Intelligent factory production scheduling method and system based on digital twinning

Abstract

The invention provides an intelligent factory production scheduling method and system based on digital twinning, and the method comprises the steps: firstly obtaining a real-time monitoring data set, including equipment operation states, material flow tracks and environment sensing data, of a physical workshop; generating a three-dimensional digital twinborn model containing an equipment dynamic operation topology, a material real-time distribution thermodynamic diagram and an environment state simulation layer, then performing operation state correlation analysis on the three-dimensional digital twinborn model, and generating a feature set of equipment operation efficiency, material flow bottleneck, environment interference and the like; generating a dynamic scheduling parameter set containing equipment load balancing, material priority distribution and environment adaptation compensation parameters, and finally calling a preset scheduling strategy optimization algorithm to perform multi-objective optimization processing on the dynamic scheduling parameter set to generate an equipment scheduling, material distribution and environment regulation and control instruction set. And synchronizing to a physical workshop to execute real-time scheduling operation, thereby realizing intelligent and efficient production scheduling.

Description

Technical Field

[0001] The present invention relates to the technical field of digital factories, and more particularly, to an intelligent factory production scheduling method and system based on digital twin. Background Art

[0002] In the context of the rapid development of intelligent manufacturing technology, traditional factory production scheduling methods are facing increasingly severe challenges. Existing technologies mainly rely on static predefined rules or offline optimization models based on historical data, and their essential characteristics are manifested as a one-way control mode of the production process, that is, passive allocation of equipment, materials, and environmental elements through preset fixed parameters.

[0003] Such solutions have significant technical defects: First, due to the lack of the ability to dynamically perceive the real-time operating status of the physical workshop, they cannot respond in a timely manner to dynamic interference events such as sudden equipment failures, abnormal material flows, or fluctuations in environmental parameters, resulting in frequent deviations between production plans and actual executions; Second, existing scheduling models mostly use two-dimensional planar or locally discrete representations of the workshop state, making it difficult to establish a full-element association model among equipment-material-environment elements, and lacking a global perspective for identifying production bottlenecks and optimizing resource allocation; Third, traditional optimization algorithms usually take equipment utilization rate, material turnover rate, and energy consumption indicators as independent optimization objectives, lacking a multi-objective collaborative optimization mechanism, resulting in the fact that when improving one indicator, the scheduling scheme often sacrifices other indicators, and it is difficult to achieve the optimal solution of the overall production efficiency.

[0004] In current industrial practices, although digital twin technology has been initially applied to the fields of production monitoring and visualization, existing applications are mostly limited to single-dimensional data mapping or static scenario simulation, and have not yet formed the ability of dynamic modeling and real-time optimization for production scheduling. Specifically, existing digital twin systems either only implement the mirror display of equipment operating states, or only provide the trajectory playback of material flows, lacking the technical means to perform three-dimensional fusion modeling of equipment dynamic topology, real-time material distribution, and environmental states, and have not established a closed-loop optimization mechanism for mapping from digital twin models to dynamic scheduling parameters. This technical fragmentation results in the digital twin model being unable to effectively support the real-time decision-making needs of production scheduling, and its application value is limited to the level of post-event analysis and auxiliary diagnosis. Summary of the Invention

[0005] In view of the above-mentioned problems, in combination with the first aspect of the present invention, embodiments of the present invention provide an intelligent factory production scheduling method based on digital twin, and the method includes: Obtain a set of real-time monitoring data of the physical workshop, where the set of real-time monitoring data includes equipment operating state data, material flow trajectory data, and environmental sensing data; Generate a three-dimensional digital twin model of the physical workshop based on the real-time monitoring data set, where the three-dimensional digital twin model includes an equipment dynamic operation topology, a real-time material distribution heat map, and an environmental state simulation layer; Perform an associated analysis of the operating state of the three-dimensional digital twin model to generate an equipment operation efficiency feature set, a material flow bottleneck feature set, and an environmental interference feature set; Generate a dynamic scheduling parameter set based on the equipment operation efficiency feature set, the material flow bottleneck feature set, and the environmental interference feature set. The dynamic scheduling parameter set includes equipment load balancing parameters, material priority allocation parameters, and environmental adaptation compensation parameters; Call a preset scheduling strategy optimization algorithm to perform multi-objective optimization processing on the dynamic scheduling parameter set, generate an equipment scheduling instruction set, a material allocation instruction set, and an environmental regulation instruction set, and synchronize them to the physical workshop to perform real-time scheduling operations.

[0006] In a possible implementation manner of the first aspect, the obtaining of the real-time monitoring data set of the physical workshop includes: Collect the real-time energy consumption data, equipment vibration spectrum data, and equipment temperature change data of the equipment through edge sensors deployed at the equipment end of the physical workshop, and associate the real-time energy consumption data, the equipment vibration spectrum data, and the equipment temperature change data with the unique equipment identifier; Obtain the real-time position coordinate sequence of the material in the physical workshop through the material positioning tag, and calculate the residence time interval and the moving speed change rate of the material between adjacent processes based on the real-time position coordinate sequence; Collect the temperature distribution data, humidity gradient data, and air particle concentration data in the physical workshop through the environmental monitoring device, and map the temperature distribution data, the humidity gradient data, and the air particle concentration data to the workshop space grid coordinates based on the time stamp; Aggregate the real-time energy consumption data, the equipment vibration spectrum data, and the equipment temperature change data associated with the unique equipment identifier, the residence time interval and the moving speed change rate, and the temperature distribution data, the humidity gradient data, and the air particle concentration data associated with the workshop space grid coordinates into the real-time monitoring data set according to a preset time window.

[0007] In a possible implementation manner of the first aspect, the generating of the three-dimensional digital twin model of the physical workshop based on the real-time monitoring data set includes: Match the predefined three-dimensional geometric model and physical property parameters of the device according to the device unique identifier, update the dynamic operating state of the three-dimensional geometric model based on the real-time energy consumption data, vibration spectrum data and temperature change data of the device, and generate the dynamic operating topology of the device; Overlay the real-time position coordinate sequence with the process layout map, calculate the aggregation density and flow path deviation degree of materials at process nodes, and generate the real-time distribution heat map of the materials with color gradient identification based on the aggregation density and the flow path deviation degree; Construct an environmental state surface equation according to the temperature distribution data, humidity gradient data and air particulate matter concentration data associated with the workshop space grid coordinates, and generate the environmental state simulation layer covering the entire physical workshop through an interpolation algorithm; Fuse the device dynamic operation topology, the real-time distribution heat map of the materials and the environmental state simulation layer in a unified spatial coordinate system, and render and generate the three-dimensional digital twin model in the visualization interface.

[0008] In a possible implementation manner of the first aspect, the performing correlation analysis on the operating state of the three-dimensional digital twin model to generate a set of device operating efficiency characteristics, a set of material flow bottleneck characteristics and a set of environmental interference characteristics includes: Extract a set of device operating efficiency characteristics from the device dynamic operation topology, and the set of device operating efficiency characteristics includes the production capacity matching deviation rate between adjacent process devices, the proportion of device idle time and the device failure warning index; Extract a set of material flow bottleneck characteristics from the real-time distribution heat map of the materials, and the set of material flow bottleneck characteristics includes the material accumulation duration when the material accumulation amount at the process node exceeds the threshold, the number of path crossing conflicts in the material flow path crossing conflict area, and the overtime ratio of the material waiting for processing; Extract a set of environmental interference characteristics from the environmental state simulation layer, and the set of environmental interference characteristics includes the influence coefficient of the high-temperature area on the device heat dissipation efficiency, the fluctuation range of the material moisture content in the humidity mutation area, and the wear rate of the device precision parts by the particulate matter concentration; Compare the production capacity matching deviation rate, the proportion of device idle time and the device failure warning index with the historical operation baseline to generate the root cause analysis result of the device operating efficiency decline; Match the material accumulation duration, the number of path crossing conflicts and the overtime ratio with the preset flow optimization rules to generate the grading result of the material flow bottleneck; Associate and query the influence coefficient, the fluctuation range and the wear rate with the environmental regulation strategy library to generate the compensation priority ranking result of the environmental interference.

[0009] In a possible implementation manner of the first aspect, generating the dynamic scheduling parameter set based on the device operation efficiency feature set, the material flow bottleneck feature set, and the environmental interference feature set includes: According to the root cause analysis result of the decline in device operation efficiency, calculate the device load balancing parameters, where the device load balancing parameters include the task diversion ratio of high-load devices, the task additional amount of low-load devices, and the replacement device selection weight of faulty devices; According to the grading result of the material flow bottleneck, calculate the material priority allocation parameters, where the material priority allocation parameters include the path preemption permission of emergency order materials, the reallocation path planning of stacked materials, and the adjustment of the material passage time window in the conflict area; According to the compensation priority sorting result of the environmental interference, calculate the environmental adaptation compensation parameters, where the environmental adaptation compensation parameters include the increased number of heat dissipation devices in high-temperature areas, the desiccant delivery frequency in areas with sudden humidity changes, and the working intensity of filtering devices in areas with high particulate matter concentration; Encapsulate the task diversion ratio, the task additional amount, and the replacement device selection weight into a device scheduling parameter unit; Encapsulate the path preemption permission, the reallocation path planning, and the passage time window adjustment into a material allocation parameter unit; Encapsulate the increased number of heat dissipation devices, the desiccant delivery frequency, and the working intensity of filtering devices into an environmental regulation parameter unit; Integrate the device scheduling parameter unit, the material allocation parameter unit, and the environmental regulation parameter unit into the dynamic scheduling parameter set according to the time series.

