Modeling and Monitoring for Improved quality of Machining Processes: Development and Application of Error Compensation Strategies

This work develops the theme of optimisation in machining processes with particular focus on the integration of the precision engineering concept of determinism to improve the final product quality, in terms of reduced machining process costs and reduced costs to assess the part quality. The results achieved in this work contributes to the recent trends of digitalisation and implementation of zero-defect manufacturing.

The fundamental objective is modelling the part errors generated by the cutting process loads, such as cutting forces and temperature gradients correlated exclusively to the surface generation action. With the part error prediction, a compensation algorithm is built capable of realising a priori correction of the process errors, applying a modification to the tool trajectory.
With regard to the cutting force modelling and trajectory compensation strategy, this work presents the application of the approach to a portable 3-axis CNC machine and to 5-axis machining centre for precision machining operation.

The portable machine tool has been designed to perform machining operation on large wind turbine parts, in particular the wind turbine hub. A cutting force model has been developed for the particular grade of spheroidal cast iron of which the wind turbine component is made of. An innovative approach is presented to calibrate a mechanistic force model in the case of significant chip segmentation/fragmentation. The part error prediction and the tool trajectory compensation have been experimentally verified on the portable machine; the process involved a thorough characterisation of the stiffness and positioning performances of the portable machine.

The implementation of the tool trajectory compensation approach for the 5-axis machining centre focused on the prediction of an entire machining sequence. As study case, the part error profile of a two-steps machining sequence has been predicted and experimentally measured. Furthermore, different compensation approaches have been experimentally validated. To complement the a priori error compensation strategy, a monitoring solution for in-line part quality assurance has been implemented for both the machine tools. To maintain a wide applicability range as force sensors the absorbed current by the axes motors arc used. The monitoring solution couples the cutting force model prediction ,vith the measured signal to perform more detailed evaluation of the cutting process conditions and the cutting process induced errors. The capability of the system for both the CNC machines have been experimentally validated.

With regard to the thermal input modelling in cutting operations, the work focused only the prediction of the heat flowing into the cutting tool. A sensori?:ed cutting tool is conceptualised and built. With the latter, the thermal input into the cutting tool in orthogonal cutting conditions is characterised for different cutting parameters, speed and feed rate. The heat flmving into the cutting tool is estimated using an inverse approach. Furthermore, a methodology to apply the obtained heat for intermittent cutting operations, such as milling, is presented and experimentally validated.