SPP2402 – Development of a grey box model for predicting the performance of PVD-coated carbide tools

Although the majority of machining processes are carried out using geometrically defined cutting edges with coated carbide tools, the actual, complex application behaviour of these tools cannot be satisfactorily measured or adequately described using models based on the current state of research. In order to be able to carry out knowledge-based qualification of coated tools, knowledge of the onset of failure, wear progression and remaining service life is required. In addition to the layer properties, the process-related loads also change as a result of tool wear. Since experimental approaches have so far provided only limited information about the load stresses, it is even more difficult to predict discontinuous layer failure.

Objectives

The overarching research goal is to understand the fundamental interactions between the transient load collective and the change in layer properties in order to derive highly accurate models for predicting the performance of coated cutting tools.

For this purpose, a model is being developed that follows the approach of combining white-box and black-box models into a grey-box model. In order to make manufacturing processes more productive, it is necessary to fundamentally understand the interactions that lead to damage and to translate them into models for highly accurate prediction. One way to achieve this is to combine white-box and black-box models to supplement purely physical models with statistical methods, thereby taking into account the highly complex interactions and thus enabling a more accurate prediction of the performance of coated carbide tools.

Benefits

In order to make manufacturing processes more productive, it is necessary to fundamentally understand the interactions that lead to damage and translate them into models for highly accurate predictions. One way to achieve this is to combine white-box and black-box models to supplement purely physical models with statistical methods, thereby taking into account the highly complex interactions and thus enabling a more accurate prediction of the performance of coated carbide tools.

Approach

The research project addresses the dynamic relationship between changing layer properties on the one hand and changing process-related loads on the other. By examining the relationship between loads and layer properties, it is possible to predict the further behaviour of the tools during use and gain new insights into the underlying mechanisms. Since the changed layer properties are difficult to map on the basis of physical relationships and use for such a prediction, data-driven approaches based on high-resolution analyses of coating properties and residual stresses are used here.

Are you also interested in a cooperation project?

Contact Malte Kraeft via email at kraeft@ifw.uni.hannover.de or by phone at +49 511 762 12321.