred to other machines. If the trigger analysis result is not a trigger for forced machine replacement, that is, the current task can be continued after adjustment and the machine does not need to be replaced immediately, the historical work data of the machine is called to extract the data such as the machine's production capacity, energy consumption, and load fluctuation under similar historical conditions, analyze its performance under similar load, working time, and energy consumption conditions, and determine the first retention factor for evaluating whether the machine can continue to keep the current task running under the current trigger condition.

[0050] Preferably, based on the current trigger conditions, a maintenance adjustment fit is performed, such as reducing the task frequency, reducing the output or reducing the load fluctuation range, to simulate the impact of these adjustments on the machine operation, and establish a second maintenance factor to evaluate whether the current machine can continue to maintain the task after adjusting the task parameters; then use the current trigger analysis results to evaluate the possibility of transferring the current task to other machines, compare the loads and constraints of different machines, find alternative machines that can better adapt to the current task requirements, and then generate a third maintenance factor to measure the feasibility and efficiency of continuing the task after the machine is replaced.

[0051] Preferably, by comprehensively analyzing the three retention factors, it is determined whether the current task can continue to run on the original machine, or whether the task needs to be adjusted or the machine needs to be replaced. Specifically, if the comprehensive result of the first retention factor and the second retention factor shows that the current machine can continue to perform the task, the task is kept unchanged; if the second retention factor (maintaining adjustment fit) shows that appropriate adjustment of the task can extend the operating capacity of the current machine, the task parameters are adjusted to ensure that the task continues; if the third retention factor shows that replacing the machine is the best solution, the current task is transferred to other machines; finally, based on the result of the task retention certification, it is determined whether a new iterative schedule needs to be generated.

[0052] Furthermore, step S434 also includes step S4341, obtaining the number of started and stopped machines under the machine replacement status, and establishing the start-stop impact characteristics; step S4342, obtaining the task execution impact under the machine replacement fitting, and establishing the task impact characteristics; step S4343, performing a balance analysis of the machine tasks based on the machine replacement fitting, and establishing the balance impact characteristics; step S4344, establishing a third retention factor according to the start-stop impact characteristics, the task impact characteristics, and the balance impact characteristics.

[0053] Preferably, when a task requires the replacement of a machine, the number of started and stopped machines in the replacement state is obtained, wherein, if the task needs to be replaced to another machine, the original machine needs to be shut down for maintenance, cooling or adjustment, and the machine required for the new task may include idle equipment. The machine is restarted, and the start-stop impact characteristics are established, which specifically refers to recording the time cost of restarting and shutting down, recording the increased power consumption and energy cost due to the start and stop of the machine, etc.; the task execution impact under the machine replacement fitting is obtained, for example, replacing the machine may cause the task startup time to be extended, and the production capacity or load limit of the new machine after the machine is replaced affects the completion speed of the task, and the quality performance of some tasks on different machines may be different. Based on this, task impact characteristics are established, including execution speed, quality assurance and order delay risk, which are used to analyze the actual impact of machine replacement on task completion.

[0054] Preferably, a balance analysis is performed on the machine tasks based on the machine replacement fitting, and the load distribution of the production line after the task replacement is evaluated to ensure that the distribution of production tasks among the machines is as balanced as possible. Specifically, the load status of each machine after the replacement is analyzed to avoid overloading of some machines while other machines are too idle; the impact of machine replacement on resource utilization is evaluated to ensure that there is no waste of resources after the replacement; and then a balance impact feature is established, including load balance (difference in load distribution after machine replacement) and equipment utilization change (the degree of increase or decrease in equipment utilization caused by machine replacement); finally, a third retention factor is established based on the start-stop impact feature, task impact feature, and balance impact feature to judge the rationality of task replacement of the machine. For example, the weight values ​​of the start-stop impact feature, task impact feature, and balance impact feature are comprehensively calculated, and the final third retention factor is obtained by weighted calculation, which indicates whether it is suitable to replace the machine under the current conditions, minimizes the start-stop cost, optimizes task execution, and ensures the balance of load distribution when replacing the machine, thereby improving the overall efficiency and resource utilization of the production line.

[0055] Furthermore, step S400 also includes step S440, recording the factory production score corresponding to the iterative schedule; step S450, establishing scheduling feedback based on the factory production score, reconstructing the machine static constraints with the scheduling feedback, and performing scheduling management based on the reconstructed machine static constraints.

[0056] Preferably, the iterative scheduling scheme is evaluated, that is, the production efficiency, resource utilization, energy consumption, output quality, etc. are comprehensively considered to quantitatively evaluate the advantages and disadvantages of the iterative scheduling scheme, and the factory production score is calculated and recorded based on the actual production data. Then, by analyzing the factory production score, the advantages and disadvantages of the current scheduling scheme are found out, and this information is fed back for subsequent scheduling optimization. Specifically, the scores of different iterative scheduling schemes are compared to find out the reasons for the score changes, such as which static constraints affect production efficiency or quality, identify the effective factors in the scheduling schemes with higher scores as a reference for subsequent scheduling optimization, and at the same time identify the factors with lower scores for adjustment.

