Dynamic scheduling method for intelligent manufacturing system

Abstract

The invention discloses a dynamic scheduling method for an intelligent manufacturing system, and particularly relates to the technical field of intelligent scheduling. Through a scheduling performance evaluation mechanism fusing a task switching interference index and a target function offset index, a polynomial regression model is utilized to realize scheduling health score prediction, and scheduling performance abnormity is accurately identified; under the abnormal condition, the system can dynamically adjust and dispatch a target function structure and algorithm disturbance parameters according to different index sources, the capability of jumping out of local optimum is effectively enhanced, and the global optimization effect is improved; through a closed-loop scheduling execution and feedback mechanism, the robustness, the self-adaptability and the execution feasibility of the scheduling system are remarkably improved, it is guaranteed that high-quality tasks are completed in time, and the overall production efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent scheduling, and in particular to a dynamic scheduling method for an intelligent manufacturing system. Background Art

[0002] Dynamic scheduling in intelligent manufacturing systems refers to the process of automatically optimizing and adjusting production tasks within an intelligent manufacturing environment based on real-time data and changes in production resources (such as equipment status, order changes, and process adjustments). By integrating technologies such as the Internet of Things, artificial intelligence, and big data analytics, it enables flexible response and efficient collaboration within the manufacturing process, thereby improving production efficiency, resource utilization, and the ability to respond to emergencies.

[0003] The existing technology has the following shortcomings: In complex real-time multi-objective scheduling, heuristic or local search algorithms often struggle to break out of existing task sequencing patterns when the search space is high-dimensional. For example, on electronics assembly lines, to reduce line changeovers, the system constantly prioritizes similar products, trapping them in local optimal solutions. This leads to persistent delays in high-priority orders and severely impacts delivery dates. The root cause is that scheduling algorithms lack effective "breakout" mechanisms, such as the perturbation strategies used in simulated annealing or the high-intensity mutations used in genetic algorithms. These mechanisms make it difficult to globally optimize under multi-objective trade-offs, thus limiting the adaptability and robustness of the scheduling system. Summary of the Invention

[0004] The purpose of the present invention is to provide a dynamic scheduling method for an intelligent manufacturing system to solve the deficiencies in the background technology.

[0005] In order to achieve the above-mentioned object, the present invention provides the following technical solution: a dynamic scheduling method for an intelligent manufacturing system, comprising: Collect real-time status data and task information in manufacturing systems; The task switching interference index is generated by analyzing the resource disturbance degree caused by frequent task switching, and the objective function offset index is generated by analyzing the dynamic deviation degree between the current scheduling result and the theoretical global optimal. Inputting the task switching interference index and the objective function offset index into a pre-built scheduling performance anomaly prediction model, the output of the model being a scheduling health score value; Comparing the scheduling health score with a preset threshold, if the scheduling health score is lower than the preset threshold, determining that the scheduling has performance anomalies; For abnormal scheduling performance, the corresponding scheduling optimization strategy is dynamically selected, including whether to introduce a task switching penalty term into the scheduling objective function based on the task switching interference index, and adjusting the disturbance parameters in the scheduling algorithm based on the objective function offset index; Execute the optimized scheduling algorithm, output the final scheduling plan, and send the scheduling results to the manufacturing execution system for production task execution.

[0006] Preferably, the status data includes equipment operating status, task progress, resource occupancy rate and material distribution status; the task information includes order parameters, process path, delivery requirements and personnel scheduling information.

[0007] Preferably, the task switching interference index is generated by: For each resource r∈R, where R represents the set of all resources participating in the current scheduling process, a fixed-length sliding time window W is set, with a unit time interval of Δt; the task execution sequence of the resource within the window is recorded as: ;in Indicates the task number that the resource performs in the i-th time slice, and n represents the total time slice; Traverse and record whether adjacent tasks have changed: ; Among them, the cumulative number of switching times ; Calculate the comprehensive switching intensity , the expression is: ; Defining resource disturbance weights , calculate the switching interference index of a single resource, the expression is: ;in is the maximum perturbation weight in the window, is the switching interference index of a single resource; the interference index of all resources is summarized to form the overall task switching interference index, which is expressed as: Where, is the task switching interference index.

