Real-time dynamic scheduling system and method for industrial production line
By integrating multi-source data and intelligent algorithms, a real-time dynamic scheduling system is constructed, which solves the problems of insufficient dynamic adaptability and low collaborative efficiency of traditional production line scheduling systems when facing real-time changes, and realizes efficient utilization of production line resources and process optimization.
Patent Information
- Application Number
- CN202511095350.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-06
- Publication Date
- 2025-11-21
AI Technical Summary
Traditional industrial production line scheduling systems lack dynamic adaptability, have low collaborative efficiency, and fail to fully utilize data when facing real-time changes in the production environment, making it difficult to quickly identify and resolve production bottlenecks.
By integrating multi-source data and intelligent algorithms, a data acquisition and preprocessing module, a dynamic evaluation and decision-making module, a collaborative control and execution module, and a digital twin monitoring module are constructed to achieve real-time dynamic scheduling. Combined with equipment status assessment models, task allocation algorithms, and path planning engines, equipment availability assessment and robot motion trajectory planning are performed, and production status is monitored in real time and self-optimization adjustments are made.
It improves the dynamic adaptability and collaborative efficiency of the production line, reduces production downtime, increases production efficiency by 15%-20%, and improves equipment resource utilization and material flow efficiency.
Smart Images
Figure CN120993843A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of industrial automation control technology, and in particular to a real-time dynamic scheduling system and method for industrial production lines, applicable to production line scenarios requiring high-precision baking processes, such as battery manufacturing and electronic component processing. Background Technology
[0002] Traditional industrial production line scheduling systems rely on preset rules or offline optimization schemes, making it difficult to cope with real-time changes in the production environment (such as equipment failures, order changes, and material shortages). Existing technologies mainly suffer from the following problems:
[0003] Insufficient dynamic adaptability: The lack of a real-time perception and linkage adjustment mechanism for equipment status, material flow, and order priority makes it difficult to quickly identify and resolve production bottlenecks.
[0004] Low collaboration efficiency: When multiple workstations and multiple robots work together, they rely on fixed path planning and do not consider real-time obstacle detection and dynamic obstacle avoidance, which can easily lead to equipment collisions or process blockages.
[0005] Insufficient data utilization: The data from IoT sensor data, Manufacturing Execution System (MES) data, and historical equipment operation and maintenance data are not effectively integrated, resulting in a lack of comprehensiveness and accuracy in scheduling decisions.
[0006] Therefore, it is necessary to invent a real-time dynamic scheduling system and method for industrial production lines to solve the above problems. Summary of the Invention
[0007] The purpose of this invention is to provide a real-time dynamic scheduling system and method for industrial production lines. Through multi-source data fusion and intelligent algorithms, the real-time dynamic scheduling of industrial production lines is realized, which effectively improves production efficiency and flexibility, thereby solving the problems of insufficient dynamic adaptability, low collaborative efficiency and insufficient data utilization in the prior art mentioned in the background.
[0008] To achieve the above objectives, the present invention provides the following technical solution: a real-time dynamic scheduling system for industrial production lines, comprising...
[0009] Data acquisition and preprocessing module: includes an IoT sensor cluster and an edge computing unit, to realize real-time acquisition, preprocessing and real-time transmission of production data to the cloud server;
[0010] Dynamic evaluation and decision-making module: integrates equipment status evaluation model, task allocation algorithm and path planning engine, calculates equipment availability based on real-time data, and generates optimal task allocation scheme and robot motion trajectory;
[0011] Collaborative control and execution module: including equipment controller and robot control cabinet, receives and executes scheduling instructions, and simultaneously feeds back the equipment status to the data acquisition and preprocessing module to form closed-loop control;
[0012] Digital twin monitoring module: Constructs a 3D virtual model of the production line, displays the production status in real time, provides a visual monitoring interface, and supports manual intervention and strategy adjustment.
[0013] This invention also discloses a real-time dynamic scheduling method for industrial production lines, specifically including the following steps:
[0014] S1. Real-time acquisition and preprocessing of multi-source data:
[0015] Using IoT sensors (such as RFID, vision cameras, and pressure sensors), data such as equipment status (speed, temperature, energy consumption), material location, and workstation load are collected in real time. The data is then processed by edge computing nodes to perform noise reduction and normalization, resulting in a standardized dataset D = {d1, d2, ..., dn}, where di represents the data of the i-th type of sensor.
[0016] The data normalization formula is: di′=max(di)-min(di)di-min(di)
[0017] In the formula, di′ represents the normalized data, which solves the problem of merging data with different dimensions;
[0018] S2. Dynamic Equipment Status Assessment and Task Allocation:
[0019] Construct a device condition assessment model, and calculate the overall availability Sk of the device by comprehensively considering parameters such as the current load Lk, historical failure rate Fk, and energy efficiency Ek.
[0020] Sk=α·(1-Lk)+β·(1-Fk)+γ·Ek
[0021] Among them, α, β and γ are weighting coefficients and α+β+γ=1. They are dynamically adjusted according to the production process requirements, and tasks are given priority to be assigned to the equipment with the highest Sk value to ensure efficient use of resources.
