Optimising properties of steel

We demonstrate a neural network that designs the composition and heat treatment of a steel that fulfils the target physical properties. The neural network model was trained from a library of experimental data from 1000 alloys.

In the first panel below please set the target yield stress, ultimate tensile strength, and elongation. Then click predict to get the neural network to design the optimal composition and heat treatment.

Click here to use this technology to predict the yield stress, ultimate tensile strength, and elongation of a steel.

This same technology was used to design nickel alloys where the composition covered 20 elements, 5 heat treatment parameters, and 11 material properties were optimised. Click here to read more about this study.

Input properties - unknowns