[0057] Preferably, suggestions for scheduling improvement are made based on the feedback information, such as adjusting certain static restriction parameters or improving resource allocation strategies to improve the production score; static restriction parameters of the machine are reset based on the generated feedback information, for example, if the scheduling feedback shows that the capacity limit is too low, resulting in a decrease in production efficiency, the capacity limit can be appropriately relaxed to improve the machine utilization rate; the upper limit of continuous working time is adjusted based on the feedback to ensure that the continuous production time is increased without damaging the life of the equipment to adapt to peak tasks; if the feedback shows that the energy consumption limit is too strict, resulting in a decrease in efficiency, the limit can be appropriately relaxed, or the energy consumption threshold can be increased according to the configuration of high-efficiency equipment; the load fluctuation limit is adjusted based on the feedback to ensure that the load adaptability of the machine is improved while meeting the product quality requirements; finally, scheduling management is performed based on the reconstructed static restrictions of the machine, that is, after the static restrictions are reconstructed, the scheduling plan is dynamically managed and optimized to adapt to the new machine restriction conditions, for example, on the basis of the reconstructed static restrictions, tasks are recalculated and assigned, task distribution is optimized, dynamic optimization and continuous improvement of scheduling are achieved, production efficiency and resource utilization are improved, and the best production effect is achieved.

[0058] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A multi-objective factory scheduling method considering machine constraints, characterized in that: The method comprises: Establishing a machine work database, wherein the machine work database is constructed by collecting historical operation data of the machine, wherein the historical operation data includes processing task data, production capacity data, energy consumption data, and processing monitoring data; Connect the monitoring sensor of the machine, establish the real-time status of the machine based on the monitoring sensor, extract the restriction features based on the real-time status of the machine and the machine working database, and establish the static restriction of the machine, wherein the static restriction of the machine includes capacity restriction, continuous working restriction, energy consumption restriction, and load fluctuation restriction; Read the order requirements of the processing control system, configure the node work coordination of the machine nodes, perform scheduling fitting optimization based on the node work coordination, order requirements and machine static restrictions, and establish the scheduling fitting optimization results; Based on the scheduling fitting optimization results, machine tasks are allocated and real-time machine operation data is established. Based on the real-time machine operation data, dynamic limit trigger analysis of the machine is performed. Based on the trigger analysis results, iterative scheduling is generated, and machine scheduling management is performed based on iterative scheduling. The scheduling fitting optimization is performed based on the node work coordination, order requirements and machine static restrictions, and the scheduling fitting optimization result is established, and further includes: Configure the optimization evaluation function as follows: ， Among them, S represents the comprehensive score, represents the node work collaboration score, n is the total number of collaborative tasks, i represents any collaborative task node, represents the waiting time of the i-th collaborative task node before connecting to the next node, represents the total completion time of the i-th collaborative task node, represents the order demand adaptation, m represents the total number of orders, j represents any order, , , are the weights of priority, delivery time, and batch size, respectively. represents the priority adaptation score of the jth order, Characterizes the delivery time adaptation score of the jth order, represents the batch adaptation score of the j-th order, Characterize the machine static limit satisfaction score, , , The weight factors that represent node work coordination, order demand adaptation, and machine static constraints respectively; Use the optimization evaluation function to perform scheduling fitting optimization and establish the scheduling fitting optimization results; The extracting of restriction features based on the real-time status of the machine and the machine working database to establish static restrictions on the machine also includes: Performing data analysis on the machine work database to establish a capacity data set, wherein the capacity data set is provided with a load mapping identifier; The dynamic factors are established using the machine work database, and the step load capacity limit is established based on the denoised capacity mean and dynamic factors of the capacity data set; Based on the real-time status of the machine, an applicable match is made to the step load capacity limit, and based on the applicable matching result, the capacity limit is established, and the static limit of the machine is constructed using the capacity limit.

2. The multi-objective factory scheduling method considering machine constraints as claimed in claim 1, characterized in that: Before establishing the step load capacity limit based on the denoised capacity mean and dynamic factors of the capacity data set, the method further includes: Divide the load into steps and establish dynamic step load weights; When calculating the denoised capacity mean, step denoising is performed by matching the denoised capacity mean to the step level corresponding to the dynamic step load weight; The denoised capacity mean of the step level mapping is established based on the step denoising processing results.

3. The multi-objective factory scheduling method considering machine constraints as claimed in claim 1, characterized in that: The method of performing a machine dynamic limit trigger analysis according to the real-time machine operation data and generating an iterative schedule based on the trigger analysis result also includes: Establishing dynamic limits for machines, including capacity trigger limits, continuous operation trigger limits, load fluctuation trigger limits, and energy consumption trigger limits; Use machine dynamic limits to trigger analysis of real-time machine operation data; Perform task retention authentication based on the trigger analysis result, and generate an iteration schedule based on the task retention authentication result.

4. The multi-objective factory scheduling method considering machine constraints as claimed in claim 3, characterized in that: The task holding authentication based on the trigger analysis result also includes: Determining whether the trigger analysis result is a trigger for forced machine replacement; If the trigger analysis result is not a trigger for forcibly replacing the machine, calling the backtracking work data of the machine to establish a first retention factor; Perform hold adjustment fitting based on the trigger analysis result to establish a second hold factor; Use the trigger analysis results to fit the machine change and establish the third hold factor; The task retention authentication is completed according to the first retention factor, the second retention factor, and the third retention factor.

5. The multi-objective factory scheduling method considering machine constraints as claimed in claim 4, characterized in that: The method of using the trigger analysis result to perform machine replacement fitting and establish the third holding factor also includes: Obtain the number of machines started and stopped during machine replacement and establish start-stop impact characteristics; Obtain the impact of task execution under machine replacement fitting and establish task impact characteristics; Perform balance analysis of machine tasks based on machine replacement fitting and establish balance impact characteristics; The third retention factor is established according to the start-stop impact characteristics, task impact characteristics, and balance impact characteristics.

6. The multi-objective factory scheduling method considering machine constraints as claimed in claim 1, characterized in that: The method further comprises: Record the factory production scores corresponding to the iterative scheduling; Establish scheduling feedback based on factory production scores, reconstruct machine static constraints based on scheduling feedback, and perform scheduling management based on the reconstructed machine static constraints.

Images