[0008] Preferably, the objective function offset index is generated as follows: Set the optimization goal of the scheduling system to m, and define ; Where: x represents a scheduling solution; Represents the value of the i-th objective function; constructs the reference Pareto frontier set , the expression is: ; z is the total number of objective functions; calculate the minimum distance between the current solution and the Pareto frontier, for the current scheduling solution , and its objective function value is , calculate its distance to each point in the Pareto frontier: Where, Represents Euclidean distance; take the minimum distance As the offset distance; after normalizing the offset distance, the objective function offset index is obtained , the expression is: ;in: The upper limit of the reference distance.

[0009] Preferably, the task switching interference index and the objective function deviation index are input into a pre-built scheduling performance anomaly prediction model, and the output of the model is a scheduling health score value, including: The task switching interference index and the objective function offset index are converted into comprehensive feature vectors, which are used as inputs of the machine learning model. The machine learning model uses the scheduling health score value label predicted by each group of comprehensive feature vectors as the prediction target, and minimizes the sum of the prediction errors of all scheduling health score value labels as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The scheduling health score value is determined based on the model output results. The machine learning model is a polynomial regression model.

[0010] Preferably, comparing the scheduling health score with a preset threshold, and if the scheduling health score is lower than the preset threshold, determining that the scheduling has a performance abnormality, includes: The obtained scheduling health score value is compared with the preset score threshold. If the scheduling health score value is greater than or equal to the score threshold, the scheduling performance is determined to be normal; if the scheduling health score value is less than the preset score threshold, the scheduling performance is determined to be abnormal.

[0011] Preferably, the optimization strategy for the task switching interference index TSDD includes: in the scheduling optimization process, modifying the objective function and introducing a task switching penalty term as an additional optimization target; the optimized objective function is expressed as: Where, is the original scheduling objective function; the task switching interference index calculated by the current TSDD system; α is the penalty weight coefficient, ,in is the reference threshold of TSDD, is the adjustment factor.

[0012] Preferably, the optimization strategy for the objective function shift index OFDR includes adjusting the temperature attenuation factor in the simulated annealing algorithm, which is expressed as: If the OFDR is greater than its reference threshold, the initial disturbance of the annealing process is increased: ; is the current temperature parameter, λ is the temperature attenuation factor, represents the default decay rate, is the adjustment coefficient, is the reference threshold of OFDR; Adjust the mutation probability in the genetic algorithm, the expression is: ; represents the current mutation probability, is the default mutation probability; is the adjustment factor; When OFDR is greater than its reference threshold, the mutation probability is automatically increased, and the ability of searching to escape the local optimum is enhanced; when OFDR is less than its reference threshold, the mutation probability gradually decreases.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: When OFDR is greater than its reference threshold, the mutation probability is automatically increased, and the ability of searching to escape the local optimum is enhanced; when OFDR is less than its reference threshold, the mutation probability gradually decreases.

[0014] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. This invention introduces two innovative parameters for real-time dynamic assessment of scheduling status: the task switching interference index and the objective function offset index. This effectively addresses the technical shortcomings of existing scheduling systems, which are prone to local optimality and lack stability control in complex multi-objective environments. Leveraging sliding window modeling, Pareto frontier distance metric, and machine learning prediction models, the system enables quantitative assessment of scheduling status and proactively identifies potential performance anomalies using the scheduling health score, significantly improving the adaptability and intelligence of the scheduling system.

[0015] 2. This invention dynamically selects an optimization strategy based on the source of scheduling anomalies. This not only improves scheduling stability by introducing a task switching penalty, but also adaptively adjusts search perturbation parameters for scheduling results that deviate from the optimal state, effectively improving the scheduling algorithm's ability to escape local optima, thereby achieving efficient, stable, and executable scheduling results. The overall solution establishes a closed-loop scheduling control system of "perception-assessment-optimization-execution-feedback," which is highly robust, real-time, and industrially applicable, and can be widely applied in various intelligent manufacturing scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0017] Figure 1 This is a mind map of the method of the present invention. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0019] For examples, see Figure 1 As shown, the dynamic scheduling method of an intelligent manufacturing system described in this embodiment includes: Collect real-time status data and task information in manufacturing systems; The task switching interference index is generated by analyzing the resource disturbance degree caused by frequent task switching, and the objective function offset index is generated by analyzing the dynamic deviation degree between the current scheduling result and the theoretical global optimal. Inputting the task switching interference index and the objective function offset index into a pre-built scheduling performance anomaly prediction model, the output of the model being a scheduling health score value; Comparing the scheduling health score with a preset threshold, if the scheduling health score is lower than the preset threshold, determining that the scheduling has performance anomalies; For abnormal scheduling performance, the corresponding scheduling optimization strategy is dynamically selected, including whether to introduce a task switching penalty term into the scheduling objective function based on the task switching interference index, and adjusting the disturbance parameters in the scheduling algorithm based on the objective function offset index; Execute the optimized scheduling algorithm, output the final scheduling plan, and send the scheduling results to the manufacturing execution system for production task execution.

