Robotic Multipass Welding with Adaptive Weld Fill Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional robotic welding systems face challenges due to the complexity of manufacturing tasks, variations in part dimensions, and the need for human-generated weld plans that are subjective and static, leading to errors in multipass welding operations.
Innovation Solution
A computer-implemented method generates a welding plan for a robotic system by identifying seams using sensors and neural networks, determining waypoints and weld profiles, and optimizing bead layers and parameters to create a validated weld fill plan, which is then executed by the robot.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If a weld plan is generated based on a single set of dimensions from a CAD model, then the welding process can be automated and executed by robots, but the weld plan becomes static and cannot adapt to parts with varying dimensions within acceptable tolerances, leading to errors in multipass welding operations
Solution Approach 1:
The weld plan is transformed from a static set of instructions into a dynamic system that continuously adapts to actual part dimensions. Sensors capture real-time dimensional data during welding operations, and the system dynamically adjusts welding parameters, bead placement, and pass sequences based on measured variations within acceptable tolerances, enabling the automated system to handle dimensionally variable parts effectively
Solution Approach 2:
A feedback mechanism is implemented where sensors continuously measure actual part dimensions during the welding process. This measurement data feeds back to the control system, which compares actual dimensions against the CAD model tolerances and automatically adjusts the weld plan parameters accordingly. This closed-loop feedback enables the automated welding system to compensate for dimensional variations without human intervention
2Manufacturing precision
If human experts manually generate weld plans based on experience and subjective criteria, then the weld plans can handle complex manufacturing tasks and variations, but the process becomes time-consuming and inconsistent between different individuals
Solution Approach 1:
The manual expert system is replaced with an automated intelligent system that uses sensors, neural networks, and optimization algorithms to generate weld plans. The system processes dimensional data from sensors and automatically determines welding parameters, bead placement strategies, and pass sequences, eliminating the time-consuming manual programming process while maintaining or improving weld quality through consistent application of optimized parameters
Solution Approach 2:
The system enables self-service weld plan generation where the robotic welding system automatically creates and adjusts its own welding instructions based on sensor measurements of the actual parts. The system uses embedded intelligence to autonomously determine optimal welding parameters and sequences without requiring external expert intervention, significantly reducing plan generation time while maintaining high manufacturing precision
3Device complexity
If conventional robotic welding systems use static weld plans, then the system complexity is reduced and easier to implement, but errors in the weld plan propagate and increase with each subsequent pass in multipass welding
Solution Approach 1:
The system implements dynamic adjustment of welding parameters between passes based on real-time sensor measurements of previously deposited beads and accumulated weld geometry. The control system continuously updates the weld plan to compensate for deviations from intended geometry, ensuring that errors do not propagate across multiple passes. This dynamic adaptation maintains reliability without requiring overly complex manual programming
Data Source
AI summary
This disclosure provides systems, methods, and apparatuses, including computer programs encoded on computer storage media, that provide for welding techniques for manufacturing robots, such multipass welding techniques for welding robots. For example, the welding techniques may enable generation of weld instructions based on a welding fill plan. The instructions may be generated based on a bead model or a table that indicates a wire feed speed, a travel speed, or a voltage. As another example, the techniques may enable generation of weld instructions based on the one or more dimensions of a seam. As another example, the techniques may enable generation of a joint model of a cross-section of a seam to be welded. The joint model may be generated based on a combination of a plurality of feature components to generate the joint model of the seam. Other aspects and features are also claimed and described.