| Abstract Scope |
A novel approach is presented to enhancing weld quality control in resistance spot welding (RSW) through the development of robust and expansible machine learning (ML) frameworks. By harnessing the power of machine learning, we have developed the ability to ensure every aspect of the welding process, from the initial process design stage to the final weld joint quality. The frameworks operate by analyzing a variety of data streams, including in-line process signals, sensing signals, and postprocessed weld joint data. Through this analysis, the models have been trained to detect deviations from optimal quality standards, leveraging their ability to identify signature data patterns and anomalies within in-line signals and construct complex correlations between these signals and weld quality parameters. Meanwhile, the machine learning framework is designed to seamlessly adapt to a variety of materials, including high strength steels and aluminum alloys, etc. Its flexible architecture facilitates the incorporation of diverse data sources and features, enabling precise modeling and prediction across a broad range of material properties and weld quality variables. The framework is not bound by the limitations of traditional modeling approaches. Instead, it embraces the complexity inherent in welding processes, allowing us to capture aspects that were previously overlooked. The expansible ML frameworks represent a promising transformation in weld quality monitoring and control, empowering industry to achieve high levels of efficiency, consistency, and reliability in manufacturing. |