Optimising the manufacturing of a ß-Ti alloy produced via direct energy deposition using small dataset machine learning.

Successful additive manufacturing involves the optimisation of numerous process parameters that significantly influence product quality and manufacturing success. One commonly used criteria based on a collection of parameters is the global energy distribution (GED). This parameter encapsulates the energy input onto the surface of a build, and is a function of the laser power, laser scanning speed and laser spot size. This study uses machine learning to develop a model for predicting manufacturing layer height and grain size based on GED constituent process parameters. For both layer height and grain size, an artificial neural network (ANN) reduced error over the data set compared with multi linear regression. Layer height predictions using ANN achieved an R2 of 0.97 and a root mean square error (RMSE) of 0.03 mm, while grain size predictions resulted in an R2 of 0.85 and an RMSE of 9.68 µm. Grain refinement was observed when reducing laser power and increasing laser scanning speed. This observation was successfully replicated in another α + ß Ti alloy. The findings and developed models show why reproducibility is difficult when solely considering GED, as each of the constituent parameters influence these individual responses to varying magnitudes.

Author Correction: Additive manufacturing of ultrafine-grained high-strength titanium alloys.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Additive manufacturing of ultrafine-grained high-strength titanium alloys.

Additive manufacturing, often known as three-dimensional (3D) printing, is a process in which a part is built layer-by-layer and is a promising approach for creating components close to their final (net) shape. This process is challenging the dominance of conventional manufacturing processes for products with high complexity and low material waste1. Titanium alloys made by additive manufacturing have been used in applications in various industries. However, the intrinsic high cooling rates and high thermal gradient of the fusion-based metal additive manufacturing process often leads to a very fine microstructure and a tendency towards almost exclusively columnar grains, particularly in titanium-based alloys1. (Columnar grains in additively manufactured titanium components can result in anisotropic mechanical properties and are therefore undesirable2.) Attempts to optimize the processing parameters of additive manufacturing have shown that it is difficult to alter the conditions to promote equiaxed growth of titanium grains3. In contrast with other common engineering alloys such as aluminium, there is no commercial grain refiner for titanium that is able to effectively refine the microstructure. To address this challenge, here we report on the development of titanium-copper alloys that have a high constitutional supercooling capacity as a result of partitioning of the alloying element during solidification, which can override the negative effect of a high thermal gradient in the laser-melted region during additive manufacturing. Without any special process control or additional treatment, our as-printed titanium-copper alloy specimens have a fully equiaxed fine-grained microstructure. They also display promising mechanical properties, such as high yield strength and uniform elongation, compared to conventional alloys under similar processing conditions, owing to the formation of an ultrafine eutectoid microstructure that appears as a result of exploiting the high cooling rates and multiple thermal cycles of the manufacturing process. We anticipate that this approach will be applicable to other eutectoid-forming alloy systems, and that it will have applications in the aerospace and biomedical industries.