[0010] In a possible implementation manner of the first aspect, invoking a preset scheduling strategy optimization algorithm to perform multi-objective optimization processing on the dynamic scheduling parameter set includes: Construct a multi-objective optimization function, where the objective variables of the multi-objective optimization function include the comprehensive device utilization rate, the total material transfer duration, and the environmental regulation energy consumption cost; Input the device scheduling parameter unit, the material allocation parameter unit, and the environmental regulation parameter unit in the dynamic scheduling parameter set into the scheduling strategy optimization algorithm, and calculate the Pareto optimal solution set that maximizes the comprehensive device utilization rate, minimizes the total material transfer duration, and minimizes the environmental regulation energy consumption cost through a constraint satisfaction algorithm; Select an optimized solution that meets the preset conditions from the Pareto optimal solution set, and map the optimized solution to a device scheduling instruction set, a material allocation instruction set, and an environmental regulation instruction set; Among them, the device scheduling instruction set includes device task shunting instructions, device task appending instructions, and device alternative operation instructions; The material distribution instruction set includes path priority adjustment instructions, material redistribution instructions, and conflict area time window distribution instructions; The environmental regulation instruction set includes heat dissipation device start / stop instructions, desiccant delivery instructions, and filter device power adjustment instructions.

[0011] In a possible implementation manner of the first aspect, the synchronization to the physical workshop to perform real-time scheduling operations includes: Issuing the device task shunting instructions, the device task appending instructions, and the device alternative operation instructions to the controllers of the corresponding devices through the industrial Internet of Things communication protocol to trigger the adjustment of device operation parameters; Sending the path priority adjustment instructions, the material redistribution instructions, and the conflict area time window distribution instructions to the automated guided vehicle and the robotic arm through the material scheduling system to trigger the real-time change of the material path; Transmitting the heat dissipation device start / stop instructions, the desiccant delivery instructions, and the filter device power adjustment instructions to the environmental execution mechanism through the environmental regulation system to trigger the adaptive adjustment of environmental parameters; In the three-dimensional digital twin model, the device operation status, the material flow path, and the environmental parameter changes are updated in real time, and the actual response data of the physical workshop is compared with the expected effects of the instruction set to generate a scheduling deviation correction coefficient; When the scheduling deviation correction coefficient exceeds the preset threshold, the generation and optimization processing of the dynamic scheduling parameter set are re-triggered.

[0012] In a possible implementation manner of the first aspect, the comparison of the actual response data of the physical workshop with the expected effects of the instruction set to generate a scheduling deviation correction coefficient includes: Obtaining the actual device utilization rate after the physical workshop executes the device scheduling instruction set, the actual material transfer duration after executing the material distribution instruction set, and the actual energy consumption cost after executing the environmental regulation instruction set; Calculating the absolute value of the difference between the actual device utilization rate and the expected comprehensive device utilization rate to obtain a device scheduling deviation value; Calculating the percentage of the difference between the actual material transfer duration and the expected total material transfer duration to obtain a material scheduling deviation value; Calculating the ratio of the actual energy consumption cost to the expected environmental regulation energy consumption cost to obtain an environmental regulation deviation value; Performing a weighted sum of the device scheduling deviation value, the material scheduling deviation value, and the environmental regulation deviation value to generate the scheduling deviation correction coefficient.

[0013] In a possible implementation of the first aspect, the method further includes: Statistical comprehensive performance indicators after the physical workshop executes the historical scheduling instruction set within a preset period, where the comprehensive performance indicators include the mean time between failures of equipment, the on-time delivery rate of materials, and the environmental regulation cost per unit output; Compare the comprehensive performance indicators with prior benchmark data to generate the intensity of parameter optimization requirements; When the intensity of parameter optimization requirements exceeds a preset threshold, adjust the generation weights of the equipment load balancing parameter, the material priority allocation parameter, and the environmental adaptation compensation parameter in the dynamic scheduling parameter set; Based on the adjusted generation weights, re-execute the operation status correlation analysis and multi-objective optimization process to generate an updated scheduling instruction set and synchronize it to the physical workshop.

[0014] For example, in a possible implementation of the first aspect, the adjustment of the generation weights of the equipment load balancing parameter, the material priority allocation parameter, and the environmental adaptation compensation parameter in the dynamic scheduling parameter set includes: According to the ratio of the difference between the mean time between failures of the equipment and the prior benchmark, increase the calculation proportion of the alternative equipment selection weight in the equipment load balancing parameter; According to the percentage of the difference between the on-time delivery rate of materials and the prior benchmark, increase the priority coefficient of the path preemption permission in the material priority allocation parameter; According to the difference amplitude between the environmental regulation cost per unit output and the prior benchmark, reduce the adjustment step of the working intensity of the filtering equipment in the environmental adaptation compensation parameter; Write the adjusted calculation proportion, priority coefficient, and adjustment step into the weight configuration file of the scheduling strategy optimization algorithm to take effect in subsequent optimization processes.

[0015] In another aspect, an embodiment of the present invention further provides an intelligent factory production scheduling system based on digital twin, including a processor and a machine-readable storage medium. The machine-readable storage medium is connected to the processor. The machine-readable storage medium is used to store programs, instructions, or codes, and the processor is used to execute the programs, instructions, or codes in the machine-readable storage medium to implement the above method.

[0016] Based on the above aspects, the embodiments of this application achieve the full-element dynamic mapping and closed-loop optimization between the physical workshop and the virtual model, significantly enhancing the global collaboration and adaptive capabilities of production scheduling. Specifically, by collecting real-time device operating status, material flow trajectories, and environmental sensing data, a three-dimensional digital twin model integrating multi-source heterogeneous information is constructed. This three-dimensional digital twin model not only realizes the multi-dimensional visual representation of device dynamic topology, material distribution heat maps, and environmental status, but also accurately extracts a composite feature set of device operating efficiency, material flow bottlenecks, and environmental disturbances through correlation analysis of operating status. On this basis, the dynamically generated set of scheduling parameters effectively solves the contradiction between local and global optimality in traditional scheduling methods through a collaborative optimization mechanism of device load balancing, material priority allocation, and environmental adaptation compensation. Further, through a preset multi-objective optimization algorithm to iteratively optimize the scheduling parameters, the finally output set of device scheduling, material allocation, and environmental regulation instructions can respond in real time to the dynamic changes in the production process, forming a two-way interaction closed-loop between the physical workshop and the digital twin model. This not only significantly improves the utilization rate of production resources and the consistency of production rhythm, but also enhances the robustness of the production system through an active compensation mechanism for environmental disturbances, thus achieving multi-dimensional collaborative improvement of production efficiency, quality stability, and energy consumption control in complex and changeable industrial scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 FIG. is a schematic execution flowchart of an intelligent factory production scheduling method based on digital twin provided by an embodiment of the present invention.

[0018] Figure 2 FIG. is a schematic diagram of exemplary hardware and software components of an intelligent factory production scheduling system based on digital twin provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] As described in the technical problems recorded in the foregoing background art, the technical bottleneck that the field urgently needs to break through is: how to build an intelligent scheduling system with multi-dimensional perception, dynamic modeling, and real-time optimization capabilities, and achieve global optimal allocation of production resources through in-depth interaction between the physical workshop and the digital space. The embodiments of the present invention innovatively integrate three-dimensional digital twin modeling, multi-source heterogeneous data correlation analysis, and multi-objective dynamic optimization technologies, and for the first time propose a full-element dynamic scheduling method for intelligent factories, effectively solving the fundamental defects of the prior art in terms of real-time performance, global performance, and collaboration.

[0020] The present invention will be specifically described below with reference to the accompanying drawings of the specification. Figure 1 FIG. is a schematic flowchart of an intelligent factory production scheduling method based on digital twin provided by an embodiment of the present invention. The intelligent factory production scheduling method based on digital twin will be introduced in detail below.

[0021] Step S110: Obtain the real-time monitoring data set of the physical workshop, where the real-time monitoring data set includes equipment operation status data, material flow trajectory data, and environmental sensing data.

[0022] Taking a physical workshop in a digital factory for the production of new energy vehicle power batteries as an example, this physical workshop undertakes the key production link of battery module assembly and is equipped with production equipment, a material conveying system, and environmental control facilities.

[0023] In this embodiment, step S110 may include: Step S111: Collect the real-time energy consumption data, vibration spectrum data, and temperature change data of the equipment through edge sensors deployed at the equipment end of the physical workshop, and associate the real-time energy consumption data, vibration spectrum data, and temperature change data of the equipment with the unique identifier of the equipment.

[0024] Specifically, there are multiple devices with different functions in this physical workshop. Taking the battery welding equipment numbered E-001 as an example. Edge sensors are installed at the key parts of this battery welding equipment to monitor various data in real time. At a certain moment, the real-time energy consumption data of the equipment shows that the power consumption of this battery welding equipment per hour is 150 degrees. At the same time, the vibration spectrum data of the equipment is collected. After being processed by a professional spectrum analysis tool, it is obtained that the vibration energy ratio in the low-frequency band (0-100Hz) is 0.4, the vibration energy ratio in the medium-frequency band (100-500Hz) is 0.3, and the vibration energy ratio in the high-frequency band (above 500Hz) is 0.3. For the temperature change data of the equipment, it is monitored that the temperature of the equipment welding head part is 60 degrees Celsius at this moment. Among them, the above data are all associated with the unique identifier of equipment E-001.

[0025] Step S112: Obtain the real-time position coordinate sequence of the material in the physical workshop through the material positioning label, and calculate the residence time interval and moving speed change rate of the material between adjacent processes based on the real-time position coordinate sequence.