[0022] S3, Multi-robot cooperative path planning and obstacle avoidance:
[0023] Based on the improved A* algorithm and combined with real-time obstacle detection data, the robot's motion trajectory is planned; the path cost function C is defined as:
[0024] C=λ1·Ddistance+λ2·Dobstacle+λ3·Twait
[0025] In the formula, Ddistance is the geometric distance of the path, Dobstacle is the obstacle influence factor (0 when there are no obstacles, and the value increases as the obstacle is closer), Twait is the waiting time at the workstation; λ1, λ2 and λ3 are cost weights, which are dynamically optimized through real-time production cycle time.
[0026] S4. Real-time production process monitoring and dynamic adjustment:
[0027] Establish a digital twin model of the production process to map the status of each workstation, material flow trajectory and equipment operating parameters in real time. Through preset production bottleneck identification rules (such as workstation load continuously exceeding 80% for 10 minutes), the scheduling strategy is automatically triggered to adjust, tasks are reassigned or equipment operating parameters are adjusted to achieve self-optimization of the production process.
[0028] The technical effects and advantages of this invention are as follows:
[0029] Highly adaptable to dynamic events: Through real-time data-driven scheduling strategies, it can quickly respond to dynamic events such as equipment failures and order changes, reducing production downtime.
[0030] High collaborative efficiency: Multi-robot collaborative path planning combined with real-time obstacle avoidance avoids equipment collisions and improves material flow efficiency.
[0031] Data-driven decision-making: By integrating multi-source data to build an evaluation model, scheduling decisions become more scientific, effectively reducing equipment energy consumption and failure rate, and improving production efficiency by 15%-20%.
[0032] In summary, this invention achieves real-time dynamic scheduling of industrial production lines through multi-source data fusion and intelligent algorithms, effectively improving production efficiency and flexibility. Attached Figure Description
[0033] Figure 1 This is a schematic diagram of the system structure of the present invention;
[0034] Figure 2 This is a schematic diagram of the method flow of the present invention. Detailed Implementation
[0035] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0036] Example 1
[0037] This invention provides, for example Figure 1The illustrated industrial production line real-time dynamic scheduling system includes
[0038] Data acquisition and preprocessing module: includes an IoT sensor cluster and an edge computing unit, to realize real-time acquisition, preprocessing and real-time transmission of production data to the cloud server;
[0039] Dynamic evaluation and decision-making module: integrates equipment status evaluation model, task allocation algorithm and path planning engine, calculates equipment availability based on real-time data, and generates optimal task allocation scheme and robot motion trajectory;
[0040] Collaborative control and execution module: including equipment controller and robot control cabinet, receives and executes scheduling instructions, and simultaneously feeds back the equipment status to the data acquisition and preprocessing module to form closed-loop control;
[0041] Digital twin monitoring module: Constructs a 3D virtual model of the production line, displays the production status in real time, provides a visual monitoring interface, and supports manual intervention and strategy adjustment.
[0042] This invention provides, for example Figure 2 The method for real-time dynamic scheduling of an industrial production line, as shown, specifically includes the following steps:
[0043] S1. Real-time acquisition and preprocessing of multi-source data:
[0044] Using IoT sensors (such as RFID, vision cameras, and pressure sensors), data such as equipment status (speed, temperature, energy consumption), material location, and workstation load are collected in real time. The data is then processed by edge computing nodes to perform noise reduction and normalization, resulting in a standardized dataset D = {d1, d2, ..., dn}, where di represents the data of the i-th type of sensor.
[0045] The data normalization formula is: di′=max(di)-min(di)di-min(di)
[0046] In the formula, di′ represents the normalized data, which solves the problem of merging data with different dimensions;
[0047] S2. Dynamic Equipment Status Assessment and Task Allocation:
[0048] Construct a device condition assessment model, and calculate the overall availability Sk of the device by comprehensively considering parameters such as the current load Lk, historical failure rate Fk, and energy efficiency Ek.
[0049] Sk=α·(1-Lk)+β·(1-Fk)+γ·Ek
[0050] Among them, α, β and γ are weighting coefficients and α+β+γ=1. They are dynamically adjusted according to the production process requirements, and tasks are given priority to be assigned to the equipment with the highest Sk value to ensure efficient use of resources.
[0051] S3, Multi-robot cooperative path planning and obstacle avoidance:
[0052] Based on the improved A* algorithm and combined with real-time obstacle detection data, the robot's motion trajectory is planned; the path cost function C is defined as:
[0053] C=λ1·Ddistance+λ2·Dobstacle+λ3·Twit
[0054] In the formula, Ddistance is the geometric distance of the path, Dobstacle is the obstacle influence factor (0 when there are no obstacles, and the value increases as the obstacle is closer), Twait is the waiting time at the workstation; λ1, λ2 and λ3 are cost weights, which are dynamically optimized through real-time production cycle time.
