A data-driven optimization method for pressure-resistant, stealthy and impact-resistant structures

CN118886287BActive Publication Date: 2024-12-10WUHAN UNIV OF TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411376100.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-30
Publication Date
2024-12-10
Estimated Expiration
2044-09-30

AI Technical Summary

Technical Problem

Traditional structural design methods are difficult to fully consider complex actual situations, resulting in low efficiency and insufficient optimization of pressure resistance, stealth and impact resistance performance, and lack of comprehensive consideration of various factors.

Method used

Using a data-driven method, combined with machine learning and optimization algorithms, through forward and reverse training models, systematically considering topology, geometry, density and rotation parameters, optimize voltage, stealth and impact resistance performance indicators, and improve design efficiency and performance .

Benefits of technology

The comprehensive optimization of structural performance has been achieved, the comprehensiveness and rationality of the design has been improved, and the pressure resistance, stealth and impact resistance have been significantly improved, ensuring the stability and safety of the design under extreme conditions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118886287B_ABST
    Figure CN118886287B_ABST
Patent Text Reader

Abstract

The present invention discloses a data-driven optimization method for a pressure-resistant stealth and shock-resistant structure, the method comprising: determining various parameters of the pressure-resistant stealth and shock-resistant structure, and obtaining a forward training model F1, a forward training model F2, a reverse training model G1 and a reverse training model G2 based thereon; independently training the forward training models F1 and F2 to form a forward training model for the pressure-resistant stealth and shock-resistant structure; directly combining the reverse training models G1 and G2 to form a reverse training model for the pressure-resistant stealth and shock-resistant structure, and training the two reverse training models simultaneously; using the output of the reverse training model for the pressure-resistant stealth and shock-resistant structure as the input of the forward training model for the pressure-resistant stealth and shock-resistant structure, comparing the target structural response and the predicted structural response, further improving the reverse training model for the pressure-resistant stealth and shock-resistant structure, and finally predicting the pressure-resistant stealth and shock-resistant structure through the reverse training model for the pressure-resistant stealth and shock-resistant structure. The present invention realizes the comprehensive optimization of the design of the pressure-resistant stealth and shock-resistant structure.
Need to check novelty before this filing date? Find Prior Art