Automobile robot resistance spot welding cycle time optimization method

Abstract

The invention discloses a robot period time optimization method for automobile robot resistance spot welding. The method comprises the following steps that S1, a 3D virtual simulation model is constructed; s2, collecting production data; s3, constructing a task sequence optimization graph; s4, robot motion planning is optimized; s5, simulating the motion trail of each robot, verifying the non-interference of the optimized task sequence, and if collision exists, adjusting the motion time sequence of the robot; s6, operating the 3D virtual simulation model; and S7, applying the optimization scheme to an actual production line, and carrying out physical verification. According to the method, simulation software 3D process simulation software is used as a core tool, and the method comprises the steps of establishing a 3D virtual simulation model matched with an actual production line, accurately extracting production data, optimizing a task sequence based on a plan review technology, planning robot motion in a classified manner, performing collision detection verification, performing multi-round simulation iteration, performing physical verification landing and the like. And efficient optimization of the resistance spot welding period time of the automobile robot is achieved.

Description

Technical Field

[0001] This invention relates to the field of automotive body manufacturing technology, specifically to a method for optimizing the cycle time of resistance spot welding for automotive robots. Background Technology

[0002] Resistance spot welding is a core joining process in the manufacturing of automotive body-in-white, and the stability of its welding quality has a decisive impact on the structural strength and safety performance of the entire vehicle. With the rapid development of the automotive manufacturing industry towards intelligence and automation, welding robots have become an indispensable key piece of equipment in modern automotive production lines. However, in the practical engineering applications of multi-robot welding systems, a series of technical bottlenecks still urgently need to be addressed.

[0003] In automotive welding production lines with multiple workstations and multiple robots working collaboratively, the rationality of task allocation and path planning directly affects overall production efficiency and equipment utilization. To address this issue, Wang Ye et al. established a mathematical model for multi-workstation, multi-robot welding task allocation and path planning, aiming to simultaneously minimize processing time and robot motion path length, thereby improving production efficiency (Reference 1). Chen Zhilan et al., by establishing a 3D model of the welding workstation and programming the controller to conduct process simulation and virtual debugging, identified and corrected trajectory interference problems in advance, optimizing the workflow sequence (Reference 2).

[0004] At the system design level of welding robot workstations, multiple factors need to be comprehensively considered, including robot motion trajectory, tooling fixture layout, and safety protection. Especially in the limited space where multiple robots are working collaboratively, the intersection and overlap of robot motion trajectories can easily lead to collision accidents. Traditional manual adjustment methods are difficult to guarantee accuracy and pose safety hazards. Zhang Yinghua proposed using heuristic algorithms to optimize motion trajectories, while integrating hardware protection such as arc light protection barriers and safety light curtains, as well as dynamic clamping compensation functions for fixtures, to achieve coordinated protection of trajectory efficiency, clamping stability, and operational safety (Reference 3).

[0005] Lin Juguang et al. studied the application of 3D simulation software in the planning and design of spot welding workstations for robot body-in-white, confirming that 3D virtual simulation can effectively identify potential collision problems in welding work units and optimize path planning before implementation (Reference 4). Dávila-Ros et al.'s research on robot simulation environments showed that 3D virtual simulation can effectively identify potential collision problems in welding work units and optimize path planning before implementing real work units (Reference 5).

[0006] Despite the progress made in existing research, three core bottlenecks still exist in practical production applications: First, path planning and welding point allocation rely on manual experience, which is time-consuming, labor-intensive, and prone to deviations due to subjective judgment, resulting in low efficiency in path determination. Patent CN202410932007 explicitly points out this problem. Second, multi-robot collaboration lacks a scientific scheduling mechanism, which easily leads to load imbalance and trajectory interference. Third, existing solutions mostly focus on single-dimensional optimization and lack a systematic analysis of efficiency and trajectory rationality. CN202211736821 points out that traditional monitoring only focuses on load protection and lacks reliable data to support trajectory and efficiency evaluation, making it difficult to improve production efficiency.

