Physics-Informed Neural Network Surrogate Model for Small-Sample Mistuned Bladed Disk

What is it about?

The data-driven surrogate model for mistuned bladed disk often needs massive data to get good accuracy. However, the data are difficult to obtain due to the complexity of the experiment. We build a new surrogate model based on Physics-Informed Neural Network to accurately predict the response of a bladed disk using only a small number of samples.

Featured Image