[0026] In this embodiment, the materials in the workshop are various components required for battery module assembly, such as battery cells, connection plates, etc. Taking battery cells as an example, each battery cell is equipped with a material positioning tag. Suppose a batch of battery cells starts from the raw material storage area and goes to the first assembly process. During this process, the material positioning tag records the position coordinates of the battery cells at each moment in real time, forming a real-time position coordinate sequence. For example, moving from position A (10, 20, 3) to position B (15, 22, 3), by analyzing this coordinate sequence and combining the recorded timestamp information, the residence time interval of the battery cell between adjacent processes can be calculated. Suppose it takes a total of 5 minutes from leaving the previous process to entering the next process, and this is the residence time interval. At the same time, according to the coordinate changes and time information, the change rate of the moving speed is calculated. If the speed is 5 unit lengths per minute in the first 2 minutes and then becomes 4 unit lengths per minute in the next 3 minutes, through the set calculation method (such as the change in speed divided by the change in time), the change rate of the moving speed is -0.2 unit lengths per minute².

[0027] Step S113: Collect the temperature distribution data, humidity gradient data, and air particulate matter concentration data in the physical workshop through the environmental monitoring device, and map the temperature distribution data, the humidity gradient data, and the air particulate matter concentration data to the workshop space grid coordinates based on the timestamp.

[0028] In this embodiment, a plurality of environmental monitoring devices are evenly distributed in the physical workshop for comprehensively collecting environmental data. At a certain moment, the temperature distribution data collected by the above environmental monitoring device shows that the temperature in the upper left corner area of the workshop is 23 degrees Celsius, the temperature in the upper right corner area is 24 degrees Celsius, the temperature in the lower left corner area is 22 degrees Celsius, the temperature in the lower right corner area is 23.5 degrees Celsius, etc., forming a multi-dimensional temperature distribution data set. In terms of humidity gradient data, it is measured by the device that the humidity shows a gradually increasing trend from one end to the other end of the workshop. For example, in the length direction of the workshop, for every 10 meters forward, the humidity increases by 2%. The collection of air particulate matter concentration data shows that in the area near the welding equipment, the particulate matter concentration is 50 micrograms per cubic meter, while in the material storage area far from the welding equipment, the particulate matter concentration is 20 micrograms per cubic meter. Among them, the above data are all mapped to the workshop space grid coordinates according to the timestamp at the time of collection. For example, the workshop is divided into a 10×10 space grid, each grid has a corresponding coordinate, and the collected data is associated with the corresponding grid coordinate for subsequent construction of the environmental state simulation layer.

[0029] Step S114: Aggregate the real-time energy consumption data, vibration spectrum data, and temperature change data of the device associated with the device unique identifier, the residence time interval and the moving speed change rate, and the temperature distribution data, humidity gradient data, and air particulate concentration data associated with the workshop space grid coordinates into the real-time monitoring data set according to a preset time window.

[0030] For example, the preset time window is set to 1 hour. Within this 1 hour, all the data associated with device E-001 can be integrated according to the set aggregation rules, including real-time energy consumption data (150 degrees of electricity per hour), vibration spectrum data (low frequency band 0.4, medium frequency band 0.3, high frequency band 0.3), temperature change data (60 degrees Celsius at the welding head), residence time interval of the battery cell (5 minutes), moving speed change rate (-0.2 unit length per minute²), temperature distribution data (temperature values in each area) based on the workshop space grid coordinates, humidity gradient data (humidity increases by 2% every 10 meters), and air particulate concentration data (concentration values in different areas). This aggregation process unifies various types of data within the same time window to form a complete real-time monitoring data set.

[0031] Step S120: Generate a three-dimensional digital twin model of the physical workshop based on the real-time monitoring data set. The three-dimensional digital twin model includes a device dynamic operation topology, a real-time distribution heat map of materials, and an environmental state simulation layer.

[0032] In this embodiment, when generating the three-dimensional digital twin model, the previously obtained and aggregated real-time monitoring data set is fully utilized. Specifically, step S120 may include: Step S121: Match the pre-defined three-dimensional geometric model and physical property parameters of the device according to the device unique identifier, and update the dynamic operation state of the three-dimensional geometric model based on the real-time energy consumption data, vibration spectrum data, and temperature change data of the device to generate the device dynamic operation topology.

[0033] For example, for device E-001, its predefined three-dimensional geometric model is a digital model that accurately represents the appearance shape and structural layout of the device. This three-dimensional geometric model also contains physical property parameters such as the mass and material of the device. Thus, based on the real-time energy consumption data (150 degrees of electricity per hour), vibration spectrum data (low frequency band 0.4, middle frequency band 0.3, high frequency band 0.3), and temperature change data (60 degrees Celsius at the welding head part) of device E-001 collected in real time, the dynamic operating state of this three-dimensional geometric model can be updated. For example, through the energy consumption data, the display related to the energy consumption of the device in the three-dimensional geometric model can be adjusted, such as using different colors or lighting effects to show the energy consumption intensity of the device; according to the vibration spectrum data, the vibration conditions of each part of the device are shown in the form of dynamic graphics in the three-dimensional geometric model; based on the temperature change data, the color of the welding head part of the device in the three-dimensional geometric model is changed in real time to intuitively reflect the temperature level. Through the above update operations, a dynamic operating topology of device E-001 is generated. For other devices in the workshop, such as the battery assembly device numbered E-002 and the detection device E-003, etc., they are also processed in the same way, and finally the dynamic operating topologies of all devices are integrated together to form the device dynamic operating topology of the entire workshop.

[0034] Step S122: Superimpose the real-time position coordinate sequence on the process layout map, calculate the aggregation density and flow path deviation degree of the material at the process node, and generate the real-time distribution heat map of the material containing color gradient markings based on the aggregation density and the flow path deviation degree.

[0035] In this embodiment, the real-time position coordinate sequence of materials such as battery cells can be superimposed and analyzed on the process layout map of the workshop. The process layout map clearly marks the positions and interconnection relationships of each process. For example, at a certain moment, it is found that a large number of battery cells are gathered near the second assembly process node. By analyzing and calculating the real-time position coordinate sequence, the aggregation density of the material at this process node is obtained. Assuming that the area of the region where this process node is located is 10 square meters and 50 battery cells are gathered at this moment, the aggregation density is 5 battery cells per square meter. At the same time, by comparing the actual flow path of the material with the preset flow path, the flow path deviation degree is calculated. For example, the preset path is to reach process B directly from process A, but the actual material flow path shows a certain bend, and the deviation degree is calculated to be 15% through a specific algorithm. Based on the above aggregation density and flow path deviation degree data, a real-time distribution heat map of the material is generated. For regions with high aggregation density, they are represented by red on the heat map; for regions with low aggregation density, they are represented by blue; regions with a large flow path deviation degree have brighter colors to intuitively display the real-time distribution and flow state of the material.

[0036] Step S123: Construct an environmental state surface equation based on the temperature distribution data, the humidity gradient data, and the air particulate matter concentration data associated with the workshop space grid coordinates, and generate the environmental state simulation layer covering the entire physical workshop through an interpolation algorithm.

[0037] In this embodiment, an environmental state surface equation can be constructed according to the temperature distribution data, the humidity gradient data, and the air particulate matter concentration data associated with the workshop space grid coordinates. Taking the temperature distribution data as an example, assume that the workshop space grid coordinates are two-dimensional coordinates (x, y), and different temperature values T1, T2, etc. are collected at different grid points (x1, y1), (x2, y2), etc. A surface equation that can describe the temperature distribution in the entire workshop space is constructed by mathematical methods. For example, a binary polynomial equation T(x, y)=a0+a1x+a2y+a3x²+a4xy+a5y² can be constructed using the method of polynomial fitting. By solving the coefficients a0, a1, a2, etc., the binary polynomial equation can describe the temperature distribution as accurately as possible. Similar methods are used to construct the corresponding surface equations for the humidity gradient data and the air particulate matter concentration data. Then, through the interpolation algorithm, data filling is performed in the blank areas of the workshop space to generate an environmental state simulation layer covering the entire physical workshop, and thus a continuous and complete environmental state simulation can be obtained, visually showing the distribution of temperature, humidity, and air particulate matter concentration in the workshop.

[0038] Step S124: Integrate the device dynamic operation topology, the real-time material distribution heat map, and the environmental state simulation layer in a unified spatial coordinate system, and render and generate the three-dimensional digital twin model in the visualization interface.

[0039] In this embodiment, the generated device dynamic operation topology, the real-time material distribution heat map, and the environmental state simulation layer can be integrated in a unified spatial coordinate system. During this process, ensure that each part accurately corresponds in spatial position. For example, the position of the device in the device dynamic operation topology matches the position of the material in the real-time material distribution heat map and the position of the environmental data in the environmental state simulation layer. Then, through professional visualization software, the integrated three-dimensional digital twin model is rendered in the visualization interface. On the visualization interface, the operator can intuitively see the operating state of the device, the real-time distribution of the material, and the simulation display of the environmental state. The device is presented in the form of a dynamic three-dimensional model, and its operating parameters are displayed through effects such as color and light and shadow; the real-time material distribution heat map shows the aggregation and flow of the material with an intuitive color gradient; the environmental state simulation layer shows the distribution of temperature, humidity, and air particulate matter concentration with a realistic graph. In this way, an intuitive and accurate three-dimensional digital twin model of the physical workshop is generated.

[0040] Step S130: Perform an analysis of the operating status association for the 3D digital twin model to generate a set of device operation efficiency characteristics, a set of material flow bottleneck characteristics, and a set of environmental interference characteristics.

[0041] In this embodiment, step S130 may include: Step S131: Extract a set of device operation efficiency characteristics from the device dynamic operation topology. The set of device operation efficiency characteristics includes the production capacity matching deviation rate between adjacent process devices, the proportion of device idle time, and the device failure warning index.