[0007] Existing technologies mostly focus on single-dimensional optimization, such as adjusting the task sequence or optimizing the motion of a single robot, lacking a comprehensive system-level solution. Therefore, there is an urgent need in this field for a systematic approach that integrates virtual simulation, task sequence optimization, motion planning, and physical verification to achieve comprehensive optimization of the cycle time of multi-robot resistance spot welding. Summary of the Invention

[0008] This invention aims to provide a method for optimizing the cycle time of resistance spot welding for automotive robots. The purpose is to overcome the problems of load imbalance among multiple robots, excessive idle time of some robots, and easy trajectory interference in the existing automotive robot resistance spot welding production line. By optimizing the task sequence of the resistance spot welding cycle of automotive robots through this method, the redundant idle time of automotive robots can be significantly reduced, and the risk of trajectory interference of automotive robots at critical workstations can be reduced.

[0009] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for optimizing the cycle time of an automotive robot resistance spot welding robot includes the following steps: Step S1: Based on 3D process simulation software, construct a 3D virtual simulation model of the automotive robot resistance spot welding production line. The 3D virtual simulation model includes a robot, a C-type welding torch, an X-type welding torch, a truck cab, and tooling fixtures. The geometric parameters of the 3D virtual simulation model do not exceed 0.5% of the actual production line. Step S2: Control the system and robot controller of the actual production line through the programmable logic controller, collect production data continuously for 24 hours at a sampling frequency of 1Hz, and extract the current task sequence and time parameters of each robot; Step S3: Construct a task sequence optimization diagram using the project review and approval technical diagram, and adjust the task sequence according to the process logic; Step S4: Optimize robot motion planning. Non-spot welding activities use joint motion, spot welding activities use linear motion, and limit the rotation angle of robot joints. Step S5: Using the collision detection function of 3D process simulation software, simulate the movement trajectory of each robot in the two key workstations of the roof docking area and the side panel welding area to verify the non-interference of the optimized task sequence. If a collision occurs, adjust the robot movement sequence. Step S6: Run the 3D virtual simulation model to obtain the optimized total cycle time and time parameters of each robot. If the total cycle time is greater than the expected time, return to steps S3 and S4 to adjust the task sequence or motion parameters until the requirements are met. Step S7: Apply the optimized solution to the actual production line and conduct physical verification for 100 production cycles to ensure that the error between the actual total cycle time and the simulation data is ≤1%, and that the solder joint temperature is maintained at 1800-2200℃ and the welding gun pressure fluctuation is ≤±5%.

[0010] According to another aspect of the present invention, the tooling fixture in step S1 includes a floor positioning fixture, a roof fixing device and a side panel clamping mechanism, wherein the positioning accuracy of the floor positioning fixture is ±0.1mm and the clamping force fluctuation of the side panel clamping mechanism is ≤±3%.

[0011] According to another aspect of the present invention, when collecting production data in step S2, the robot joint current, welding torch electrode pressure and welding time of the weld point are recorded simultaneously, and the average value of valid data within 24 hours and data during equipment downtime periods is taken as the current time parameter.

[0012] According to another aspect of the present invention, the task node logical relationship of the technical review diagram in step S3 satisfies the sequence of "floor positioning fixture clamping", "floor welding", "side panel clamping", "side panel welding", and "roof welding".

[0013] According to another aspect of the present invention, the collision detection coverage of the collision detection function in step S5 includes the robot body, welding torch, tooling fixture and truck cab of the car body to be welded, and the detection scenario covers the robot's starting position, intermediate path nodes and ending position.

[0014] According to another aspect of the present invention, when performing physical verification in step S6, an infrared thermometer is used to monitor the temperature of the weld joint in real time, and pressure data is recorded by a welding torch pressure sensor. The welding quality must meet the requirements for the strength of resistance spot welded joints in the ISO14329 standard.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention uses 3D process simulation software as the core tool. Through steps such as establishing a 3D virtual simulation model that matches the actual production line, accurately extracting production data, optimizing task sequences based on plan review technology, classifying and planning robot motion, collision detection verification, multiple rounds of simulation iteration, and physical verification implementation, it achieves efficient optimization of the resistance spot welding cycle time of automotive robots. Attached Figure Description

[0016] Appendix Figure 1 This is a sequence diagram of the robot tasks before optimization in an embodiment of the present invention; Appendix Figure 2 This is a robot task sequence diagram optimized based on plan review technique in an embodiment of the present invention; Appendix Figure 3 This is a flowchart of the multi-robot motion planning algorithm in an embodiment of the present invention; Appendix Figure 4 This is a comparison chart of the processing time and idle time of each robot before and after optimization in the embodiments of the present invention.