[0020] In the present invention, the first step is to collect real-time status data and task information in the manufacturing system. This step is used to provide an accurate and comprehensive input data basis for the subsequent scheduling algorithm, ensuring the real-time and effectiveness of the scheduling calculation.

[0021] In this embodiment, the manufacturing system is a typical discrete multi-station assembly production line, involving multiple numerically controlled machine tools (CNC), industrial robots, automatic guided vehicles (AGVs), production process stations, and auxiliary resources.

[0022] To achieve accurate collection of real-time status and task information, this embodiment adopts the following specific technical means: Adopting the Industrial Internet of Things (IIoT) architecture, the equipment operating status is collected through smart sensors and data acquisition terminals (such as PLCs and edge gateways) deployed at key nodes of the production line. The status information includes but is not limited to: equipment operating status: such as "standby", "processing", "abnormal", and "maintenance"; current task number and task stage identifier; machine tool spindle load, current, voltage, temperature and other operating condition data; fault warning information (such as abnormal vibration frequency and insufficient lubricating oil pressure); all collected data is uploaded to the Manufacturing Execution System (MES) through the OPC UA protocol and synchronized through timestamps to ensure that the scheduling system receives highly consistent and low-latency status data.

[0023] The system dynamically calculates resource utilization based on the current task schedule and equipment availability, including: the scheduled usage time of a single device within a specified time window; the occupancy and idleness of process stations; the degree of match between worker scheduling status and skills; AGV path congestion and call frequency; resource utilization is updated in real time in matrix form for use by the scheduling engine to determine resource conflicts and optimize load balancing.

[0024] Order information is connected to the enterprise ERP system in real time. Specific parameters include: order number, customer level and delivery requirements; product specifications, BOM list and process path; urgency (priority) mark; order insertion and change records; ERP and MES systems are connected through an API interface, and order information is standardized and then input into the scheduling engine to form a task breakdown list.

[0025] The scheduling system, through linkage with MES data and combining RFID tracking, barcode scanning, workstation reporting and other means, monitors the execution progress of each task in real time, including: the process node where the current task is located; the stages of completed and pending tasks; the comparison of actual start and end times with planned times; delay identification and cause analysis (such as material shortages and downtime); task progress is dynamically presented in a Gantt chart structure to determine the risk of task delays and whether the rescheduling mechanism should be triggered.

[0026] All collected data are pre-processed and uniformly modeled through the industrial big data platform, using data standardization, redundant filtering, time synchronization and other processing methods, and are finally stored in JSON format or database table structure for scheduling system calls.

[0027] At the same time, to support algorithm efficiency, the system will cache core data items in an in-memory database (such as Redis) and set a data refresh cycle (for example, updating every 10 seconds) to ensure that the scheduling algorithm can obtain the latest system snapshot for calculation.

[0028] Through the above approach, this embodiment achieves high-precision collection and unified management of real-time status data and task information in the manufacturing system. It not only provides the basic data input required for scheduling, but also provides key support for subsequent anomaly identification, parameter calculation (such as TSDD and OFDR), and scheduling optimization, reflecting the high adaptability and practicality of the present invention in industrial intelligent scheduling scenarios.

[0029] The task switching interference index is generated by analyzing the degree of resource disturbance caused by frequent task switching. By establishing a sliding time window on each resource (such as machine tools, operators, robots), the frequency, intensity and complexity of task switching within the window are counted, and the degree of disturbance to system stability is calculated in combination with the sensitive weight of the resource, thus forming the task switching interference index.

[0030] The specific generation method is: For each resource r∈R, R represents the set of all resources participating in the current scheduling process, which includes multiple resource units. Set a fixed-length sliding time window W with a unit time interval of Δt; record the task execution sequence of the resource within the window as: ;in Indicates the task number that the resource performs in the i-th time slice, and n represents the total time slice; Traverse and record whether adjacent tasks have changed: ; Among them, the cumulative number of switching times ; Every task switch There may be different perturbation complexities, and the task switching intensity function is defined as: =β1*process switching level+β2*material replacement complexity+β3*equipment adjustment time; calculate comprehensive switching intensity , the expression is: ; Each type of resource has different tolerance for switching, and defines resource disturbance weights , such as: high-precision CNC machine tools: =1.0; Automatic handling robot: =0.6; Ordinary assembly station: =0.3; After analyzing the dynamic deviation between the current scheduling result and the theoretical global optimum, the objective function deviation index is generated. The generation method is: Calculate the handover interference index of a single resource. The expression is: ;in is the maximum perturbation weight in the window (used for normalization), is the switching interference index of a single resource; the interference index of all resources is summarized to form the overall task switching interference index, which is expressed as: Where, The interference index of task switching is 1. The closer the interference index is to 1, the worse the stability of the current scheduling is and the system needs to be adjusted.