[0042] In this embodiment, taking device E-001 and its adjacent process device E-002 in the device dynamic operation topology as an example, assume that device E-001 can complete the welding work of 100 battery cells per hour, and device E-002 can process 120 welded battery cells per hour. By calculation, the production capacity matching deviation rate between adjacent process devices is (120 - 100) / 120 ≈ 16.7%. For the proportion of device idle time, the idle time of device E-001 within a certain period (such as 8 hours of working time) is counted. Assume that device E-001 is idle for 1 hour within these 8 hours, then the proportion of device idle time is 1 / 8 = 12.5%. The device failure warning index is obtained through comprehensive analysis of device vibration spectrum data, temperature change data, etc. For example, when the proportion of high-frequency vibration energy in the device vibration spectrum data continuously exceeds 0.4, and the device temperature continuously exceeds 65 degrees Celsius, through a specific algorithm and preset rules, the device failure warning index is calculated to be 0.6, indicating that there is a certain risk of device failure. Thus, the above data is integrated together to form a set of device operation efficiency characteristics.

[0043] Step S132: Extract a set of material flow bottleneck characteristics from the real-time material distribution heat map. The set of material flow bottleneck characteristics includes the material accumulation duration when the material accumulation amount at the process node exceeds the threshold, the number of path crossing conflicts in the material transfer path crossing conflict area, and the overtime ratio of material waiting for processing.

[0044] In this embodiment, taking the node of the second assembly process as an example from the real-time material distribution heat map. Assume that the threshold of the material accumulation amount at this process node is set to 30 battery monomers per square meter. During a certain period, the material accumulation amount at this process node exceeds the threshold, and the duration is 30 minutes, which is the material accumulation duration. For the area where the material flow path intersects and conflicts, by analyzing the real-time position coordinate sequence of the material and the flow path information, it is found that there are 3 areas with the situation of material flow path intersection and conflict, which is the number of path intersection and conflicts. Assume that the preset maximum waiting time for materials to be processed is 20 minutes. Among a batch of counted materials, 20% of the materials have a waiting processing time exceeding 20 minutes, and the overtime ratio of materials waiting for processing is obtained as 20%. Thus, after organizing the above data, a set of material flow bottleneck characteristics is formed.

[0045] Step S133: Extract the environmental interference feature set from the environmental state simulation layer, where the environmental interference feature set includes the influence coefficient of the high-temperature area on the heat dissipation efficiency of the equipment, the fluctuation range of the moisture content of the material in the humidity mutation area, and the wear rate of the precision components of the equipment by the particulate matter concentration.

[0046] In this embodiment, assume that it is found from the environmental state simulation layer that a corner area in the workshop has a temperature continuously higher than 30 °C, which belongs to the high-temperature area. Through experiments and analysis, it is obtained that the influence coefficient of this high-temperature area on the heat dissipation efficiency of equipment E-001 is 0.2, that is, the high temperature reduces the heat dissipation efficiency of the equipment by 20%. In another area of the workshop, the humidity changes suddenly. For example, the humidity increases from 50% to 70% in a short time. Through the monitoring and analysis of the moisture content of the materials in this area, it is obtained that the fluctuation range of the moisture content of the materials in the humidity mutation area is 5%. For the wear rate of the precision components of the equipment by the particulate matter concentration, taking a precision welding head in equipment E-001 as an example, in the area with a higher particulate matter concentration, after a period of operation and monitoring, it is found that the wear rate of the welding head increases by 0.1 mm / day compared with that in the environment with a normal particulate matter concentration, which is the wear rate of the precision components of the equipment by the particulate matter concentration. Thus, the above data are summarized to form an environmental interference feature set.

[0047] Step S134: Compare the production capacity matching deviation rate, the equipment idle time ratio, and the equipment failure warning index with the historical operation baseline to generate the root cause analysis result of the decline in equipment operation efficiency.

[0048] In this embodiment, the calculated production capacity matching deviation rate (16.7%), the proportion of equipment idle time (12.5%), and the equipment failure warning index (0.6) of equipment E-001 can be compared with the historical operation baseline. The historical operation baseline data shows that the production capacity matching deviation rate should generally be controlled within 10%, the proportion of equipment idle time should be below 10%, and the equipment failure warning index should be lower than 0.5. Through comparison, it is found that the relatively high production capacity matching deviation rate may be due to the aging of some components of equipment E-001, resulting in a decrease in welding speed; the high proportion of equipment idle time may be due to unreasonable production plan arrangements and loose process connections; the equipment failure warning index exceeding the threshold is because the equipment has been running in a high-temperature environment for a long time, affecting the stability of the equipment. Based on the above analysis, the root cause analysis result of the decrease in equipment operation efficiency is generated.

[0049] Step S135: Match the material stacking duration, the number of path crossing conflicts, and the overtime ratio with the preset flow optimization rules to generate a grading result of the material flow bottleneck.

[0050] In this embodiment, the material stacking duration (30 minutes), the number of path crossing conflicts (3), and the overtime ratio (20%) can be matched with the preset flow optimization rules. The preset flow optimization rules stipulate that the material stacking duration within 15 minutes is normal, 15 - 30 minutes is a mild bottleneck, and more than 30 minutes is a moderate bottleneck; the number of path crossing conflicts within 2 is normal, 2 - 4 is a mild bottleneck, and more than 4 is a moderate bottleneck; the overtime ratio within 15% is normal, 15% - 25% is a mild bottleneck, and more than 25% is a moderate bottleneck. According to the above rules, it is comprehensively judged that the level of this material flow bottleneck is a mild bottleneck.

[0051] Step S136: Perform an associated query on the influence coefficient, the fluctuation range, and the wear rate with the environmental regulation strategy library to generate a sorting result of the compensation priorities for environmental interference.

[0052] In this embodiment, the influence coefficient of the high-temperature area on the heat dissipation efficiency of the equipment (0.2), the fluctuation range of the moisture content of the material in the humidity mutation area (5%), and the wear rate of the precision components of the equipment by the particulate matter concentration (0.1 mm / day) are associated and queried with the environmental control strategy library. In the environmental control strategy library, there are corresponding control strategies and compensation priorities for different degrees of influence coefficient, fluctuation range, and wear rate. Through query, it is found that the influence of the high-temperature area on the heat dissipation efficiency of the equipment is relatively serious, and the compensation priority is the highest; the fluctuation range of the moisture content of the material in the humidity mutation area is the second; the wear rate of the precision components of the equipment by the particulate matter concentration is relatively low. Therefore, the generated compensation priority ranking result for environmental interference is: the heat dissipation problem in the high-temperature area is processed first, followed by the moisture content problem of the material in the humidity mutation area, and finally the wear problem of the precision components of the equipment by the particulate matter concentration.

[0053] Step S140: Generate a dynamic scheduling parameter set based on the equipment operation efficiency feature set, the material flow bottleneck feature set, and the environmental interference feature set, where the dynamic scheduling parameter set includes equipment load balancing parameters, material priority allocation parameters, and environmental adaptation compensation parameters.

[0054] In this embodiment, step S140 may include: Step S141: Calculate equipment load balancing parameters according to the root cause analysis result of the decrease in equipment operation efficiency, where the equipment load balancing parameters include the task diversion ratio of high-load equipment, the task addition amount of low-load equipment, and the alternative equipment selection weight of faulty equipment.

[0055] For example, according to the root cause analysis result of the decrease in equipment operation efficiency, for equipment E-001, since its production capacity matching deviation rate is relatively high and there is a risk of failure, it belongs to high-load equipment. After comprehensive evaluation, the task diversion ratio of high-load equipment is determined to be 30%, that is, 30 tasks are diverted from the current 100 battery cell welding tasks per hour of equipment E-001 to other equipment. The equipment E-004 in the workshop currently has a low load, and the actual task processing volume per hour is 60. 20 of the 30 tasks diverted from the high-load equipment E-001 are added to the equipment E-004, that is, the task addition amount of the low-load equipment E-004 is 20. For the possible failure situation of equipment E-001, after evaluating the performance, compatibility, etc. of all equipment in the workshop, the selection weight of equipment E-005 as its alternative equipment is determined to be 0.6. This means that when equipment E-001 fails, there is a 60% possibility of preferentially selecting equipment E-005 to replace its work. After sorting out the task diversion ratio, the task addition amount, and the alternative equipment selection weight, an equipment scheduling parameter unit is formed as part of the equipment load balancing parameters.

[0056] Step S142: According to the classification result of the material flow bottleneck levels, calculate the material priority allocation parameters, where the material priority allocation parameters include the path preemption permission for emergency order materials, the reallocation path planning for piled-up materials, and the adjustment of the material passage time window in the conflict area.

[0057] In this embodiment, since the material flow bottleneck level is classified as a mild bottleneck, calculate the material priority allocation parameters based on this result. For emergency order materials, grant them the path preemption permission. Suppose there is an emergency order currently, which requires assembling a batch of battery modules in a short time. After the materials for this emergency order enter the workshop, they have the right to preferentially use the paths originally occupied by other regular order materials. For example, the materials for regular orders were originally transported from the raw material area to the first process according to a fixed path, but the materials for the emergency order can, under specific system instructions, temporarily preempt this path to speed up their circulation speed.

[0058] For piled-up materials, conduct reallocation path planning. In the case of material accumulation at the second assembly process node, re-plan the path for the piled-up battery cells and other materials. Originally, the above-mentioned materials were transported to this process node along a straight path, but due to the accumulation, a new path is planned to bypass the current congested area and reach this process node from another direction in the workshop to relieve the accumulation situation.

