Multi-axis plating line control system and method based on distributed group intelligence

Abstract

The invention discloses a multi-shaft plating line control system and method based on distributed group intelligence, and particularly relates to the technical field of plating line distributed control, the system comprises a physical execution layer, the physical execution layer comprises a pay-off shaft, a plurality of transmission guide shafts, a plurality of plating bath units and a take-up shaft which are arranged in sequence along a production line; the distributed intelligent control layer is composed of intelligent agent nodes which are in one-to-one correspondence with the execution units in the physical execution layer and have independent sensing, decision making and communication capabilities; according to the group intelligent communication network, all intelligent agent nodes are connected into a peer-to-peer network through a high-speed real-time industrial Ethernet, and low-delay and high-bandwidth information interaction between the nodes is achieved. According to the method, single-point faults can be eliminated, extremely high cooperative control precision is achieved, meanwhile, high flexibility and robustness are achieved, when a certain node in the system breaks down, the adjacent node can immediately sense and adjust self behaviors for compensation, and fault diffusion is avoided.

Description

Technical Field

[0001] This invention relates to the field of distributed control technology for electroplating lines, and specifically to a multi-axis electroplating line control system and method based on distributed swarm intelligence. Background Technology

[0002] In the photovoltaic industry chain, silicon wafer cutting is one of the key processes. Currently, the mainstream technology uses diamond wire for cutting, and copper-plated busbars are a core component of diamond wire. During its production, a layer of metallic copper or nickel needs to be uniformly plated onto the high-speed running busbar, and then diamond micropowder is bonded onto it through subsequent processes. The quality and efficiency of the electroplating process directly determine the performance, lifespan, and cost of the diamond wire.

[0003] Currently, existing multi-axis electroplating line control systems typically employ, for example... Figure 1 The centralized control architecture shown uses a central industrial computer as the sole brain, responsible for all calculations and decisions, while each axis controller is merely an instruction execution unit with no direct communication with each other. This centralized architecture has the following inherent drawbacks: The system is rigid and lacks flexibility: production formulas such as linear speed, tension, and current density are set uniformly by a central controller. Once a small fluctuation or malfunction occurs in the power supply of a certain piece of equipment, such as a certain anode tank, the entire system cannot adjust quickly and adaptively. It can only be intervened uniformly by the central controller, which often leads to a reduction in speed or shutdown of the entire line, seriously affecting production efficiency. High risk of single point of failure: The central control industrial computer or backbone network is the "brain" and "nerve center" of the entire system. Once a failure occurs, it will paralyze the entire production line, resulting in low reliability.

[0004] Insufficient precision in coordinated control: Under high-speed operation, the stability of bus tension and linear speed is crucial. Centralized systems have a delay in sensing the real-time status of each execution unit, making it difficult to achieve precise coordination at the millisecond level. For example, when the pay-off shaft experiences slight speed fluctuations due to mechanical resistance, the take-up shaft and intermediate drive shafts cannot instantly compensate, leading to unstable tension, which in turn affects the uniformity of the electroplating layer and may even cause wire breakage. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-axis electroplating line control system and method based on distributed swarm intelligence to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a multi-axis electroplating line control system based on distributed swarm intelligence, comprising: The physical execution layer includes a wire feeding shaft, multiple transmission guide shafts, multiple electroplating tank units, and a wire take-up shaft arranged sequentially along the production line. The distributed intelligent control layer consists of intelligent agent nodes that correspond one-to-one with each execution unit in the physical execution layer and have independent perception, decision-making, and communication capabilities. The swarm intelligence communication network uses high-speed real-time industrial Ethernet to connect all intelligent agent nodes into a peer-to-peer network, enabling low-latency, high-bandwidth information exchange between nodes. Each smart agent node includes: The environmental sensing module is used to collect key parameters of this execution unit in real time, such as linear velocity, tension, current, voltage, temperature, and liquid level. The local decision-making module has built-in speed consistency rules, tension balance rules, and energy optimization rules. It is used to calculate local control commands that only apply to this unit based on the global objective, the parameters of this unit, and the "state intention" of adjacent nodes. The communication interaction module is used to periodically send and receive "status intent" messages containing linear velocity, tension, health status, and timestamps to adjacent nodes.

