Cutting with an abrasive water jet is a typical machining method for carbon fiber-reinforced plastics (CFRP). However, the cutting principal of multi-materials including CFRP has not been clarified. Here, the cutting principal with an abrasive water jet of a multi-material using CFRP and A5052 was investigated. The experimental results showed that the cutting mechanism of the multi-material was completely different when the multi-material was turned upside down. In addition, the surface roughness of A5052/CFRP was strongly influenced by the nozzle feed speed, the abrasive flow rate, and the pump pressure in comparison with CFRP/A5052.

The development of methods to monitor the state of polished surfaces during automated polishing would further optimize these processes. In this study, a machine learning model for estimation of surface roughness by extracting various features was constructed to estimate surface roughness by supervised learning based on images of the polished surface obtained with a camera loaded onto an articulated robot. The accuracy of the constructed model was verified by accuracy evaluation with cross-validation. Furthermore, the generality of constructed model was verified by estimating the surface roughness of unknown polished surfaces. The results verified the effectiveness of the constructed model and showed that features of polished surfaces could be extracted accurately.

This paper discusses the mechanism of groove generation on small bore internal surfaces using ultrasonic assisted grinding for texturing. Single-diamond grain tools were manufactured and single grain ultrasonic assisted internal grinding experiments were performed. The influence of ultrasonic vibration on microgrinding groove formation in macrogrinding grooves was confirmed. The grain cutting edge form and cross-section of the bottom of the microgrinding groove were confirmed. The material removal effect was improved and short convergence time of grinding force was achieved in ultrasonic assisted grinding by improving grinding conditions.