Mechanical properties of a high-strength cupronickel alloy-Bayesian neural network analysis

In this work the mechanical properties of a highly alloyed cupronickel have been analyzed using a neural network technique within a Bayesian framework. In this way the mechanical properties can be represented as an empirical function of the compositional variables. This method has been used to analyze the relative contributions of the various elements to the mechanical properties. Whilst the method is entirely empirical, it will be shown that the predictions made are of metallurgical significance.