Aero-engine combustion chamber creep fatigue life prediction method

Abstract

The invention discloses an aero-engine combustion chamber creep fatigue life prediction method. The method comprises: obtaining historical data and an initial feature subset of a to-be-predicted aero-engine combustion chamber; training the extracted initial feature subset by adopting an iterative decision tree algorithm to obtain a creep fatigue life prediction model of the aero-engine combustion chamber, and predicting the creep fatigue life of the aero-engine combustion chamber by utilizing the creep fatigue life prediction model of the aero-engine combustion chamber obtained in the step 2; and performing model calculation by taking the aero-engine combustion chamber load spectrum obtained in the step 1 as an input quantity of a prediction model so as to predict the creep fatigue residual life of the aero-engine combustion chamber. The invention further discloses a system for predicting the creep fatigue life of the aero-engine combustion chamber.

Description

technical field [0001] The invention belongs to the field of deep learning algorithms, in particular to a method and system for predicting creep fatigue life of an aero-engine combustion chamber based on a progressive gradient regression tree. Background technique [0002] Engineering structures often fail due to creep-fatigue loads. Accurate prediction of structural creep-fatigue life progress is of great engineering application value for rationally formulating structural maintenance cycles. [0003] The traditional fatigue life prediction methods are all based on its unique failure mechanism, such as Manson-Coffin model for low cycle fatigue life prediction, Larson-Miller model for creep rupture life prediction and creep fatigue life prediction Time Fractional Model. Especially the prediction of creep fatigue life has attracted the attention of many researchers due to its complex interaction mechanism. The linear damage accumulation (LDS) principle is the most widely use...

AI Technical Summary

Problems solved by technology

The above machine learning methods are all shallow structural algorithms. In the case of limited computing units, due to insufficient learning ability, the ability to express multi-variable complex time series is limited. When making predictions, its generalization ability is limited and it is difficult to be more accurate. Fatigue life prediction of aeroengine combustor under multivariable conditions

Images