Application of random threshold neural networks for diagnostics of micro machine tool condition

Abstract

Micro equipment based manufacturing requires a higher automation level than traditional manufacturing, because a human operator should be able to supervise the parallel operation of many micro equipment units. We briefly describe the prototypes of micro machine tools and the random threshold neural network classifiers created by us for the experimental investigation of problems on the way to fully automated micro manufacturing. The problems of diagnostics of micro machine tools and machining modes are discussed as well as the results of experiments with the acoustic diagnostics of cutting modes using the random threshold classifier.