Materials Joining Research Group

Data-driven research and development has become increasingly important in the field of materials with the advances in AI technology. Our research group is trying to realize “manufacturing consulting” by proposing and developing various AI tools. As an example, we have developed an AI application that predicts material properties from microstructure images. Users can easily build prediction models from a database of images and their corresponding properties. These models can predict the target properties from any given microstructure image. In addition, the application can generate heatmaps that visualize the regions of the structure that the model considers most important for its predictions.

Ultrasonic welding is commonly used to join dissimilar metals. However, variations in energy transmission during the welding process or the formation of brittle intermetallic compounds can lead to inconsistencies in joint strength. This results in a significant loss of cost and time, requiring measures such as setting an excessive safety factor and inspecting all the parts. To address this issue, our research group has developed a technology that accurately determines the joint quality by analyzing acoustic data generated during the welding process. This technology offers a promising solution to reduce the inspection burden and improve manufacturing efficiency.