[0059] For the adjustment of the material passage time window in the conflict area, after discovering a conflict area where 3 material flow paths cross, adjust the material passage time in the above area. For example, originally all materials were allowed to pass freely in each time period, but now the passage time is divided into multiple small segments, and different types of materials pass in different small time segments. For example, first allow battery cell materials to pass in the first time period, and then allow connecting piece materials to pass in the second time period. In this way, conflicts are reduced and the material circulation efficiency is improved. Integrate the path preemption permission, reallocation path planning, and passage time window adjustment together to form a material allocation parameter unit, which is an important part of the material priority allocation parameters.

[0060] Step S143: According to the sorting result of the compensation priorities for environmental disturbances, calculate the environmental adaptation compensation parameters, where the environmental adaptation compensation parameters include the increased number of heat dissipation devices in high-temperature areas, the dosing frequency of desiccants in areas with sudden humidity changes, and the working intensity of filtration devices in areas with high particulate matter concentrations.

[0061] In this embodiment, according to the sorting result of the compensation priority for environmental interference, for the high-temperature area, the influence coefficient of the high-temperature area at the corner of the current workshop on the heat dissipation efficiency of the equipment is 0.2. In order to reduce the impact of high temperature on the equipment, it is decided to add heat dissipation equipment. After comprehensively considering factors such as the area of this region and the heat generation of the equipment, it is calculated that 2 heat dissipation equipment need to be added to ensure that the equipment can operate in a suitable temperature environment.

[0062] For the area with sudden humidity change, the humidity increases from 50% to 70% in a short time, resulting in a 5% fluctuation range in the moisture content of the material. In order to stabilize the moisture content of the material, according to factors such as the area of the region and the humidity change rate, the desiccant dosing frequency is determined to be 3 times per hour. By regularly dosing the desiccant, the excess moisture in the air is absorbed to reduce the impact of humidity on the material.

[0063] For the area with high particulate concentration, the wear rate of the particulate concentration on the precision components of the equipment is 0.1 mm / day. In order to reduce the wear of the particulate matter on the equipment, the working intensity of the filtration equipment is increased. Originally, the filtration equipment runs for 20 minutes per hour, and now its working intensity is adjusted to run for 30 minutes per hour. By increasing the working time of the filtration equipment, the particulate matter in the air is filtered more effectively to protect the precision components of the equipment. After sorting out the number of additional heat dissipation equipment, the desiccant dosing frequency, and the working intensity of the filtration equipment, an environmental control parameter unit is formed as part of the environmental adaptation compensation parameters.

[0064] Step S144: Package the task splitting ratio, the task additional quantity, and the alternative equipment selection weight into an equipment scheduling parameter unit; package the path preemption permission, the re-allocated path planning, and the passage time window adjustment into a material distribution parameter unit; package the number of additional heat dissipation equipment, the desiccant dosing frequency, and the working intensity of the filtration equipment into an environmental control parameter unit; integrate the equipment scheduling parameter unit, the material distribution parameter unit, and the environmental control parameter unit in time series into the dynamic scheduling parameter set.

[0065] In this embodiment, the calculated task splitting ratio (30%), task additional quantity (20), and alternative equipment selection weight (0.6) are packaged to form an equipment scheduling parameter unit. This equipment scheduling parameter unit will be used as a whole for precise control of equipment scheduling in the future.

[0066] Package the path preemption permission (urgent order materials can preempt the path preferentially), the re-allocated path planning (plan a new path for the stacked materials), and the passage time window adjustment (pass through the conflict area in different time periods) into a material distribution parameter unit so that the system can reasonably allocate the flow path and time of the materials according to the actual situation of the materials.

[0067] Package the increased number of heat dissipation devices (2 units), the desiccant feeding frequency (3 times per hour), and the working intensity of the filtration device (running for 30 minutes per hour) into an environmental control parameter unit to provide accurate parameters for the environmental control system and achieve effective control of the workshop environment.

[0068] Finally, integrate the equipment scheduling parameter unit, the material distribution parameter unit, and the environmental control parameter unit according to the time series. For example, at the beginning of each scheduling cycle, first allocate and adjust tasks for the equipment according to the equipment scheduling parameter unit; then, plan and control the flow of materials based on the material distribution parameter unit; finally, start the corresponding environmental control equipment according to the environmental control parameter unit. Through this integration method according to the time series, a dynamic scheduling parameter set is formed, providing comprehensive and accurate data support for the subsequent optimization of the scheduling strategy.

[0069] Step S150: Call a preset scheduling strategy optimization algorithm to perform multi-objective optimization on the dynamic scheduling parameter set, generate an equipment scheduling instruction set, a material distribution instruction set, and an environmental control instruction set, and synchronize them to the physical workshop to perform real-time scheduling operations.

[0070] In this embodiment, step S150 may include: Step S151: Construct a multi-objective optimization function, and the objective variables of the multi-objective optimization function include the comprehensive equipment utilization rate, the total material flow time, and the environmental control energy consumption cost.

[0071] In this embodiment, the constructed multi-objective optimization function aims to balance the three key objective variables of the comprehensive equipment utilization rate, the total material flow time, and the environmental control energy consumption cost. The comprehensive equipment utilization rate refers to the ratio of the actual running time of the equipment to the available running time. It is hoped that through optimized scheduling, the equipment can be utilized as fully as possible to improve production efficiency. The total material flow time refers to the total time spent by the material from entering the workshop to completing production and leaving the workshop. It is necessary to strive to shorten the total material flow time and reduce the production cycle by reasonably planning the material path and time. The environmental control energy consumption cost refers to the energy cost consumed to maintain a suitable production environment in the workshop, such as adjusting temperature, humidity, filtering particulate matter, etc. It is necessary to reduce the environmental control energy consumption cost on the premise of meeting production requirements. For example, the goal of the comprehensive equipment utilization rate is set to be above 85%, the total material flow time is expected to be controlled within 3 hours, and the environmental control energy consumption cost is to be reduced by 10% on the current basis. By incorporating the above objective variables into the multi-objective optimization function, comprehensive optimization of production scheduling is achieved.

[0072] Step S152: Input the equipment scheduling parameter unit, the material allocation parameter unit and the environmental control parameter unit in the dynamic scheduling parameter set into the scheduling strategy optimization algorithm, and calculate the Pareto optimal solution set that satisfies the maximization of the comprehensive utilization rate of the equipment, the minimization of the total material circulation time and the minimization of the environmental control energy consumption cost through the constraint satisfaction algorithm.

[0073] In this embodiment, the packaged equipment scheduling parameter unit, material distribution parameter unit and environmental control parameter unit can be input into the preset scheduling strategy optimization algorithm. The scheduling strategy optimization algorithm uses the constraint satisfaction algorithm to calculate, and under the premise of meeting various production conditions and restrictions, it seeks the Pareto optimal solution set that can maximize the comprehensive utilization rate of the equipment, minimize the total material flow time and minimize the environmental control energy consumption cost. For example, during the calculation process, the algorithm will consider various constraints such as the capacity limit of the equipment, the processing sequence requirements of the materials, and the power limit of the environmental control equipment. Through the continuous adjustment and optimization of the above parameter units, a set of Pareto optimal solution sets is finally obtained. The Pareto optimal solution set contains multiple different solutions, each of which represents a scheduling scheme that balances the three target variables to varying degrees. For example, one of the solutions may make the comprehensive utilization rate of the equipment reach 88%, the total material flow time is 2.8 hours, and the environmental control energy consumption cost is reduced by 8%; another solution may make the comprehensive utilization rate of the equipment 86%, the total material flow time is 2.5 hours, and the environmental control energy consumption cost is reduced by 6%.

[0074] Step S153: Select an optimization solution that meets preset conditions from the Pareto optimal solution set, and map the optimization solution into an equipment scheduling instruction set, a material allocation instruction set and an environmental control instruction set; wherein the equipment scheduling instruction set includes equipment task diversion instructions, equipment task addition instructions and equipment alternative operation instructions; the material allocation instruction set includes path priority adjustment instructions, material reallocation instructions and conflict area time window allocation instructions; the environmental control instruction set includes heat dissipation equipment start and stop instructions, desiccant delivery instructions and filtering equipment power adjustment instructions.

[0075] In this embodiment, an optimized solution that best meets the production requirements can be selected from the Pareto optimal solution set according to preset conditions. The preset conditions may include focusing on a certain target variable or comprehensive consideration of overall performance. For example, if the current production task has a high requirement for the total material flow time and hopes to shorten the time as much as possible, then the solution with the shortest total material flow time and other target variables that can also meet the basic requirements will be selected.

[0076] Then, map the selected optimal solutions to generate specific instruction sets. For the equipment scheduling instruction set, generate equipment task diversion instructions according to the equipment scheduling parameters in the optimal solutions. For example, instruct equipment E-001 to divert 30% of the welding tasks to other equipment; generate equipment task append instructions, requiring equipment E-004 to append 20 tasks; if equipment E-001 fails, generate equipment alternative operation instructions, specifying that equipment E-005 replaces its operation with a weight of 60%.

[0077] For the material distribution instruction set, generate path priority adjustment instructions according to the material distribution parameters in the optimal solutions, giving emergency order material paths preemption rights; generate material redistribution instructions to plan new paths for the materials piled up at the second assembly process node; generate conflict area time window allocation instructions to adjust the passing time of the material flow path intersection conflict area.

[0078] For the environmental control instruction set, generate heat dissipation equipment start / stop instructions according to the environmental control parameters in the optimal solutions, starting 2 heat dissipation equipment in high-temperature areas; generate desiccant injection instructions, specifying that the desiccant is injected 3 times per hour in areas with sudden humidity changes; generate filtration equipment power adjustment instructions to adjust the working intensity of the filtration equipment in areas with high particulate concentrations to run for 30 minutes per hour.

