Aircraft auxiliary power device air entraining performance prediction method and system based on incremental learning
The method comprises the following steps: acquiring parameter data of the aircraft auxiliary power device under different working conditions in a test process of a high-altitude test bed; the method comprises the following steps: preprocessing parameter data under different working conditions by ado...
Saved in:
Main Authors | , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
05.03.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The method comprises the following steps: acquiring parameter data of the aircraft auxiliary power device under different working conditions in a test process of a high-altitude test bed; the method comprises the following steps: preprocessing parameter data under different working conditions by adopting empirical mode decomposition and an improved wavelet threshold denoising fusion algorithm, carrying out normalization processing, constructing an input matrix required by a model by utilizing a time sequence of normalized data, and dividing the input matrix into a training set, a verification set and a test set; performing model training and evaluation on the CNN-GRU hybrid model according to a training set and a verification set under the nth working condition obtained by preprocessing to obtain a CNN-GRU initial hybrid model; the CNN-GRU initial hybrid model is retrained according to the training set and the verification set obtained through preprocessing under the (n + 1) th working condition, in the retra |
---|---|
Bibliography: | Application Number: CN202311787728 |