Towards Material-Batch-Aware Tool Condition Monitoring

In subtractive manufacturing, process monitoring systems are used to observe the manufacturing process, predict maintenance actions and suggest optimizations. One challenge, however, is that observable signals influenced not only by degradation of cutting tool, but also deviations in machinability among material batches. Thus it necessary first respective batch before making decisions. this study, an approach shown for batch-aware tool condition using feature extraction unsupervised learning analyze high-frequency control data order detect clusters materials with different machinability, subsequently optimize process. This validated experiments implemented as edge framework.