[0031] After analyzing the dynamic deviation between the current scheduling result and the theoretical global optimum, the objective function deviation index is generated. The generation method is: Set the optimization objectives of the scheduling system to m (e.g. m=2, most commonly the total completion time Makespan and energy consumption Energy), and define ; Where: x represents a scheduling solution; Represents the value of the i-th objective function. Construct the reference Pareto frontier set , the expression is: ; z is the total number of objective functions; the set is obtained by: using evolutionary algorithms (such as NSGA-II and MOEA / D) to run in similar task scenarios and save non-dominated solutions; building based on the known "scheduling optimal solution" in historical data; and obtaining the ideal scheduling boundary in the simulation model.

[0032] Calculate the minimum distance between the current solution and the Pareto frontier. , and its objective function value is , calculate its distance to each point in the Pareto frontier: Where, Represents Euclidean distance; take the minimum distance as the offset distance; After normalizing the offset distance, the objective function offset index is obtained, which is expressed as: ;in: The upper limit of the reference distance can be the maximum possible objective function difference of the current problem or the maximum distance between two points in the Pareto boundary point set; OFDR represents the objective function deviation index, which ranges from [0,1]. The closer it is to 1, the more serious the deviation.

[0033] The task switching interference index and the objective function deviation index are input into a pre-built scheduling performance anomaly prediction model. The output of the model is a scheduling health score value, including: The task switching interference index and the objective function offset index are converted into comprehensive feature vectors, which are used as inputs of the machine learning model. The machine learning model uses the scheduling health score value label predicted by each group of comprehensive feature vectors as the prediction target, and minimizes the sum of the prediction errors of all scheduling health score value labels as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The scheduling health score value is determined based on the model output results. The machine learning model is a polynomial regression model.

[0034] Comparing the scheduling health score with a preset threshold, if the scheduling health score is lower than the preset threshold, determining that the scheduling has performance anomalies; Set a scoring threshold Hth∈(0,1) to determine the scheduling performance status. Compare the obtained scheduling health score with the preset scoring threshold. If the scheduling health score is greater than or equal to the scoring threshold, the scheduling performance is considered normal; if the scheduling health score is less than the preset scoring threshold, the scheduling performance is considered abnormal.

[0035] For abnormal scheduling performance, the corresponding scheduling optimization strategy is dynamically selected, including whether to introduce a task switching penalty term into the scheduling objective function based on the task switching interference index, and adjusting the disturbance parameters in the scheduling algorithm based on the objective function offset index; The optimization strategy for the task switching interference index TSDD is used to reduce the system disturbance caused by task switching and improve the stability of scheduling.

[0036] During the scheduling optimization process, the objective function is modified and the task switching penalty term is introduced as an additional optimization objective.

[0037] The optimized objective function is expressed as: Where, is the original scheduling objective function (such as total completion time, total delay time, etc.); TSDD is the task switching interference index calculated by the current system; α is the penalty weight coefficient, which is used to adjust the importance of TSDD in the objective function and can be adjusted dynamically.

[0038] If TSDD is greater than its reference threshold, α is appropriately increased to strengthen the stability guidance; it can be set ,in is the reference threshold of TSDD, is the adjustment factor.

[0039] The optimization strategy for the objective function shift index OFDR is used to enhance the scheduling algorithm's ability to escape from local optimality and improve the global search capability.

[0040] Dynamically adjust the disturbance control parameters in the scheduling algorithm, which is applicable to heuristic search algorithms (such as genetic algorithms, simulated annealing, particle swarm optimization, etc.).

[0041] Adjust the temperature attenuation factor in the simulated annealing algorithm. The expression is: If the OFDR is greater than its reference threshold, the initial disturbance of the annealing process is increased: ; is the current temperature parameter, λ is the temperature attenuation factor, Indicates the default decay rate (such as 0.95), is the adjustment coefficient, which determines the response speed of the disturbance intensity. is the reference threshold of OFDR.