[0079] Step S154: Send the equipment task diversion instructions, the equipment task append instructions, and the equipment alternative operation instructions to the controllers of the corresponding equipment through the industrial Internet of Things communication protocol to trigger the adjustment of equipment operation parameters; send the path priority adjustment instructions, the material redistribution instructions, and the conflict area time window allocation instructions to the automatic guided vehicle and the robotic arm through the material scheduling system to trigger the real-time change of the material path; send the heat dissipation equipment start / stop instructions, the desiccant injection instructions, and the filtration equipment power adjustment instructions to the environmental execution agency through the environmental control system to trigger the adaptive adjustment of environmental parameters.

[0080] In this embodiment, through the industrial Internet of Things communication protocol, the equipment task diversion instructions, the equipment task append instructions, and the equipment alternative operation instructions are accurately sent to the controllers of the corresponding equipment. For example, send the task diversion instruction of equipment E-001 to its controller. After receiving the instruction, the controller automatically adjusts the task allocation settings of the equipment and distributes 30 out of the original 100 battery cell welding tasks per hour. After receiving the task append instruction, the controller of equipment E-004 increases its own task volume and starts to process 80 tasks per hour. If equipment E-001 fails, the controller of equipment E-005 receives the equipment alternative operation instruction, quickly starts and adjusts the operation parameters, and starts to execute part of the tasks originally undertaken by equipment E-001.

[0081] Moreover, by using the material scheduling system, the path priority adjustment instruction, the material reallocation instruction, and the conflict area time window allocation instruction are sent to the automated guided vehicle and the robotic arm. After receiving the path priority adjustment instruction, for the materials of the emergency order, the automated guided vehicle preferentially selects the preempted path for transportation; after receiving the material reallocation instruction, it transports the materials piled up at the second assembly process node according to the newly planned path. After receiving the conflict area time window allocation instruction, the robotic arm grabs and transports materials within a suitable time period according to the adjusted time window, realizing the real-time change of the material path.

[0082] Meanwhile, through the environmental control system, the heat dissipation device start-stop instruction, the desiccant delivery instruction, and the filter device power adjustment instruction are transmitted to the environmental execution mechanism. In the high-temperature area, after receiving the heat dissipation device start-stop instruction, the environmental execution mechanism immediately starts 2 heat dissipation devices to reduce the area temperature. In the area with sudden humidity change, according to the desiccant delivery instruction, the execution mechanism delivers desiccant 3 times per hour to adjust the humidity. In the area with high particulate concentration, according to the filter device power adjustment instruction, the execution mechanism adjusts the working intensity of the filter device to run for 30 minutes per hour, effectively filtering the particulate matter in the air and realizing the adaptive adjustment of the environmental parameters.

[0083] Step S155: In the three-dimensional digital twin model, the device operation status, the material flow path, and the environmental parameter changes are updated in real time, and the actual response data of the physical workshop is compared with the expected effect of the instruction set to generate a scheduling deviation correction coefficient; when the scheduling deviation correction coefficient exceeds the preset threshold, the generation and optimization process of the dynamic scheduling parameter set are triggered again.

[0084] In this embodiment, in the three-dimensional digital twin model, the device operation status, the material flow path, and the environmental parameter changes are updated in real time. For example, in terms of the device operation status, according to the actual task execution situation of the device, such as after the task of device E-001 is split, its task progress and energy consumption in the model are updated in real time; in terms of the material flow path, as the materials are transported along the new path, the real-time position and transfer trajectory of the materials are accurately displayed in the model; in terms of the environmental parameter changes, according to the regulation results of the environmental execution mechanism, the data such as the temperature, humidity, and particulate concentration in the workshop are updated in the model display.

[0085] Moreover, the actual response data of the physical workshop can be compared with the expected effects of the instruction set to generate a scheduling deviation correction coefficient. For example, obtain the actual equipment utilization rate after the physical workshop executes the equipment scheduling instruction set, the actual material flow duration after executing the material allocation instruction set, and the actual energy consumption cost after executing the environmental control instruction set. Assume that the expected overall equipment utilization rate is 88%, and the actual equipment utilization rate is 85%. Calculate the absolute value of the difference between the actual equipment utilization rate and the expected overall equipment utilization rate, and the equipment scheduling deviation value is 3%. The expected total material flow duration is 2.8 hours, and the actual material flow duration is 3 hours. Calculate the percentage difference between the actual material flow duration and the expected total material flow duration, and the material scheduling deviation value is (3 - 2.8) / 2.8 ≈ 7.1%. The expected environmental control energy consumption cost is reduced by 8%, and the actual environmental control energy consumption cost is reduced by 6%. Calculate the ratio of the actual energy consumption cost to the expected environmental control energy consumption cost, and the environmental control deviation value is 6% / 8% = 0.75. Perform a weighted sum of the equipment scheduling deviation value, the material scheduling deviation value, and the environmental control deviation value. Assume that the weight of the equipment scheduling deviation value is 0.4, the weight of the material scheduling deviation value is 0.3, and the weight of the environmental control deviation value is 0.3. The generated scheduling deviation correction coefficient is 3% × 0.4 + 7.1% × 0.3 + 0.75 × 0.3 = 0.012 + 0.0213 + 0.225 = 0.2583.

[0086] Furthermore, when the scheduling deviation correction coefficient exceeds a preset threshold (assuming the preset threshold is 0.2), it indicates that there is a large deviation in the current scheduling plan, and it is necessary to re-trigger the generation and optimization of the dynamic scheduling parameter set. The system will automatically re-collect the real-time monitoring data of the physical workshop, generate a three-dimensional digital twin model again, conduct an operation state correlation analysis, generate a new dynamic scheduling parameter set, and re-perform multi-objective optimization to adjust the scheduling plan to make it more in line with production requirements and improve production efficiency and quality.

[0087] Furthermore, in a possible implementation manner, the above method may further include: Step S210: Statistically analyze the comprehensive performance indicators after the physical workshop executes the historical scheduling instruction set within a preset period. The comprehensive performance indicators include the mean time between failures of equipment, the on-time delivery rate of materials, and the environmental control cost per unit of output.

[0088] For example, within a preset one - month cycle, the comprehensive efficiency index after the physical workshop executes the historical scheduling instruction set is statistically analyzed. For the mean time between failures (MTBF) of equipment, the failure occurrence times of all equipment in the workshop are statistically analyzed. For example, equipment E - 001 had 2 failures in the past month. The first failure occurred on the 5th day and the second failure occurred on the 20th day. Then the time intervals between failures of equipment E - 001 are 15 days and 10 days respectively, and the calculated MTBF is (15 + 10) / 2 = 12.5 days. The same statistical analysis and calculation are performed on other equipment in the workshop, and finally the MTBF of all equipment in the workshop is obtained.

[0089] Regarding the on - time delivery rate of materials, the delivery situations of all orders are statistically analyzed. Suppose that within the past month, a total of 50 orders were received, and 45 of them were delivered on time. Then the on - time delivery rate of materials is 45 / 50 = 90%.

[0090] The environmental control cost per unit of production is the cost consumed for environmental control during the production of a certain quantity of products. For example, within the past month, 1000 battery modules were produced, and the total energy cost consumed by environmental control equipment (such as heat dissipation equipment, desiccant dispensing equipment, filtration equipment, etc.) was 5000 yuan. Then the environmental control cost per unit of production is 5000 / 1000 = 5 yuan per unit. The above - mentioned comprehensive efficiency indexes are sorted out and recorded to provide data support for subsequent parameter optimization.

[0091] Step S220: Compare the comprehensive efficiency index with the prior benchmark data to generate the intensity of parameter optimization requirements; when the intensity of parameter optimization requirements exceeds the preset threshold, adjust the generation weights of the equipment load balancing parameter, material priority allocation parameter, and environment adaptation compensation parameter in the dynamic scheduling parameter set; based on the adjusted generation weights, re - execute the operation state correlation analysis and multi - objective optimization processing to generate an updated scheduling instruction set and synchronize it to the physical workshop.

[0092] In this embodiment, the statistically obtained comprehensive efficiency index can be compared in detail with the prior benchmark data. The prior benchmark data is the ideal index value set according to a large amount of past production experience and industry standards.

[0093] For the mean time between failures of equipment, the prior benchmark data is set to 15 days. The currently statistically obtained mean time between failures of workshop equipment is 12.5 days. By comparison, it can be seen that the actual value is lower than the benchmark value, indicating that the stability of the equipment needs to be improved. Calculate the gap ratio, that is, (15 - 12.5) / 15≈16.7%.

[0094] The prior benchmark data for the on-time delivery rate of materials is 95%, while the actual on-time delivery rate of materials statistically is 90%. The percentage difference is (95% - 90%) / 95% ≈ 5.3%, indicating that there is still room for optimization in material scheduling and production coordination.

[0095] The prior benchmark data for the environmental control cost per unit output is 4 yuan per unit, and the actual statistical value is 5 yuan per unit. The difference amplitude is (5 - 4) / 4 = 25%, indicating that the current environmental control cost is relatively high and needs to be optimized.

[0096] Based on the above comparison results, the intensity requirement for parameter optimization is generated through specific algorithms and rules. Suppose the weight of the difference ratio of the mean time between failures of the equipment is set to 0.4, the weight of the percentage difference in the on-time delivery rate of materials is set to 0.3, and the weight of the difference amplitude of the environmental control cost per unit output is set to 0.3. Calculate the intensity requirement for parameter optimization as 16.7% × 0.4 + 5.3% × 0.3 + 25% × 0.3 = 0.0668 + 0.0159 + 0.075 = 0.1577.

