Abstract
Artificial intelligence techniques have been widely applied in many domains, such as image /sound/text recognition, manufacturing monitoring, etc. One of the requirements for an artificial intelligence modeling is massive datasets. However, it is often limited knowns in the beginning of the design phase.This paper studied the methods and the influence of building an artificial intelligence model from a limited number of inputs. The application of the artificial neural network (ANN) and the recurrent neural network (RNN) has been applied to the nonlinear mechanical FE, steady-state thermal FE and transient FE model, and a rather simple neural network model and accuracy/application of these models has been reported.