[0017] For clarity and convenience, the references are listed here together. References are cited in this application by their respective reference numbers.

[0018] 1. Wang Ye, Wang Xuewu, Gu Xingsheng. Task allocation and path planning for multi-station, multi-robot welding [J]. Journal of East China University of Science and Technology (Natural Science Edition), 2025, 51(5): 633-644.

[0019] 2. Chen Zhilan, Sun Xin. Research on digital process simulation of welding path of body-in-white [J]. Mechanical Design and Research, 2024, 40(6): 188-193.

[0020] 3. Zhang Yinghua. Design of a welding robot workstation [J]. Electric Welding Machine, 2014, 44(8): 104-106.

[0021] 4. Lin Juguang, Tang Donghua. Application of DELMIA in the planning and design of spot welding workstation for robot body-in-white [J]. Mechanical Design and Manufacturing, 2010(12): 90-92.

[0022] 5.Dávila-Ros I, Torres-Treviño LM, López-Juárez I. On the implementation of a robotic welding process using 3D simulation environment[C] / / 2008 Electronics, Robotics and Automotive Mechanics Conference. 2008:283-287. Detailed Implementation

[0023] To make the objectives and technical solutions of the embodiments of the present invention clearer, 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, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0024] The following description, in conjunction with the accompanying drawings and an example of a factory's four-robot resistance spot welding production line, details the implementation process of this invention.

[0025] Step S1: Construction of 3D Virtual Simulation Model (1) Model import and configuration: The robot model was retrieved from the 3D process simulation software model library. The end effector configuration was modified according to the actual welding torch interface. 3D models of C-type and X-type welding torches were imported, and the electrode pressure adjustment range was set to 3-5 kN. 3D models of the floor positioning fixture (positioning accuracy ±0.1 mm), roof fixing device, side panel clamping mechanism (clamping force fluctuation ≤ ±3%), and truck cab (material Q235 steel, thickness 1.5-2.0 mm) were also imported. The models were placed according to the actual production line layout, ensuring the spacing error did not exceed 0.5%. Specific model parameters are shown in Table 1 below. Table 1 Detailed list of parameters for each piece of equipment Equipment Category quantity Key parameters robot 4 units Load capacity: 280kg; Number of joints: 6 C-type welding torch 2 Electrode diameter: 6mm X-type welding torch 2 Electrode travel: 50mm (2) Solder joint planning and parameter setting: 209 weld points were precisely marked on the 3D model of the truck cab. Coordinate and process parameters were input for each weld point. Taking the floor weld point P1 as an example: coordinates (X=1200mm, Y=800mm, Z=300mm), welding current 10kA, welding time 1.0 second. The motion parameters of each joint of the robot were set in the 3D process simulation software to ensure that the simulated motion matches the actual robot performance. A detailed breakdown of the weld point distribution is shown in Table 2 below. Table 2 Detailed List of Welding Points Welding area Number of solder joints (pieces) percentage Typical welding parameters Welding requirements Floor area 60 28.7% Current: 9-11kA Duration: 1.0-1.2s High positioning accuracy is required, necessitating the use of floor clamps. Side panel area 80 38.3% Current: 10-12kA Duration: 0.9-1.1s Welding reliability needs to be verified. Roof area 69 33.0% Current: 8-10kA Duration: 1.0-1.2s In areas where multiple robots work together, interference is likely to occur. total 209 100% - Covering the main structural parts of the truck cab Step S2: Production Data Collection and Processing Through the data communication link established with the robot controller, continuous 24-hour production data collection is carried out, with the sampling frequency set to 1Hz. The collected parameters comprehensively cover key indicators such as process execution time, idle time, joint current, welding torch electrode pressure, and welding point temperature of each industrial robot.