[0007] Furthermore, the speed consistency rule ensures that each drive shaft, pay-off shaft, and take-up shaft agent node adjusts its own axis speed to a deviation of less than ±1% from the average linear speed of adjacent upstream and downstream agent nodes; The mathematical expression for the speed consistency rule is: V_i_desired(t) =α× V_global+(1-α)×V_{i-1}(t)+V_{i+1}(t)] / 2, Where V_i_desired(t) is the desired velocity of agent node i at time t, V_global is the global target velocity, V_{i-1}(t) and V_{i+1}(t) are the linear velocities of adjacent upstream and downstream agent nodes, respectively, and α is the global target weight coefficient.

[0008] Furthermore, the tension balance rule monitors the tension by converting motor torque. When the bus tension between any two adjacent agent nodes deviates from the target value by more than the ideal value, the relevant agent will compensate by fine-tuning the speed.

[0009] Furthermore, the energy optimization rule enables each electroplating tank agent node to dynamically reduce the current setting value based on the real-time copper ion concentration, temperature, and pH value of the tank, while ensuring that the coating quality is not lower than the set threshold, so as to minimize the energy consumption of a single tank. The energy optimization rule is locally optimized through the built-in benefit function Efficiency = f(coating quality, energy consumption). When the tank condition is good, the proxy node can appropriately reduce the current to the critical quality point to achieve energy saving.

[0010] Furthermore, the local decision-making module of the proxy node of the electroplating tank unit is also used to dynamically optimize its operating current based on the real-time operating parameters of the tank to achieve optimal local energy consumption.

[0011] A multi-axis electroplating line control method based on distributed swarm intelligence, applied to the above system, includes the following steps: Step 1, System Initialization: The upper-level management system issues global production targets to all intelligent agent nodes; Step 2, State Awareness and Broadcasting: Each intelligent agent node periodically collects key parameters of its unit and encapsulates them into "state intent" messages, which are then broadcast to neighboring nodes. Step 3, Local Decision-Making and Execution: Each node makes local decisions based on the received "state intent" messages from neighboring nodes, combined with the global goal and coordination rules, and drives the executor to take action; Step 4, Continuous Coordination and Adaptation: Repeat steps 2 and 3, so that steps 2 and 3 are executed cyclically in each control cycle, enabling the system to continuously respond to internal and external disturbances and achieve dynamic equilibrium.

[0012] Furthermore, in step three, the coordination rules include speed consistency rules, tension balance rules, and energy optimization rules.

[0013] Furthermore, in step four, the duration of the control cycle is 1 to 10 milliseconds.

[0014] Furthermore, in step one, the global production target includes the target linear velocity and total output.

[0015] Furthermore, in step two, the message includes the local axis ID, current speed, tension, and status flag.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. High flexibility and robustness: If a node in the system fails, such as a power fluctuation, its neighboring nodes can immediately sense it and adjust their behavior to compensate, preventing the fault from spreading. The system can operate in a degraded manner with the fault instead of shutting down completely, which greatly improves availability. 2. Eliminates single points of failure: The distributed architecture has no central bottleneck, and the failure of any node does not affect the normal operation and local collaboration of other nodes, thus significantly improving system reliability; 3. Extremely high collaborative control precision: Based on millisecond-level information interaction of peer-to-peer network, local fine-tuning of speed and tension can be completed almost in real time, ensuring the ultra-stable state of the busbar under high-speed operation, and greatly improving electroplating uniformity and product quality. 4. Easy to expand and maintain: Adding a new electroplating unit simply requires connecting its intelligent agent node to the network; it will automatically integrate into the group and participate in collaboration according to established rules. Highly modular and easy to debug; 5. Energy saving and consumption reduction: Each electroplating tank agent can independently optimize energy consumption, realize refined energy management at the system level, and reduce production costs. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a schematic diagram of a centralized control architecture in existing technology; Figure 2 This is a diagram illustrating the architecture of a distributed swarm intelligence collaborative control system according to an embodiment of the present invention. Figure 3 This is a functional block diagram of an intelligent agent node according to the present invention; Figure 4 This is a flowchart illustrating the collaborative process of speed consistency rules in an embodiment of the present invention. Figure 5 This is a flowchart illustrating the coordination process of the tension balance rules in an embodiment of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solution of the present invention, the specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0020] I. Hardware Implementation refer to Figure 2 An electroplating line for producing copper-plated busbars consists of, from left to right, a feed shaft, drive shaft 1, anode roller shaft corresponding to electroplating tank 1, drive shaft 2, anode roller shaft corresponding to electroplating tank 2, ..., drive shaft N, and take-up shaft.