[0042] Adjust the mutation probability in the genetic algorithm, the expression is: ; represents the current mutation probability, is the default mutation probability (such as 0.05); is the adjustment factor.

[0043] When OFDR is greater than its reference threshold, the system automatically increases the mutation probability and enhances the ability of searching to escape the local optimum; when OFDR is less than its reference threshold, the mutation probability can gradually decrease.

[0044] Execute the optimized scheduling algorithm, output the final scheduling plan, and send the scheduling results to the manufacturing execution system for production task execution.

[0045] Based on the updated scheduling objective function or algorithm perturbation parameters, the system re-calls the scheduling engine to perform scheduling calculations. At this point, the scheduling algorithm has incorporated task switching penalties or enhanced global search capabilities, making it more suitable for resource allocation and optimization goals in the current state.

[0046] The scheduling algorithm uses the real-time task list, device availability, resource constraints, and optimization strategy as input data to perform task allocation and sorting calculations. The scheduling engine aims for a globally optimal or near-optimal solution and re-searches the solution space for the optimal scheduling path based on the updated objective function.

[0047] During execution, the system also monitors the convergence status of the scheduling calculation process and the resource load balance in real time to avoid new problems such as resource overload or scheduling imbalance. After the scheduling solution is completed, the system automatically generates a new set of task scheduling solutions.

[0048] The scheduling engine outputs information such as the allocated resources for each task, planned start and end times, task priority ranking, and critical path identification. All scheduling information is packaged into a unified data structure for easy integration with subsequent systems.

[0049] In addition, the system can simultaneously generate scheduling visualization diagrams, such as task Gantt charts, resource utilization charts, conflict warning charts, etc., for operators to view and verify the rationality of the scheduling results in real time.

[0050] The scheduling plan will also indicate whether there are high-risk task nodes that need to be observed, such as tasks that are at resource bottlenecks, close to delivery deadlines, or are susceptible to disturbances. The system will mark and prompt them for key monitoring in subsequent operation stages.

[0051] Once the scheduling plan is confirmed, the system sends the task assignment results to the Manufacturing Execution System (MES) via a standard interface. This process typically connects data with the MES through message queues, APIs, or industrial bus communication protocols such as OPC UA and MQTT.

[0052] After receiving the task plan, the manufacturing execution system immediately triggers the on-site control logic and dispatches the task to the specific equipment or operator position. For example, the automated equipment will receive the processing path, process parameters, and task execution time, while the operator will receive task scheduling and material delivery instructions through the terminal.

[0053] During this process, the system ensures that the scheduling results are synchronized with the actual resource status in the MES, avoiding data inconsistencies, duplicate scheduling, or resource conflicts. At the same time, the scheduling engine retains all scheduling records and result data for subsequent retrospective analysis and scheduling model optimization.

[0054] The manufacturing execution system begins operating according to the new schedule, continuously monitoring task execution status and actual feedback data. If resource status changes, urgent orders are added, or task anomalies occur during execution, the system triggers the adaptive scheduling process again based on real-time feedback, forming a closed-loop control mechanism.

[0055] Through these steps, the system seamlessly integrates optimization strategy generation, scheduling solution output, and execution, ensuring that scheduling results are not only theoretically optimal but also efficiently and stably implemented on-site. This entire process emphasizes real-time, closed-loop, and operability, providing the core support for intelligent manufacturing systems to achieve dynamic scheduling capabilities.

[0056] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0057] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0058] It should be understood that the term "and / or" herein is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent three situations: A exists alone, A and B exist at the same time, and B exists alone, where A and B may be singular or plural. In addition, the character " / " herein generally indicates that the objects associated with each other are in an "or" relationship, but it may also indicate an "and / or" relationship, which can be understood by referring to the context. A person of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0059] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A dynamic scheduling method for an intelligent manufacturing system, characterized by: include: Collect real-time status data and task information in manufacturing systems; The task switching interference index is generated by analyzing the resource disturbance degree caused by frequent task switching, and the objective function offset index is generated by analyzing the dynamic deviation degree between the current scheduling result and the theoretical global optimal. Inputting the task switching interference index and the objective function offset index into a pre-built scheduling performance anomaly prediction model, the output of the model being a scheduling health score value; Comparing the scheduling health score with a preset threshold, if the scheduling health score is lower than the preset threshold, determining that the scheduling has performance anomalies; For abnormal scheduling performance, the corresponding scheduling optimization strategy is dynamically selected, including whether to introduce a task switching penalty term into the scheduling objective function based on the task switching interference index, and adjusting the disturbance parameters in the scheduling algorithm based on the objective function offset index; Execute the optimized scheduling algorithm, output the final scheduling plan, and send the scheduling results to the manufacturing execution system for production task execution.