[0026] During the data processing phase, invalid data from periods such as 2 hours of equipment maintenance and 1 hour of electrode replacement were first removed. Then, statistical analysis was performed on the remaining 21 hours of valid data to obtain the robot's generated data, as shown in Table 3. The analysis results show that: the first robot's processing time was 204 seconds, with an idle time of 328 seconds; the second robot's processing time was 400 seconds, with an idle time of 132 seconds; the third robot's processing time was 304 seconds, with an idle time of 228 seconds; and the fourth robot's processing time was 367 seconds, with an idle time of 165 seconds. The total production line cycle time was 532 seconds, corresponding to a capacity of 25 units / hour. These data provide an accurate benchmark for subsequent optimization work.

[0027] Table 3. Production data collection table for each robot before optimization name Processing time s Idle time s First Robot 204 328 Second robot 400 132 Third Robot 304 228 The Fourth Robot 367 165 Step S3: Plan Review, Technical Chart, Task Sequence Optimization, and Implementation Based on the collected production data, a complete task sequence optimization model was constructed using the Plan Evaluation and Review Technique (PEP). First, the PEP diagram was drawn: using "Welding torch pickup (A)," "Floor welding (B)," "Side panel welding (C)," "Roof welding (D)," "Welding torch placement (E)," and "Jig operation (F)" as nodes, and arrows to indicate logical relationships (F→B→F→C→F→D→E). Based on this, strict node dependencies were established according to the process logic of "Floor welding point → Side panel welding point → Roof welding point," resulting in the robot motion sequence diagram before optimization, as shown below. Figure 1 .

[0028] By deeply analyzing the task logic in the project review technical diagram, two key redundant links were identified: After completing the "roof fixing" operation, the second robot needs to wait for the third robot to complete the "side panel positioning" operation before it can continue working, with a single waiting time of 16 seconds; the fourth robot has an 11-second idle time after the "front panel welding" process. To address these issues, targeted optimization measures were implemented: the start time of the third robot's "side panel welding" task was adjusted from "10 seconds after floor welding" to "immediately after floor welding," effectively eliminating the 16-second waiting time; simultaneously, the "side panel repair welding" process was adjusted from "after roof welding" to "after side panel welding," making full use of the 11-second idle time. The optimized robot motion sequence diagram was obtained through analysis of the project review technical diagram, see [link to diagram]. Figure 2 .

[0029] Regarding task redistribution, the workload of the robots was systematically balanced. The original task of "spot welding of the roof and rear panel" (25 seconds) of the second robot was split into two parts: the first robot performed "test welding of the rear panel" (15 seconds), and the third robot performed "repair welding of the rear panel" (18 seconds). At the same time, the task of "repair welding of the roof" (28 seconds) of the fourth robot was redistributed to the first robot (optimized to 22 seconds). After this adjustment, the workload of the first robot increased from 10 processes to 12 processes, while the workload of the second robot decreased from 13 processes to 10 processes, effectively achieving load balancing among the four robots.

[0030] Step S4: Robotic Arm Motion Planning Based on the different characteristics of the welding tasks, the robot's motion mode was finely set: for non-spot welding activities such as gripping, load handling, and idle movement, joint motion mode was adopted to shorten unnecessary movement time and improve motion efficiency; for spot welding activities such as floor welding and side panel welding, linear motion mode was adopted to ensure that the welding gun positioning accuracy reaches the process requirement of ±0.05mm.

[0031] See the flowchart of the robot motion planning algorithm. Figure 3 .according to Figure 3 First, motion planning is performed on the robotic arm of a single robot. In the 3D process simulation software, reasonable joint rotation limits are set based on the robot's mechanical performance. The robot's joints play a key role in adjusting the welding torch's posture, and their rotation upper limit is set to 85% (actually controlled at 80%). The rotation angles of the other joints are limited to within 88% (actually controlled at 85%). These limiting parameters ensure both motion flexibility and effectively prevent the risk of joint overload.

