Prediction of Spectral Response for Explosion Separation Based on DeepONet

Strong shock waves generated during the pyrotechnic separation process of aerospace vehicles can cause high-frequency damage or even structural failure to the vehicle’s structure. Existing structural designs for shock attenuation typically rely on shock response spectra methods, which require multip...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaoqi Chen, Zhanlong Qu, Yuxi Wang, Zihao Chen, Ganchao Chen, Xiao Kang, Ying Li
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Aerospace
Subjects:
Online Access:https://www.mdpi.com/2226-4310/12/4/310
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850183746482864128
author Xiaoqi Chen
Zhanlong Qu
Yuxi Wang
Zihao Chen
Ganchao Chen
Xiao Kang
Ying Li
author_facet Xiaoqi Chen
Zhanlong Qu
Yuxi Wang
Zihao Chen
Ganchao Chen
Xiao Kang
Ying Li
author_sort Xiaoqi Chen
collection DOAJ
description Strong shock waves generated during the pyrotechnic separation process of aerospace vehicles can cause high-frequency damage or even structural failure to the vehicle’s structure. Existing structural designs for shock attenuation typically rely on shock response spectra methods, which require multiple finite element calculations to determine the optimal geometric parameters, leading to relatively low efficiency. In this work, we propose a spectral response prediction method for spacecraft structures using the Deep Operator Network (DeepONet). This method preserves the physical relationships between input variables, modularizes geometric and positional input data, and outputs the spectral response. We integrate this neural model to analyze the impact of spacecraft structural parameters on shock resistance performance, revealing that circumferential reinforcement has the most significant influence on shock resistance. Then, we conduct a detailed analysis of the DeepONet model, noting that models with a higher number of neurons per layer train more quickly but are prone to overfitting. Additionally, we find that focusing on specific frequency bands for spectral response prediction yields more accurate results.
format Article
id doaj-art-e3cd54d6156449bb8dd61002b3a7b38e
institution OA Journals
issn 2226-4310
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Aerospace
spelling doaj-art-e3cd54d6156449bb8dd61002b3a7b38e2025-08-20T02:17:14ZengMDPI AGAerospace2226-43102025-04-0112431010.3390/aerospace12040310Prediction of Spectral Response for Explosion Separation Based on DeepONetXiaoqi Chen0Zhanlong Qu1Yuxi Wang2Zihao Chen3Ganchao Chen4Xiao Kang5Ying Li6School of Naval Architecture, Ocean and Energy Power Engineering, Wuhan University of Technology, Wuhan 430064, ChinaChina Academy of Launch Vehicle Technology, Beijing 100076, ChinaInstitute of Advanced Structure Technology, Beijing Institute of Technology, Beijing 100081, ChinaInstitute of Advanced Structure Technology, Beijing Institute of Technology, Beijing 100081, ChinaInstitute of Advanced Structure Technology, Beijing Institute of Technology, Beijing 100081, ChinaInstitute of Advanced Structure Technology, Beijing Institute of Technology, Beijing 100081, ChinaInstitute of Advanced Structure Technology, Beijing Institute of Technology, Beijing 100081, ChinaStrong shock waves generated during the pyrotechnic separation process of aerospace vehicles can cause high-frequency damage or even structural failure to the vehicle’s structure. Existing structural designs for shock attenuation typically rely on shock response spectra methods, which require multiple finite element calculations to determine the optimal geometric parameters, leading to relatively low efficiency. In this work, we propose a spectral response prediction method for spacecraft structures using the Deep Operator Network (DeepONet). This method preserves the physical relationships between input variables, modularizes geometric and positional input data, and outputs the spectral response. We integrate this neural model to analyze the impact of spacecraft structural parameters on shock resistance performance, revealing that circumferential reinforcement has the most significant influence on shock resistance. Then, we conduct a detailed analysis of the DeepONet model, noting that models with a higher number of neurons per layer train more quickly but are prone to overfitting. Additionally, we find that focusing on specific frequency bands for spectral response prediction yields more accurate results.https://www.mdpi.com/2226-4310/12/4/310explosive separationDeepONetfrequency domain machine learningneural network analysis
spellingShingle Xiaoqi Chen
Zhanlong Qu
Yuxi Wang
Zihao Chen
Ganchao Chen
Xiao Kang
Ying Li
Prediction of Spectral Response for Explosion Separation Based on DeepONet
Aerospace
explosive separation
DeepONet
frequency domain machine learning
neural network analysis
title Prediction of Spectral Response for Explosion Separation Based on DeepONet
title_full Prediction of Spectral Response for Explosion Separation Based on DeepONet
title_fullStr Prediction of Spectral Response for Explosion Separation Based on DeepONet
title_full_unstemmed Prediction of Spectral Response for Explosion Separation Based on DeepONet
title_short Prediction of Spectral Response for Explosion Separation Based on DeepONet
title_sort prediction of spectral response for explosion separation based on deeponet
topic explosive separation
DeepONet
frequency domain machine learning
neural network analysis
url https://www.mdpi.com/2226-4310/12/4/310
work_keys_str_mv AT xiaoqichen predictionofspectralresponseforexplosionseparationbasedondeeponet
AT zhanlongqu predictionofspectralresponseforexplosionseparationbasedondeeponet
AT yuxiwang predictionofspectralresponseforexplosionseparationbasedondeeponet
AT zihaochen predictionofspectralresponseforexplosionseparationbasedondeeponet
AT ganchaochen predictionofspectralresponseforexplosionseparationbasedondeeponet
AT xiaokang predictionofspectralresponseforexplosionseparationbasedondeeponet
AT yingli predictionofspectralresponseforexplosionseparationbasedondeeponet