Artificial nueral network technology for the data processing of hydrogen attack

Artificial neural networks are forms of artificial intelligence that learn correlative patterns between input and output information without a specific model. Then they use the learned relationships to make predictions. Five back - propagation artificial neural networks were constructed to recognize certain relationships in hydrogen attack to predict the time to incubate fissuring of 0.5Mo and carbon steel. Given the two parameters, environmental temperature and hydrogen pressure, these neural network models can offer a prediction with certain accuracy. The feasibility of the model was verified by the data from papers. It is proved that the ANN technology is applicable to the evaluation of the complicated system of hydrogen attack, and also gives a base to develop the expert system and residual life evaluation system for hydrogen attack.