[0032] Meanwhile, the robot's motion path was smoothed and optimized. By reducing rapid acceleration and deceleration, the motion efficiency was improved and the mechanical stress of the equipment was reduced. In particular, in areas where multiple robots work together, such as the roof docking area and the side panel welding area, the robot's motion trajectory was optimized to ensure the coordination of motion between the robots.

[0033] Step S5: Collision Detection Comprehensive collision detection verification was conducted using 3D process simulation software, with a detection accuracy set to 0.1mm. The collision detection range covered all key components, including the robot body, welding torch, tooling fixtures, and the car body to be welded, holding the cab. The focus was on simulating the coordinated motion trajectory of each robot in two key workstations: the roof docking area and the side panel welding area.

[0034] During the verification process, a 0.3mm motion interference risk was discovered in the roof docking area when the second robot was performing the "roof fixing" task and the third robot was performing the "side panel welding" task. Through in-depth analysis of the cause of the interference, it was decided to advance the start time of the third robot's relevant tasks by 5 seconds. This adjustment successfully eliminated the potential interference. To ensure reliability, ten repeated collision verification tests were conducted, and robot motion under different production cycle times, including ±5% periodic fluctuations, was simulated, confirming that no interference occurred under any operating conditions.

[0035] Step S6: Simulation Iteration After initial optimization, the first round of simulation results showed a total cycle time of 482 seconds, which did not yet meet the expected target. Further analysis of the simulation data revealed that the third robot still had 150 seconds of idle time. A second round of simulation optimization was performed by reassigning the "front panel repair welding" task (20 seconds) originally handled by the fourth robot to the third robot.

[0036] See the robot runtime and idle time allocation before and after optimization. Figure 4 .Depend on Figure 4 As can be seen, after optimization, the total production cycle time was reduced to 464 seconds, and the time parameters of each robot reached a balanced state: the first robot's processing time was 241 seconds and idle time was 223 seconds; the second robot's processing time was 341 seconds and idle time was 123 seconds; the third robot's processing time was 338 seconds and idle time was 126 seconds; and the fourth robot's processing time was 327 seconds and idle time was 137 seconds. This optimization result represents a reduction of 68 seconds compared to the original cycle time of 532 seconds, an optimization margin of 12.8%.

[0037] During the production line verification phase, the optimized task sequence and motion parameters (including joint rotation limits, motion type settings, etc.) were first imported into the controllers of the four robots, and the control logic of the controllers was updated simultaneously to ensure that the simulation parameters were completely consistent with the actual control parameters.

[0038] Step S7: Physical Verification Subsequently, the production line was started for on-site verification testing of 100 consecutive production cycles. Monitoring results showed that the actual total cycle time averaged 466 seconds, with an error of only 0.4% compared to the simulated value of 464 seconds, demonstrating the high reliability of the simulation model. Process parameter monitoring data showed that the weld point temperature was stably maintained within the ideal range of 1850-2150℃, and the welding torch pressure fluctuation was controlled between 3.2-4.8kN (fluctuation range ≤±5%), fully meeting the process specification requirements.

[0039] The quality department randomly sampled 50 weld joints for strength testing. The tensile strength test results were within the range of 350-380 MPa, meeting the stringent requirements of ISO 14329 standard for the strength of resistance spot welded joints. Throughout the verification process, no robot collisions occurred, and operators did not need to enter hazardous areas for manual intervention, effectively improving production safety. Specific performance indicators of the production line before and after optimization are shown in Table 4 below: Table 4 Comparison of Production Line Indicators Before and After Optimization Performance indicators Production line data before optimization Simulation optimization data Optimized production line data range of change Total cycle time (seconds) 532 464 466 ↓12.4% Production line capacity (units / hour) 25 28.4 28 ↑12.0% Load balance (standard deviation) 83.5 seconds 47.8 seconds 48.5 seconds ↑42.0% Total idle time (seconds) 853 609 623 ↓27.0% Solder joint temperature (°C) 1800-2200 - 1850-2150 Meets standards Welding torch pressure fluctuation ≤±5% - 3.2-4.8kN Meets standards After the optimization plan was fully implemented, the production line capacity significantly increased from 25 units / hour to 28 units / hour. Following a month of continuous production operation monitoring, the production line remained stable and reliable, with no equipment failures or quality issues. This achievement fully demonstrates the effectiveness and engineering practical value of this optimization method in solving the cycle time optimization problem of multi-robot resistance spot welding production lines.

