The application provides a
creep stress relaxation response model parameter calibration method and device based on a deep neural network, and relates to the technical field of computers. The method constructs a
creep stress relaxation simulation model and generates multiple sets of parameter samples, calculates and parameter fits the
creep stress relaxation response results corresponding to each set of parameter samples, and trains a parameter prediction model based on the fitted parameters, so that in the subsequent analysis stage,
finite element simulation calculation and parameter fitting processes do not need to be repeatedly performed, the
creep stress relaxation response under a target working condition can be obtained, the calculation time and the calculation
resource consumption are significantly reduced, and the analysis efficiency is improved; the
creep stress relaxation response results corresponding to the multiple sets of parameter samples are uniformly parameter fitted, a parameter prediction model is trained based on the fitted parameters, the automatic prediction of the
creep stress relaxation
response model parameters is realized, the uncertainty caused by the repeated adjustment of the
model parameters depending on artificial experience is reduced, and the parameter calibration efficiency is improved.