Fig. 2: Pairwise plot and distribution demonstrating correlations between five input parameters and two mechanical properties. | Nature Communications

Fig. 2: Pairwise plot and distribution demonstrating correlations between five input parameters and two mechanical properties.

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

Fig. 2: Pairwise plot and distribution demonstrating correlations between five input parameters and two mechanical properties.

a Correlations between five input parameters and UTS (in pink) and correlations between them and TE (in blue). The individual effects of each parameter were visualized without holding other parameters constant, due to the inherent variability and interdependencies within the initial dataset, which comprises data from multiple studies and experimental conditions. The shaded areas represent the distribution of data points, illustrating the variability and spread within each dataset. The solid lines indicate the fitted trends derived from linear regression of the distributions, capturing the overall relationship between variables. b Distribution of all Ti-6Al-4V samples in the initial dataset marked in gray, that of those with the top 10% of TE marked in blue, and that of those with the top 10% of UTS marked in red based on the values of laser power, HT temperature, and HT time. Source data are provided as Source Data file.

Back to article page