[0040] This invention uses 3D process simulation software as its core tool. Through steps such as establishing a 3D virtual simulation model that matches the actual production line, accurately extracting production data, optimizing task sequences based on plan review technology, classifying and planning robot motion, collision detection verification, multiple rounds of simulation iteration, and physical verification implementation, it achieves efficient optimization of cycle time. It is especially suitable for resistance spot welding production lines for components such as car cabs and body frames that operate in a multi-robot collaborative manner.

Claims

1. A method for optimizing the cycle time of an automotive robot resistance spot welding robot, characterized in that: Includes the following steps: Step S1: Based on 3D process simulation software, construct a 3D virtual simulation model of the automotive robot resistance spot welding production line. The 3D virtual simulation model includes a robot, a C-type welding torch, an X-type welding torch, a truck cab, and tooling fixtures. The geometric parameters of the 3D virtual simulation model do not exceed 0.5% of the actual production line. Step S2: Control the system and robot controller of the actual production line through the programmable logic controller, collect production data continuously for 24 hours at a sampling frequency of 1Hz, and extract the current task sequence and time parameters of each robot; Step S3: Construct a task sequence optimization diagram using the project review and approval technical diagram, and adjust the task sequence according to the process logic; Step S4: Optimize robot motion planning. Non-spot welding activities use joint motion, spot welding activities use linear motion, and limit the rotation angle of robot joints. Step S5: Using the collision detection function of 3D process simulation software, simulate the movement trajectory of each robot in the two key workstations of the roof docking area and the side panel welding area to verify the non-interference of the optimized task sequence. If a collision occurs, adjust the robot movement sequence. Step S6: Run the 3D virtual simulation model to obtain the optimized total cycle time and time parameters of each robot. If the total cycle time is greater than the expected time, return to steps S3 and S4 to adjust the task sequence or motion parameters until the requirements are met. Step S7: Apply the optimized solution to the actual production line and conduct physical verification for 100 production cycles to ensure that the error between the actual total cycle time and the simulation data is ≤1%, and that the solder joint temperature is maintained at 1800-2200℃ and the welding gun pressure fluctuation is ≤±5%.

2. The method for optimizing the cycle time of an automotive robot resistance spot welding robot according to claim 1, characterized in that: The tooling fixture mentioned in step S1 includes a floor positioning fixture, a roof fixing device, and a side panel clamping mechanism. The positioning accuracy of the floor positioning fixture is ±0.1mm, and the clamping force fluctuation of the side panel clamping mechanism is ≤±3%.

3. The method for optimizing the cycle time of an automotive robot resistance spot welding robot according to claim 1, characterized in that: When collecting production data in step S2, the robot joint current, welding torch electrode pressure, and welding time of the weld point are recorded simultaneously. The average value of valid data within 24 hours and data during equipment downtime periods is taken as the current time parameter.

4. The method for optimizing the cycle time of an automotive robot resistance spot welding robot according to claim 1, characterized in that: The task node logical relationship of the technical review diagram in step S3 satisfies the following sequence: "floor positioning fixture clamping", "floor welding", "side panel clamping", "side panel welding", and "roof welding".

5. The method for optimizing the cycle time of an automotive robot resistance spot welding robot according to claim 1, characterized in that: The collision detection function described in step S5 covers the robot body, welding torch, tooling fixtures, and the car body to be welded in the truck cab. The detection scenario covers the robot's starting position, intermediate path nodes, and ending position.

6. The method for optimizing the cycle time of an automotive robot resistance spot welding robot according to claim 1, characterized in that: In step S6, during physical verification, an infrared thermometer is used to monitor the temperature of the weld joint in real time, and pressure data is recorded by a welding torch pressure sensor. The welding quality must meet the requirements for the strength of resistance spot welded joints in the ISO 14329 standard.

Images