[0021] Each axis is driven by a servo motor or a variable frequency motor and is equipped with a high-precision encoder and torque sensor. Each electroplating tank is equipped with an independent programmable DC power supply, temperature sensor, liquid level sensor, etc.

[0022] The hardware carrier of each intelligent agent node can be an embedded industrial controller (such as an ARM or x86-based embedded PC) that integrates a CPU, I / O interface and network interface, a high-performance PLC (such as a soft PLC that supports CODESYS) or even an advanced servo drive (which itself has powerful computing and communication capabilities).

[0023] These controllers are connected in a ring or linear network via an EtherCAT bus. EtherCAT is well-suited for this application due to its high real-time performance (cycle times up to 100μs) and flexible topology.

[0024] II. Software Implementation Each agent node's software runs a lightweight real-time operating system (such as VxWorks, FreeRTOS, or Linux with RT-Preempt Patch). The software logic mainly includes: 1. Communication Protocol: Define a concise "status intent" message format, for example: struct AgentMessage { uint16_t AgentID; / / Node ID float CurrentSpeed; / / Current linear speed (m / min) float CurrentTorque; / / Current motor torque (Nm) float SetPoint; / / Current setpoint value of this node uint8_t HealthStatus; / / Health status code uint32_t Timestamp; / / Timestamp }; Each control cycle (e.g., set to 5ms), the node broadcasts this message to the entire network or a specific group of neighboring nodes.

[0025] 2. Local decision-making algorithm (taking drive shaft proxy as an example, see reference) Figure 4 ): Attached image description: Figure 4 It describes how proxy nodes make decisions based on speed consistency rules to ensure that local speeds are synchronized with upstream and downstream nodes.

[0026] 3. Implementation of the tension balance algorithm (refer to) Figure 5 ): Attached image description: Figure 5 It demonstrates how relevant agent nodes coordinate to restore tension balance when an anomaly in tension is detected.

[0027] III. Workflow Example Assume the production line is running stably at a speed of 600 meters per minute. Suddenly, the drive shaft in area 3 of the electroplating tank experiences a slight increase in resistance torque due to foreign matter becoming entangled.

[0028] In a traditional system: A slight decrease in the speed of the drive shaft leads to an increase in tension between it and the upstream shaft. The tension sensor signal is transmitted to the central PLC. After processing, the central PLC commands all shafts in the line to fine-tune their speed. The response delay may cause the tension to exceed the limit, triggering an alarm and shutdown.

[0029] In the system of this invention: 1. Agent 3 of the electroplating tank detected an abnormal increase in tension on its inlet side.

[0030] 2. It immediately calculates a tiny acceleration command for itself, while simultaneously sending a message containing the tension deviation to its direct upstream agent (driveshaft 2 agent).

[0031] 3. After receiving the message, the agent of drive shaft 2 calculates a tiny acceleration command based on its own status.

[0032] 4. Almost simultaneously, the downstream agent of electroplating tank 3 (drive shaft 4 agent) senses the decrease in tension on the outlet side and will slow down slightly.

[0033] 5. These adjustments are completed within 5ms, and tension fluctuations are confined to a localized area and quickly subside. The take-up and pay-off agents barely notice the disturbance, and the production line maintains high-speed, stable operation.

[0034] IV. Extended Implementation To add an electroplating tank unit, simply: 1. Install the new tank, anode roller shaft and its drive motor.

[0035] 2. Connect the new intelligent agent controller to the existing EtherCAT network.

[0036] 3. Configure basic parameters for the agent (such as ID, logical position in the production line, etc.).

[0037] 4. After the system powers on, the new agent will automatically discover its upstream and downstream neighbors and begin to participate in collaboration in accordance with the group rules. No modification to the central program is required.

[0038] In summary, this invention, through its distributed swarm intelligence architecture, effectively solves the core pain points of existing electroplating line control systems, providing strong technical support for the high-quality, efficient, and low-cost production of high-end copper-plated busbars in the photovoltaic industry.

[0039] V. Experimental Control By comparing and testing with traditional central PLC control systems, the significant advantages of this invention in three key performance indicators—tension control accuracy, system energy consumption level, and fault tolerance and self-healing capability—are quantitatively verified, thus demonstrating the inventiveness, novelty, and practicality of this invention.