2. The method for dynamic scheduling of an intelligent manufacturing system according to claim 1, characterized in that: The status data includes equipment operation status, task progress, resource utilization rate and material delivery status; the task information includes order parameters, process path, delivery requirements and personnel scheduling information.

3. The method for dynamic scheduling of an intelligent manufacturing system according to claim 1, characterized in that: The task switching interference index is generated as follows: For each resource r∈R, where R represents the set of all resources participating in the current scheduling process, a fixed-length sliding time window W is set, with a unit time interval of Δt; the task execution sequence of the resource within the window is recorded as: ;in Indicates the task number that the resource performs in the i-th time slice, and n represents the total time slice; Traverse and record whether adjacent tasks have changed: ; Among them, the cumulative number of switching times ; Calculate the comprehensive switching intensity , the expression is: ; Defining resource disturbance weights , calculate the switching interference index of a single resource, the expression is: ;in is the maximum perturbation weight in the window, is the switching interference index of a single resource; the interference indexes of all resources are summarized to form the overall task switching interference index, which is expressed as: Where, is the task switching interference index.

4. The method for dynamic scheduling of an intelligent manufacturing system according to claim 3, characterized in that: The objective function shift index is generated as follows: Set the optimization objectives of the scheduling system to m, and define ; Where: x represents a scheduling solution; Represents the value of the i-th objective function; constructs the reference Pareto frontier set , the expression is: ; z is the total number of objective functions; calculate the minimum distance between the current solution and the Pareto frontier, for the current scheduling solution , and its objective function value is , calculate its distance to each point in the Pareto frontier: Where, Represents Euclidean distance; take the minimum distance As the offset distance; after normalizing the offset distance, the objective function offset index is obtained , the expression is: ;in: The upper limit of the reference distance.

5. The method for dynamic scheduling of an intelligent manufacturing system according to claim 4, characterized in that: The task switching interference index and the objective function deviation index are input into a pre-built scheduling performance anomaly prediction model. The output of the model is a scheduling health score value, including: The task switching interference index and the objective function offset index are converted into comprehensive feature vectors, which are used as inputs of the machine learning model. The machine learning model uses the scheduling health score value label predicted by each group of comprehensive feature vectors as the prediction target, and minimizes the sum of the prediction errors of all scheduling health score value labels as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The scheduling health score value is determined based on the model output results. The machine learning model is a polynomial regression model.

6. The method for dynamic scheduling of an intelligent manufacturing system according to claim 5, characterized in that: Comparing the scheduling health score with a preset threshold, if the scheduling health score is lower than the preset threshold, determining that the scheduling has performance anomalies, including: The obtained scheduling health score value is compared with the preset score threshold. If the scheduling health score value is greater than or equal to the score threshold, the scheduling performance is determined to be normal; if the scheduling health score value is less than the preset score threshold, the scheduling performance is determined to be abnormal.

7. The method for dynamic scheduling of an intelligent manufacturing system according to claim 6, characterized in that: The optimization strategy for the task switching interference index TSDD includes: modifying the objective function during the scheduling optimization process and introducing a task switching penalty term as an additional optimization target; the optimized objective function is expressed as: Where, is the original scheduling objective function; the task switching interference index calculated by the current TSDD system; α is the penalty weight coefficient, ,in is the reference threshold of TSDD, is the adjustment factor.

8. The method for dynamic scheduling of an intelligent manufacturing system according to claim 7, characterized in that: The optimization strategy for the objective function shift index OFDR includes adjusting the temperature attenuation factor in the simulated annealing algorithm, which is expressed as: If the OFDR is greater than its reference threshold, the initial disturbance of the annealing process is increased: ; is the current temperature parameter, λ is the temperature attenuation factor, represents the default decay rate, is the adjustment coefficient, is the reference threshold of OFDR; Adjust the mutation probability in the genetic algorithm, the expression is: ; represents the current mutation probability, is the default mutation probability; is the adjustment factor; When OFDR is greater than its reference threshold, the mutation probability is automatically increased, and the ability of searching to escape the local optimum is enhanced; when OFDR is less than its reference threshold, the mutation probability gradually decreases.

Images