The invention relates to the technical field of novel
metal materials, in particular to a novel light high-strength single-phase
refractory high-entropy
alloy based on
machine learning design and a design and preparation method thereof, the
alloy is composed of five elements of Cr, Mo, Ti, V and W, and the composition ranges (at.%) of the
alloy are as follows: 5-25% of Cr, 5-15% of Mo, 25-35% of Ti, 20-35% of V and 5-20% of W. The invention further provides a design method of the single-phase
refractory high-entropy alloy, according to the method, the yield strength of the single-phase
refractory high-entropy alloy is predicted through a
machine learning regression model, and the
valence electron concentration and density are calculated through a
mixing rule. And drawing a performance map based on the prediction data, and screening out a target component from the performance map for
microstructure and
mechanical property characterization. The as-cast structure of the high-entropy alloy provided by the invention is a single-phase BCC structure, and the high-entropy alloy has
low density, high yield strength and good compression
plasticity at
room temperature.