[0040] Experimental setup ① Comparison System: Control group (traditional system): A centralized control architecture based on a single industrial PLC. All sensor data is uploaded to the PLC, and all control commands are calculated by the PLC and then sent to each actuator.

[0041] Experimental Group (This Invention): A control architecture based on distributed swarm intelligence. Each motion axis and electroplating tank acts as an intelligent agent node, communicating peer-to-peer via an EtherCAT network, and making local decisions according to speed consistency rules and tension balance rules.

[0042] ②Testing platform: An experimental platform for a 16-axis electroplating line simulating the production of photovoltaic copper-plated busbars includes a feed shaft, a take-up shaft, 6 drive / anode roller shafts, and 2 independently controlled electroplating tank units.

[0043] Key instruments: high-precision torque sensor (for tension measurement), servo driver and motor, high-precision energy meter, data acquisition system.

[0044] ③ Test scenario: Scenario 1: Tension stability test under dynamic disturbance; Scenario 2: Comprehensive Energy Consumption Comparison Test; Scenario 3: Fault Injection and System Fault Tolerance Testing; Experimental Results and Data Analysis: 1. Comparative Experiment on Tension Fluctuation Control Objective: To verify that, under simulated typical disturbances, the system of this invention, based on local cooperative rules, can achieve faster and smoother tension recovery, highlighting its extremely high cooperative control accuracy.

[0045] Method: When the system is running stably at 40m / min (typical speed for photovoltaic diamond wire production), a momentary mechanical blockage (lasting 0.5 seconds) is simulated between drive shafts #5 and #6, and the tension change curves at key measurement points are recorded.

[0046] Results: Data Analysis Table:

[0047] Conclusion: Experimental data fully demonstrate the effectiveness of the velocity consistency rule and the tension balance rule. The distributed architecture of this invention enables disturbances to be rapidly absorbed and compensated locally, preventing them from propagating throughout the entire line. This significantly reduces the amplitude and duration of tension fluctuations, laying a solid foundation for obtaining a uniform electroplating layer.

[0048] 2. System Energy Consumption Comparison Experiment Objective: To verify the comprehensive energy-saving effect brought about by the electroplating tank agent optimizing local energy consumption based on real-time operating conditions (energy optimization rule) and the transmission system reducing dynamic adjustment energy consumption through intelligent collaboration in the system of the present invention.

[0049] Method: Two control systems each performed the same 8-hour production task, including startup, steady-state operation at three different speeds, periodic acceleration and deceleration, and normal shutdown. The total power consumption of the production line was recorded.

[0050] Results: Comprehensive Energy Consumption Comparison Table:

[0051] Conclusion: The system of this invention achieves an average energy saving of 8.5%, with particularly significant energy savings (up to 13.0%) under dynamic operating conditions. This is due to: ① Local optimization of electroplating tank: The current is dynamically fine-tuned to the optimal value based on real-time tank conditions (such as temperature and ion concentration), thus avoiding energy waste.

[0052] ② Coordination efficiency of the transmission system: Coordination among intelligent agents reduces energy impact and ineffective work during dynamic processes such as acceleration and deceleration.

[0053] 3. Fault Injection and Fault Tolerance Testing Objective: To verify the high flexibility and robustness of the distributed architecture of this invention when system components fail, as well as its ability to eliminate single points of failure.

[0054] Method: During the period of stable system operation, the following faults were artificially injected, and the system behavior was observed and recorded.

[0055] Results: Fault Test Records and Comparison Table:

[0056] Overall Conclusion This experiment, through three sets of quantifiable comparative tests, strongly demonstrates that the present invention, "A Multi-Axis Electroplating Line Control System and Method Based on Distributed Swarm Intelligence," represents a significant advancement over existing technologies. ① In terms of control performance, the tension control accuracy has been improved by more than 60%, and the response speed has been improved by 75%, providing a key guarantee for achieving high-speed and high-quality electroplating processes; ② In terms of energy efficiency, it has achieved an average energy saving of 8.5%, and the energy saving effect is even better in complex dynamic production processes, which meets the requirements of green manufacturing; ③ In terms of system reliability, it perfectly solves the problems of "single point of failure" and "system rigidity" in traditional centralized architecture, demonstrating excellent fault tolerance and self-healing ability, and can significantly improve the overall equipment efficiency (OEE). These experimental data, together with the technical solutions, beneficial effects, and claims described in this invention specification, form a complete chain of evidence, fully demonstrating the technical superiority and industrial application value of this invention.