[0097] Among them, the preset threshold is set to 0.1. Since the current intensity requirement for parameter optimization, 0.1577, exceeds the preset threshold, it is necessary to adjust the generation weights of the parameters in the dynamic scheduling parameter set.

[0098] For example, in a possible implementation manner, step S220 may include: Step S221: According to the difference ratio between the mean time between failures of the equipment and the prior benchmark, increase the calculation proportion of the selection weight of the alternative equipment in the equipment load balancing parameter; For example, since there is a 16.7% difference ratio between the mean time between failures of the equipment and the prior benchmark, in order to improve the stability of the equipment operation and reduce the impact of equipment failures on production, it is decided to increase the calculation proportion of the selection weight of the alternative equipment in the equipment load balancing parameter. The original calculation proportion of the selection weight of the alternative equipment in the overall calculation of the equipment load balancing parameter is 30%. Now, according to the difference ratio, it is increased to 30% + 16.7% × 0.5 = 30% + 8.35% = 38.35%. This means that in the subsequent calculation of the equipment load balancing parameter, the importance of the selection weight of the alternative equipment in the overall consideration increases, and the system will evaluate more carefully and comprehensively when selecting alternative equipment to ensure more effective replacement in case of equipment failures and reduce production interruption time.

[0099] Step S222: According to the percentage difference between the on-time delivery rate of materials and the prior benchmark, increase the priority coefficient of the path preemption permission in the material priority allocation parameter; For example, there is a percentage difference of 5.3% between the on-time delivery rate of materials and the prior benchmark. To improve the on-time delivery rate of materials, the priority coefficient of the path preemption privilege among the material priority allocation parameters is adjusted. The original priority coefficient of the path preemption privilege was 1, and now it is increased to 1 + 5.3%×0.8 = 1 + 0.0424 = 1.0424. In the subsequent scheduling process, when there are situations such as emergency orders that require rapid material flow, materials with the path preemption privilege will obtain path resources more preferentially, reducing delays caused by path conflicts and other problems, thereby increasing the likelihood of on-time material delivery.

[0100] Step S223: According to the difference amplitude between the unit output environmental regulation cost and the prior benchmark, reduce the adjustment step size of the working intensity of the filtering equipment in the environmental adaptation compensation parameter; For example, there is a difference amplitude of 25% between the unit output environmental regulation cost and the prior benchmark. To reduce the environmental regulation cost, the adjustment step size of the working intensity of the filtering equipment in the environmental adaptation compensation parameter is adjusted. The original adjustment step size of the working intensity of the filtering equipment was to change by 5 working intensity units each time it was adjusted. Now, the adjustment step size is reduced to 5 - 5×25%×0.6 = 5 - 0.75 = 4.25 working intensity units. This means that when the working intensity of the filtering equipment is adjusted subsequently, the adjustment amplitude each time becomes smaller, avoiding energy waste caused by over-adjustment, and reducing the energy consumption cost of environmental regulation while ensuring the environmental regulation effect.

[0101] Step S224: Write the adjusted calculation ratio, priority coefficient, and adjustment step size into the weight configuration file of the scheduling policy optimization algorithm to take effect in subsequent optimization processing; For example, write the adjusted alternative equipment selection weight calculation ratio of 38.35%, the path preemption privilege priority coefficient of 1.0424, and the filtering equipment working intensity adjustment step size of 4.25 accurately into the weight configuration file of the scheduling policy optimization algorithm. This weight configuration file is an important part of the scheduling policy optimization algorithm, which determines the relative importance and action mode of each parameter in subsequent calculations and optimization processes. After writing the above adjusted parameters, when the running state correlation analysis and multi-objective optimization processing are executed again, the scheduling policy optimization can calculate according to the new weight settings, thereby generating a scheduling plan that better meets the current production requirements.

[0102] Step S225: Based on the adjusted generated weights, re-execute the running state correlation analysis and multi-objective optimization processing, generate an updated set of scheduling instructions and synchronize them to the physical workshop; In this embodiment, based on the adjusted generation weights, the execution of the running state correlation analysis is restarted. The device operation efficiency feature set, material flow bottleneck feature set, and environmental interference feature set are extracted again from the three-dimensional digital twin model. For example, the device operation efficiency features such as the production capacity matching deviation rate between adjacent process devices, the proportion of device idle time, and the device failure warning index are recalculated; the material flow bottleneck features such as the material accumulation duration when the material accumulation amount at the process node exceeds the threshold, the number of path intersections in the material transfer path intersection conflict area, and the overtime ratio of materials waiting for processing are re-analyzed; and the environmental interference features such as the influence coefficient of the high-temperature area on the device heat dissipation efficiency, the fluctuation range of the material moisture content in the humidity mutation area, and the wear rate of the device precision components by the particulate matter concentration are re-determined.

[0103] Next, the above updated feature sets are used for multi-objective optimization processing. The multi-objective optimization function is constructed again, with the device comprehensive utilization rate, total material transfer duration, and environmental control energy consumption cost as the target variables. The updated device scheduling parameter unit, material distribution parameter unit, and environmental control parameter unit are input into the scheduling strategy optimization algorithm, and the Pareto optimal solution set that maximizes the device comprehensive utilization rate, minimizes the total material transfer duration, and minimizes the environmental control energy consumption cost is calculated through the constraint satisfaction algorithm.

[0104] An optimization solution that meets the preset conditions is selected from the new Pareto optimal solution set and mapped to the device scheduling instruction set, material distribution instruction set, and environmental control instruction set. For example, new device task diversion instructions, device task append instructions, and device alternative operation instructions are generated; new path priority adjustment instructions, material re-distribution instructions, and conflict area time window allocation instructions; and new heat dissipation device start-stop instructions, desiccant injection instructions, and filter device power adjustment instructions.

[0105] Finally, through the industrial Internet of Things communication protocol, the material scheduling system, and the environmental control system, the above updated scheduling instruction set is synchronized to the physical workshop. After receiving the new instructions, the device controllers, automatic guided vehicles, robotic arms, and environmental execution mechanisms in the physical workshop adjust their operating parameters and working modes according to the instruction requirements, realizing the optimization and adjustment of production scheduling, so as to improve the production efficiency, product delivery rate of the workshop, and reduce the environmental control cost, making the entire production process more efficient, stable, and economical.

[0106] Figure 2 FIG. shows a schematic diagram of exemplary hardware and software components of a digital twin-based intelligent factory production scheduling system 100 that can implement the ideas of the present application provided by some embodiments of the present application. For example, the processor 120 can be used on the digital twin-based intelligent factory production scheduling system 100 and is used to execute the functions in the present application.

[0107] The intelligent factory production scheduling system 100 based on digital twin can be a general-purpose server or a special-purpose server, both of which can be used to implement the intelligent factory production scheduling method based on digital twin of this application. Although only one server is shown in this application, for convenience, the functions described in this application can be implemented in a distributed manner on multiple similar platforms to balance the processing load.

[0108] For example, the intelligent factory production scheduling system 100 based on digital twin can include a network port 110 connected to a network, one or more processors 120 for executing program instructions, a communication bus 130, and different forms of storage media 140, such as disks, ROM, or RAM, or any combination thereof. Exemplarily, the intelligent factory production scheduling system 100 based on digital twin can also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of this application can be implemented according to the above program instructions. The intelligent factory production scheduling system 100 based on digital twin also includes an input / output (I / O) interface 150 between the computer and other input / output devices.

[0109] For ease of illustration, only one processor is described in the intelligent factory production scheduling system 100 based on digital twin. However, it should be noted that the intelligent factory production scheduling system 100 based on digital twin in this application can also include multiple processors. Therefore, the steps executed by one processor described in this application can also be jointly executed or separately executed by multiple processors. For example, if the processor of the intelligent factory production scheduling system 100 based on digital twin executes step A and step B, it should be understood that step A and step B can also be jointly executed by two different processors or separately executed in one processor. For example, the first processor executes step A, the second processor executes step B, or the first processor and the second processor jointly execute steps A and B.

[0110] In addition, an embodiment of the present invention also provides a readable storage medium, in which computer-executable instructions are preset. When the processor executes the computer-executable instructions, the intelligent factory production scheduling method based on digital twin as described above is implemented.

[0111] It should be noted that, in order to simplify the expression of the disclosure of the present invention and thus help the understanding of one or more embodiments of the invention, in the foregoing description of the embodiments of the present invention, sometimes multiple features are merged into one embodiment, drawing, or description thereof.

Claims

1. An intelligent factory production scheduling method based on digital twin, characterized in that, The method includes: Obtaining a real-time monitoring data set of the physical workshop, where the real-time monitoring data set includes equipment operation status data, material flow trajectory data, and environmental sensing data; Generating a three-dimensional digital twin model of the physical workshop based on the real-time monitoring data set, where the three-dimensional digital twin model includes an equipment dynamic operation topology, a material real-time distribution heat map, and an environmental state simulation layer; Performing an operation status correlation analysis on the three-dimensional digital twin model to generate an equipment operation efficiency feature set, a material flow bottleneck feature set, and an environmental interference feature set; Generating a dynamic scheduling parameter set based on the equipment operation efficiency feature set, the material flow bottleneck feature set, and the environmental interference feature set, where the dynamic scheduling parameter set includes equipment load balancing parameters, material priority allocation parameters, and environmental adaptation compensation parameters; Invoking a preset scheduling strategy optimization algorithm to perform multi-objective optimization processing on the dynamic scheduling parameter set, generating an equipment scheduling instruction set, a material allocation instruction set, and an environmental regulation instruction set, and synchronizing them to the physical workshop to perform real-time scheduling operations.