[0055] S4. Real-time production process monitoring and dynamic adjustment:
[0056] Establish a digital twin model of the production process to map the status of each workstation, material flow trajectory and equipment operating parameters in real time. Through preset production bottleneck identification rules (such as workstation load continuously exceeding 80% for 10 minutes), the scheduling strategy is automatically triggered to adjust, tasks are reassigned or equipment operating parameters are adjusted to achieve self-optimization of the production process.
[0057] Example 2
[0058] Unlike the above embodiments, this embodiment uses an automotive parts production line as an example to illustrate the specific implementation process of the present invention:
[0059] S1. Data Acquisition:
[0060] Temperature and current sensors are installed in each machining center, and RFID tags and visual recognition systems are deployed on the conveyor line to collect equipment operating parameters (such as spindle speed and feed rate), material position and order information in real time. The edge computing unit normalizes the sensor data, for example, converting the temperature data range [50℃, 100℃] into [0, 1].
[0061] S2. Equipment Status Assessment:
[0062] Given weighting coefficients α = 0.5 (load), β = 0.3 (failure rate), and γ = 0.2 (energy consumption), a certain processing center has a current load Lk = 0.6, a historical failure rate Fk = 0.1, and an energy efficiency Ek = 0.8. Its overall availability Sk = 0.5 × (1 - 0.6) + 0.3 × (1 - 0.1) + 0.2 × 0.8 = 0.63, which ranks high among similar equipment, and it is given priority in being assigned high-priority tasks.
[0063] S3, Path Planning and Obstacle Avoidance:
[0064] When the robot is transporting materials, the vision system detects a temporary stack of tooling fixtures (Dobstacle = 0.8) 3 meters ahead. The path planning algorithm automatically adjusts the trajectory, selecting a path with a 15% increase in distance but no obstacles, and updates the cost function C to ensure the safe completion of the transport task.
[0065] S4. Monitoring and Adjusting the Digital Twin Model:
[0066] If the load on the assembly station remains too high, the system will automatically trigger task redistribution, transferring some assembly tasks to idle standby stations, and restoring production line balance within 10 minutes.
[0067] In summary, this invention achieves real-time dynamic scheduling of industrial production lines through multi-source data fusion and intelligent algorithms, effectively improving production efficiency and flexibility.
[0068] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. An industrial production line real-time dynamic scheduling system, characterized in that: Comprising Data acquisition and preprocessing module: including Internet of Things sensor cluster, edge computing unit, realizing real-time acquisition, preprocessing of production data and real-time transmission to cloud server; Dynamic evaluation and decision module: integrating device state evaluation model, task allocation algorithm and path planning engine, calculating device availability based on real-time data, generating optimal task allocation scheme and robot motion trajectory; Cooperative control and execution module: including device controller and robot control cabinet, receiving and executing scheduling instructions, and feeding back device state to data acquisition and preprocessing module to form a closed-loop control; Digital twin monitoring module: building a three-dimensional virtual model of the production line, displaying the production state in real time, providing a visual monitoring interface, and supporting manual intervention and strategy adjustment.
2. A method for real-time dynamic scheduling of an industrial production line, implemented using the real-time dynamic scheduling system of claim 1, characterized in that, Specifically comprising the following steps: S1, real-time acquisition and preprocessing of multi-source data: Real-time acquisition of device state (speed, temperature, energy consumption), material position, workstation load, etc. by Internet of Things sensors (such as RFID, visual camera, pressure sensor), noise reduction and normalization processing by edge computing nodes to obtain standardized data set D = {d1, d2, …, dn}, where di represents the data of the ith sensor; The data normalization formula is: di' = max(di) - min(di) di - min(di) In the formula, di' is the normalized data, which solves the fusion problem of different dimensional data; S2, dynamic device state evaluation and task allocation: Construct a device state evaluation model to calculate the comprehensive availability Sk of the device based on the current load Lk, historical failure rate Fk, energy consumption efficiency Ek, etc. Sk = a · (1 - Lk) + β · (1 - Fk) + γ · Ek Where, α, β and γ are weight coefficients and α + β + γ = 1, dynamically adjusted according to production process requirements, preferentially assigning tasks to devices with the highest Sk value to ensure efficient use of resources; S3, multi-robot cooperative path planning and obstacle avoidance: Based on the improved A* algorithm, combined with real-time obstacle detection data, the robot motion trajectory is planned; Define the path cost function C as: C = λ1 · Ddistance + λ2 · Dobstacle + λ3 · Twait In the formula, Ddistance is the path geometric distance, Dobstacle is the obstacle influence factor (0 when there is no obstacle, the closer the obstacle, the larger the value), Twait is the workstation waiting time; λ1, λ2 and λ3 are cost weights, dynamically optimized by real-time production rhythm; S4, real-time production process monitoring and dynamic adjustment: Establish a digital twin model of the production process to real-time map the state of each workstation, material flow trajectory and device operating parameters, automatically trigger scheduling strategy adjustment to redistribute tasks or adjust device operating parameters based on pre-set production bottleneck identification rules (such as workstation load exceeding 80% for more than 10 minutes), and realize self-optimization of the production process.