Fig. 4: Results of active learning iterations. | Nature Communications

Fig. 4: Results of active learning iterations.

From: Active learning framework to optimize process parameters for additive-manufactured Ti-6Al-4V with high strength and ductility

Fig. 4: Results of active learning iterations.

a First, (b) second, (c) third, (d) fourth, and (e) fifth iterations where yellow diamonds correspond to predicted properties and red triangles correspond to experiment results. f Ashby plot of UTS vs. TE for Ti-6Al-4V alloys from the initial dataset (supplementary data 1), DED-processed, and wrought Ti-6Al-4V samples (supplementary data 2). The shaded regions represent the property space of UTS and TE for different datasets, illustrating the distribution of data points across various processing methods. The orange region with square markers corresponds to data derived from the initial dataset. The blue region with triangle markers represents data obtained from DED. The green shaded area with pentagon markers indicates data from wrought Ti-6Al-4V samples. Lastly, the red circles denote the optimized properties achieved through the proposed framework. Source data are provided as Source Data file.

Back to article page