[0057] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A multi-axis electroplating line control system based on distributed swarm intelligence, characterized in that, Comprise: A physical execution layer comprising a pay-off spool, a plurality of transmission guide spools, a plurality of electroplating tank units and a take-up spool arranged in sequence along a production line; A distributed intelligent control layer composed of intelligent agent nodes corresponding to each execution unit in the physical execution layer and having independent perception, decision-making and communication capabilities; A swarm intelligence communication network that connects all intelligent agent nodes into a peer-to-peer network using high-speed real-time industrial Ethernet to achieve low-latency and high-bandwidth information exchange between nodes; Each intelligent agent node comprises: An environment perception module for real-time acquisition of key parameters such as line speed, tension, current, voltage, temperature and liquid level of the execution unit; A local decision-making module with built-in speed consistency rules, tension balancing rules and energy optimization rules for calculating local control instructions that only act on the unit based on global targets, unit parameters and adjacent node "state intention" messages; A communication interaction module for periodically sending and receiving "state intention" messages containing line speed, tension, health status and time stamp to adjacent nodes.

2. The multi-axis electroplating line control system based on distributed swarm intelligence according to claim 1, wherein: The speed consistency rule adjusts the line speed of each transmission spool, pay-off spool and take-up spool agent node to a deviation of less than ±1% from the average line speed of adjacent upstream and downstream agent nodes. The mathematical expression of the speed consistency rule is: V_i_desired(t) =α× V_global+(1-α)×[V_{i-1}(t)+V_{i+1}(t)] / 2, Where V_i_desired(t) is the desired speed of agent node i at time t, V_global is the global target speed, V_{i-1}(t) and V_{i+1}(t) are the line speeds of adjacent upstream and downstream agent nodes, and α is the global target weight coefficient.

3. The multi-axis electroplating line control system based on distributed swarm intelligence according to claim 1, wherein: The tension balancing rule monitors tension by motor torque conversion. When the busbar tension between any two adjacent agent nodes deviates from the target value by more than the ideal value, the relevant agent will compensate by fine-tuning the speed.

4. The multi-spindle electroplating line control system based on distributed swarm intelligence according to claim 1, wherein: The energy optimization rule enables each electroplating tank agent node to dynamically reduce the current set value based on real-time copper ion concentration, temperature and pH value in the tank to minimize single-tank energy consumption while ensuring that the plating layer quality does not fall below a set threshold. The energy optimization rule performs local optimization through the built-in efficiency function Efficiency = f. When the tank condition is good, the agent node can appropriately reduce the current to the critical quality point to achieve energy saving.

5. The multi-axis electroplating line control system based on distributed swarm intelligence according to claim 1, wherein: The local decision-making module of the agent node of the electroplating tank unit is also used to dynamically optimize its working current based on real-time operating parameters of the tank to achieve local energy optimization.

6. A multi-axis electroplating line control method based on distributed swarm intelligence, applied to the system of claim 1, characterized in that, The steps are as follows: Step one, system initialization: the upper management system issues global production targets to all intelligent agent nodes; Step two, state perception and broadcast: each intelligent agent node periodically acquires key parameters of the unit and encapsulates them into "state intention" messages to broadcast to adjacent nodes; Step three, local decision-making and execution: each node makes local decisions based on received "state intention" messages from adjacent nodes, combined with global targets and coordination rules, and drives the actuator to act. Step four, continuous coordination and self-adaptation: repeating step two and step three, so that step two and step three are executed in each control cycle, so that the system can continuously cope with internal and external disturbances and achieve dynamic balance.

7. The multi-spindle electroplating line control method based on distributed swarm intelligence according to claim 6, characterized in that: In the step three, the coordination rules include speed consistency rules, tension balance rules and energy optimization rules.

8. The multi-spindle electroplating line control method based on distributed swarm intelligence according to claim 6, characterized in that: In the step four, the length of the control cycle is 1-10 milliseconds.

9. The multi-spindle electroplating line control method based on distributed swarm intelligence according to claim 6, characterized in that: In the step one, the global production target includes target linear speed and total output.

10. The multi-spindle electroplating line control method based on distributed swarm intelligence according to claim 6, characterized in that: In the step two, the message contains the axis ID, current speed, tension, and status flag.

Images