2. The intelligent factory production scheduling method based on digital twin according to claim 1, wherein, The obtaining of the real-time monitoring data set of the physical workshop includes: Collecting equipment real-time energy consumption data, equipment vibration spectrum data, and equipment temperature change data through edge sensors deployed at the equipment end of the physical workshop, and associating the equipment real-time energy consumption data, the equipment vibration spectrum data, and the equipment temperature change data with the unique equipment identifier; Obtaining the real-time position coordinate sequence of the material in the physical workshop through a material positioning tag, and calculating the residence time interval and the moving speed change rate of the material between adjacent processes based on the real-time position coordinate sequence; Collecting temperature distribution data, humidity gradient data, and air particulate matter concentration data in the physical workshop through an environmental monitoring device, and mapping the temperature distribution data, the humidity gradient data, and the air particulate matter concentration data to the workshop space grid coordinates based on the time stamp; Aggregating the equipment real-time energy consumption data, the equipment vibration spectrum data, and the equipment temperature change data associated with the unique equipment identifier, the residence time interval and the moving speed change rate, and the temperature distribution data, the humidity gradient data, and the air particulate matter concentration data associated with the workshop space grid coordinates into the real-time monitoring data set according to a preset time window.

3. The intelligent factory production scheduling method based on digital twin according to claim 2, wherein The generating of the three-dimensional digital twin model of the physical workshop based on the real-time monitoring data set includes: Matching the pre-defined three-dimensional geometric model and physical attribute parameters of the equipment according to the unique equipment identifier, and updating the dynamic operation status of the three-dimensional geometric model based on the equipment real-time energy consumption data, the equipment vibration spectrum data, and the equipment temperature change data to generate the equipment dynamic operation topology; Overlaying the real-time position coordinate sequence with the process layout map, calculating the aggregation density and the flow path deviation degree of the material at the process nodes, and generating the material real-time distribution heat map with color gradient identification based on the aggregation density and the flow path deviation degree; Construct an environmental state surface equation based on the temperature distribution data, the humidity gradient data, and the air particulate matter concentration data associated with the workshop space grid coordinates, and generate the environmental state simulation layer covering the entire physical workshop through an interpolation algorithm; Fuse the device dynamic operation topology, the real-time material distribution heat map, and the environmental state simulation layer in a unified spatial coordinate system, and render and generate the three-dimensional digital twin model in a visualization interface.

4. The intelligent factory production scheduling method based on digital twin according to claim 1, characterized in that, Perform an operation state correlation analysis on the three-dimensional digital twin model to generate a set of device operation efficiency characteristics, a set of material flow bottleneck characteristics, and a set of environmental interference characteristics, including: Extract a set of device operation efficiency characteristics from the device dynamic operation topology, and the set of device operation efficiency characteristics includes the production capacity matching deviation rate between adjacent process devices, the proportion of device idle time, and the device fault warning index; Extract a set of material flow bottleneck characteristics from the real-time material distribution heat map, and the set of material flow bottleneck characteristics includes the material accumulation duration when the material accumulation amount at the process node exceeds the threshold, the number of path crossing conflicts in the material transfer path crossing conflict area, and the overtime ratio of materials waiting for processing; Extract a set of environmental interference characteristics from the environmental state simulation layer, and the set of environmental interference characteristics includes the influence coefficient of the high-temperature area on the device heat dissipation efficiency, the fluctuation range of the material moisture content in the humidity mutation area, and the wear rate of the device precision components by the particulate matter concentration; Compare the production capacity matching deviation rate, the proportion of device idle time, and the device fault warning index with the historical operation baseline to generate the root cause analysis result of the device operation efficiency decline; Match the material accumulation duration, the number of path crossing conflicts, and the overtime ratio with the preset flow optimization rules to generate the grading result of the material flow bottleneck; Perform an associated query on the influence coefficient, the fluctuation range, and the wear rate with the environmental control strategy library to generate the compensation priority sorting result of the environmental interference.

5. The intelligent factory production scheduling method based on digital twin according to claim 1, characterized in that, Generate a set of dynamic scheduling parameters based on the set of device operation efficiency characteristics, the set of material flow bottleneck characteristics, and the set of environmental interference characteristics, including: Calculate the device load balancing parameters according to the root cause analysis result of the device operation efficiency decline, and the device load balancing parameters include the task diversion ratio of high-load devices, the task addition amount of low-load devices, and the replacement device selection weight of faulty devices; Calculate the material priority allocation parameters according to the grading result of the material flow bottleneck, and the material priority allocation parameters include the path preemption permission of emergency order materials, the reallocation path planning of accumulated materials, and the adjustment of the material passage time window in the conflict area; Calculate the environmental adaptation compensation parameters according to the compensation priority sorting result of the environmental interference, and the environmental adaptation compensation parameters include the number of additional heat dissipation devices in the high-temperature area, the desiccant injection frequency in the humidity mutation area, and the working intensity of the filtration devices in the high particulate matter concentration area; Package the task diversion ratio, the task addition amount, and the replacement device selection weight into a device scheduling parameter unit; Encapsulate the path preemption permission, the reallocation path planning, and the passage time window adjustment into a material distribution parameter unit; Encapsulate the increased quantity of the heat dissipation device, the desiccant feeding frequency, and the working intensity of the filtering device into an environmental control parameter unit; Integrate the device scheduling parameter unit, the material distribution parameter unit, and the environmental control parameter unit into the dynamic scheduling parameter set according to the time series; 6. The intelligent factory production scheduling method based on digital twin according to claim 5, characterized in that, Call the preset scheduling strategy optimization algorithm to perform multi-objective optimization on the dynamic scheduling parameter set, including: Construct a multi-objective optimization function, where the objective variables of the multi-objective optimization function include the comprehensive utilization rate of devices, the total material flow time, and the environmental control energy consumption cost; Input the device scheduling parameter unit, the material distribution parameter unit, and the environmental control parameter unit in the dynamic scheduling parameter set into the scheduling strategy optimization algorithm, and calculate the Pareto optimal solution set that maximizes the comprehensive utilization rate of devices, minimizes the total material flow time, and minimizes the environmental control energy consumption cost through the constraint satisfaction algorithm; Select the optimized solutions that meet the preset conditions from the Pareto optimal solution set, and map the optimized solutions to the device scheduling instruction set, the material distribution instruction set, and the environmental control instruction set; Among them, the device scheduling instruction set includes device task diversion instructions, device task append instructions, and device alternative operation instructions; The material distribution instruction set includes path priority adjustment instructions, material reallocation instructions, and conflict area time window allocation instructions; The environmental control instruction set includes heat dissipation device start / stop instructions, desiccant feeding instructions, and filtering device power adjustment instructions; 7. The intelligent factory production scheduling method based on digital twin according to claim 6, characterized in that, Synchronize to the physical workshop to perform real-time scheduling operations, including: Send the device task diversion instructions, the device task append instructions, and the device alternative operation instructions to the controllers of the corresponding devices through the industrial Internet of Things communication protocol to trigger the adjustment of device operation parameters; Send the path priority adjustment instructions, the material reallocation instructions, and the conflict area time window allocation instructions to the automatic guided vehicle and the robotic arm through the material scheduling system to trigger the real-time change of the material path; Send the heat dissipation device start / stop instructions, the desiccant feeding instructions, and the filtering device power adjustment instructions to the environmental execution mechanism through the environmental control system to trigger the adaptive adjustment of environmental parameters; In the three-dimensional digital twin model, update the device operation status, the material flow path, and the environmental parameter changes in real time, and compare the actual response data of the physical workshop with the expected effects of the instruction set to generate a scheduling deviation correction coefficient; When the scheduling deviation correction coefficient exceeds the preset threshold, re-trigger the generation and optimization of the dynamic scheduling parameter set; 8. The intelligent factory production scheduling method based on digital twin according to claim 7, wherein, Compare the actual response data of the physical workshop with the expected effects of the instruction set to generate a scheduling deviation correction coefficient, including: Obtain the actual device utilization rate after the physical workshop executes the device scheduling instruction set, the actual material flow time after executing the material distribution instruction set, and the actual energy consumption cost after executing the environmental control instruction set; Calculate the absolute value of the difference between the actual equipment utilization rate and the expected overall equipment utilization rate to obtain the equipment scheduling deviation value; Calculate the percentage of the difference between the actual material flow duration and the expected total material flow duration to obtain the material scheduling deviation value; Calculate the ratio of the actual energy consumption cost to the expected energy consumption cost for environmental control to obtain the environmental control deviation value; Perform a weighted sum of the equipment scheduling deviation value, the material scheduling deviation value, and the environmental control deviation value to generate the scheduling deviation correction coefficient.

9. The intelligent factory production scheduling method based on digital twin according to claim 1, characterized in that The method further includes: Statistically analyze the comprehensive performance indicators after the physical workshop executes the historical scheduling instruction set within a preset period. The comprehensive performance indicators include the mean time between failures of equipment, the on-time delivery rate of materials, and the environmental control cost per unit output; Compare the comprehensive performance indicators with the prior benchmark data to generate the intensity of the parameter optimization requirement; When the intensity of the parameter optimization requirement exceeds a preset threshold, adjust the generation weights of the equipment load balancing parameter, the material priority allocation parameter, and the environmental adaptation compensation parameter in the dynamic scheduling parameter set; Based on the adjusted generation weights, re-execute the operation state correlation analysis and the multi-objective optimization process to generate an updated scheduling instruction set and synchronize it to the physical workshop.

10. An intelligent factory production scheduling system based on digital twin, characterized in that, It includes a processor and a memory. The memory is connected to the processor. The memory is used to store programs, instructions, or codes. The processor is used to execute the programs, instructions, or codes in the memory to implement the digital twin-based intelligent factory production scheduling method according to any one of claims 1-9 above.

Images