Creep fatigue life prediction method based on fused physical neural network

Abstract

The invention relates to a creep fatigue life prediction method based on a fusion physical neural network. The creep fatigue life prediction method comprises the following steps: S1, obtaining initial features and creep fatigue life of a target component material; S2, calculating extended features by fusing physical feature engineering; S3, carrying out sensitivity analysis on the input features and the output features; S4, preprocessing the input features and the output features; S5, building a deep neural network model; S6, building a fusion physical neural network model; S7, performing model training and parameter optimization; S8, predicting the creep fatigue life; and S9, updating the model. According to the creep fatigue life prediction method based on the fusion physical neural network, the fusion physical loss function is incorporated into the deep neural network model, and the output value range of the model is limited, so that physical constraints are incorporated into the neural network model, and the prediction result is more accurate.

Description

technical field [0001] The invention relates to the field of creep fatigue life prediction, and more particularly relates to a creep fatigue life prediction method based on a fusion physical neural network. Background technique [0002] During the operation of high-temperature equipment such as steam turbines and heat exchangers, there are working conditions such as start-up, shutdown, and steady-state operation, which will cause the accumulation of creep-fatigue damage and lead to equipment failure. Usually, fatigue damage is manifested as the initiation and propagation of transgranular cracks, while creep damage is manifested as the initiation, propagation and connection of creep holes. The interaction of the two failure mechanisms makes the creep-fatigue failure behavior very complex. In order to reduce economic losses and major threats caused by creep-fatigue failure, it is necessary to carry out creep-fatigue life prediction of high-temperature equipment and components...

AI Technical Summary

Problems solved by technology

Although machine learning has some applications in engineering, as a black box model, lack of physical constraints, people have doubts about its reliability

Images