Research Article
A Reduced Order Model Based on ANN-POD Algorithm for Steady-State Neutronics and Thermal-Hydraulics Coupling Problem
Table 4
The settings of the training parameters of each BPNN.
| Physical quantity | | | | | | |
| Number of hidden layers | 1 | 1 | 1 | 1 | 1 | 1 | Number of neurons at hidden layer | 1 | 1 | 1 | 1 | 1 | 4 | Training function | Levenberg–Marquardt | Transfer function in the hidden layer | tanh | Transfer function in the output layer | Identity | Learning rate | 0.1 | Learning goal | 0.00004 |
|
|