Methodes de suivi de l'etat des machines-outils par palpage d'artefacts

Five-axis machine tools are key players in manufacturing which offer a significant reduction in the number of setups required by not only positioning but also orienting the cutting tool relative to the workpiece with three linear and two rotary axes. This allows high productivity for the machining of complex parts such as those found in the aerospace industry. Although, there are various methods to verify a five-axis machine tool's performance, they are mostly beyond the reach of the machine shops in terms of specialised personnel and implementation time.;This thesis aims to develop a new verification method of the volumetric state of a five-axis machine. The proposed method is based on the measurement of a new artefact design by the touch probe found on most machines. The proposed artefact is formed of master balls. It is reconfigurable and uncalibrated and allows flexibility in number and positions of the balls. A mathematical model generates the volumetric errors of the machine by identifying and separating the artefact and the probe setup errors from the probing data. The link errors are included in the mathematical model in order to prevent the setup errors unduly absorb a genuine machine error source. Furthermore, a volumetric error analysis separates the errors originating from the linear and the rotary axes by performing a least square adjustment between the measured and the predicted ball centres.;An uncertainty analysis on the estimated artefact geometry is performed. The probing repeatability, the stability of the artefact geometry and the link errors of the machine tool are included in the analysis.;The results show that the proposed method is capable of excluding the setup errors from the volumetric probing data. The artefact's estimated geometry is validated on a CMM with a maximum error norm of 7.4 mum. The largest norm of the remaining volumetric distortion is 112.1 mum. This error is composed of a maximum local distortion norm of 7.50 mum resulting from the linear axes and a maximum norm of linear and angular relocalisation of 103.89 mum and 54.14 murad, respectively, resulting from the linear and rotary axes.