Product Development and Process Optimisation

In AM, product development and process optimisation are strongly correlated. This is due to the fact that a change in the product design can have a significant effect on the build of the part itself and on the different deviations or defects that can arise during the process. In this context, the accurate acquisition of holistic 3D models representing the geometry of actual products and their characteristics (for example, dimensions, structural deviations, surface defects and internal porosity) is necessary to improve the quality of the manufacturing technology and the final product. Holistic models are also needed to improve process simulations (for example, structural simulations that can be used to improve the part design or process simulations that are used to tune the process parameters for improving the final product properties) and process modelling, which are - together with AM itself - enabling requirements to realise Industry 4.0 (see Chapters 3 and 4). As specified in the previous sections, CT allows the reconstruction of 3D high-density digitisation models of AM parts, including both accessible and non-accessible geometries and features; hence, it is an optimal tool to be used directly by product designers and engineers in the above-depicted framework.

CT for Product Development

The product development process can benefit from the combination of AM and CT, because designers are no longer constrained by the limitations of conventional manufacturing processes and conventional measuring techniques. In fact, AM allows the fabrication of parts with highly complex shapes and structures that can subsequently be verified by CT. One of the main advantages of CT for product development is the possibility of obtaining holistic information on external as well as internal structures, and subsequently to export the scanned volume in a manner which enables direct use by product designers and engineers, for example, to re-engineer the manufactured parts (Akdogan et al. 2019). The use of AM and CT for reverse engineering purposes is another relevant aspect to be taken into account when dealing with product development (Favata et al. 2018). The CT reconstructed volume of a real part - for which the nominal model or drawings are not available - can be converted into an STL model and used to print a prototype using an AM machine. Both the CT virtual model and AM physical prototype can be used as starting points to improve the design of that part. Several design iterations can be based on this approach: re-designed parts can be printed again using AM and scanned with CT to be compared with the initial model in order to investigate experimentally the improvements. For example, Bauer et al. (2019) showed a full process flow for the reverse engineering of an AM turbine foil, using CT for acquiring the inner features in combination with optical CMS for measuring the external surface.

Product design can also be improved using CT models to feed software for mechanical/ structural simulations, with the benefit of basing the simulations not on an ideal geometry but on the actual one (including discrepancies with respect to the nominal model, structural deviations and internal defects). For example, Dallago et al. (2019) used CT models of as-built lattice structures as input for finite element analyses of the stress distribution at the strut junctions. They found that as-built junctions induce a higher stress concentration with respect to the nominal one, with results that are closer to those obtained from experimental analyses.

Another example of CT-based improvement of AM design is the use of CT for scanning biological samples to learn from nature how to build light but stable and rigid structures or other biologically inspired functional designs. This approach is already being applied for many AM parts, and the number of publications combining AM and design, inspired by nature, is rapidly increasing (du Plessis and Broeckhoven 2019, du Plessis et al. 2019b, du Plessis et al. 2018b).

 
Source
< Prev   CONTENTS   Source   Next >