Conventional high-temperature titanium alloys: A data-driven analysis and future directions

Conventional high-temperature titanium alloys, more commonly known as near-α alloys, find themselves in advanced aero engines for applications up to the temperatures of 600°C mainly as compressor components, namely blades, discs, shafts, cases, etc., owing to their superior combination of ambient and elevated-temperature mechanical properties and oxidation resistance. A plethora of alloys under various processing conditions have been investigated and reported in the literature. The immanent mechanical properties are known to be sensitive to chemical composition, thermomechanical processing, and microstructural constituents such as Ti₃Al, silicides, and others. We adopted a data-driven approach to analyze the mechanical properties of several grades and variants of high-temperature Ti alloys viz. IMI829, IMI 834, Ti-1100, TA 29 and so on reported in the literature to date. We applied a novel methodology combining advanced statistical analysis: cluster analysis (CA) and principal component analysis (PCA), and multiple-attribute decision making (MADM) to hear the voice of the data. Both objective (Shannon's entropy method) and subjective methods were adopted to evaluate the weights of various properties. PCA and CA not only consolidated the MADM ranks of the alloys but also grouped similar alloys. The rank assigned by several MADMs viz. selective additive weighting (SAW), WEDBA (weighed Euclidean distance-based approach), TOPSIS (Technique for the order of preference by similarity to ideal solution), and others were consistent. The investigation highlights similarities (and differences) across several grades/variants of the alloys, suggests potential replacement or substitute for existing alloys, and also provides directions for improvement and/or development of titanium alloys over the current ones to push out some of the heavier alloys and thus help in reducing the weight of the engine to advantage. This methodology further demonstrates that it is possible to try different weightage scenarios and decide on the optimum alloy(s) for specific intended application(s).