Cutting tool condition monitoring using multiple sensors and artificialintelligence techniques on a computer numerical controlled milling machine

Abstract

This work documents an investigation of the degradation of a variety of different tools whilst conducting milling operations on a computer numerical controlled (CNC) milling machine. The potential of a range of sensors to detect tool degradation has been investigated and the outputs have been incorporated into a monitoring system. Progressive degradation under nominal rough and finish face milling and rough groove milling has been investigated using a two point grooving tool and four and eight point face milling tools on En8, En24 and En24T workpiece materials. Rapid degradation of the cutting tool has also been observed under rough milling conditions using four and eight point face milling tools, whilst machining n8 and En24T materials in a variety of simulated and actual tool breakage situations. A limited investigation of the effect of the individual wear geometries associated with both progressive and instantaneous tool degradation has been conducted by simulating these geometries and carrying out rough miffing tests using a four point face milling tool on a workpiece of En8 material. Similarly, a limited investigation of the effect of machining on different machines has also been undertaken. A number of different sensing technologies have been used, including conventional sensors such as spindle current and cutting force but also novel sensing techniques such as Acoustic Emission. These have been combined using artificial intelligence techniques to provide automatic recognition of the tool wear state. Similarly, the feasibility of breakage detection/prediction has also been demonstrated.