Application of Bayesian neural network to materials diagnosis and life assessment

The purpose of this paper is to introduce some examples of the application of Bayesian neural network to materials diagnosis and life assessment. The concept of Bayesian inference is added to the traditional neural network, enabling its reliable application to problems of materials diagnosis and life assessment. As an example, the fatigue crack growth rate of nickel base superalloys has been modeled as a function of some 51 variables, including stress intensity range Delta K, log Delta K, chemical composition, temperature, grain size, heat treatment, frequency, load waveform, atmosphere, R-ratio, the distinction between short crack growth and long crack growth, sample thickness and yield strength. The Bayesian method puts error bars on the predicted value of the rate and allows the significance of each individual factor to be estimated. In addition, it has been possible to estimate the isolated effect of particular variables such as the grain size, which cannot in practice be varied independently.