What does this research mean for the field?

A hybrid Proper Orthogonal Decomposition (POD) and Long Short-Term Memory (LSTM) network framework accurately predicts unsteady hydrogen mixing and supersonic flow fields in scramjet engines with significantly lower computational cost than full-order simulations. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To evaluate a hybrid reduced-order model combining Proper Orthogonal Decomposition and Long Short-Term Memory networks for predicting unsteady hydrogen mixing and supersonic flow fields behind multi-lobe strut injectors in scramjet engines.

September 3, 2026Acta AstronauticaOpen Access

Prediction of Hydrogen Mixing and Velocity Fields Behind a Strut Injector with Multi-Lobe Nozzles at scramjet engine using POD+LSTM technique

View Full Paper

Ask AI

Bookmark

Authors

MGM. Barzegar Gerdroodbary ISIman Shiryanpoor JPJosé Páscoa

Discussion

Member takes

Overview

Simulation study demonstrates accurate prediction of supersonic hydrogen mixing downstream of multi-lobe strut injectors, indicating high efficiency for scramjet engine design.

Key Points

To evaluate a hybrid reduced-order model combining Proper Orthogonal Decomposition and Long Short-Term Memory networks for predicting unsteady hydrogen mixing and supersonic flow fields behind multi-lobe strut injectors in scramjet engines.
Generated high-fidelity simulation datasets using Unsteady Reynolds-averaged Navier–Stokes (URANS) models for 2-lobe, 3-lobe, and 4-lobe injector configurations under supersonic, non-reacting hydrogen injection.
Extracted dominant spatial flow modes via Proper Orthogonal Decomposition and modeled temporal modal coefficients using Long Short-Term Memory networks across 70%, 80%, and 90% training-to-testing splits.
The POD–LSTM framework produced accurate velocity and scalar field predictions when trained on at least 80% of the dataset, achieving near-exact reconstruction at 90% training.
The reduced-order model faithfully captured jet penetration, shock structures, and shear mixing layer growth across complex multi-lobe configurations with substantially reduced computational requirements compared to full simulations.

Cite This Study

Gerdroodbary et al. (2026) studied this question.

synapsesocial.com/papers/6a99352e636c6408cfa7d2fa https://doi.org/10.1016/j.actaastro.2026.08.076

View Full Paper

Ask AI

Bookmark

September 3, 2026Acta AstronauticaOpen Access

Prediction of Hydrogen Mixing and Velocity Fields Behind a Strut Injector with Multi-Lobe Nozzles at scramjet engine using POD+LSTM technique

View Full Paper

Ask AI

Bookmark

Authors

MGM. Barzegar Gerdroodbary ISIman Shiryanpoor JPJosé Páscoa

Discussion

Member takes

Overview

Simulation study demonstrates accurate prediction of supersonic hydrogen mixing downstream of multi-lobe strut injectors, indicating high efficiency for scramjet engine design.

Key Points

To evaluate a hybrid reduced-order model combining Proper Orthogonal Decomposition and Long Short-Term Memory networks for predicting unsteady hydrogen mixing and supersonic flow fields behind multi-lobe strut injectors in scramjet engines.
Generated high-fidelity simulation datasets using Unsteady Reynolds-averaged Navier–Stokes (URANS) models for 2-lobe, 3-lobe, and 4-lobe injector configurations under supersonic, non-reacting hydrogen injection.
Extracted dominant spatial flow modes via Proper Orthogonal Decomposition and modeled temporal modal coefficients using Long Short-Term Memory networks across 70%, 80%, and 90% training-to-testing splits.
The POD–LSTM framework produced accurate velocity and scalar field predictions when trained on at least 80% of the dataset, achieving near-exact reconstruction at 90% training.
The reduced-order model faithfully captured jet penetration, shock structures, and shear mixing layer growth across complex multi-lobe configurations with substantially reduced computational requirements compared to full simulations.

Cite This Study

Gerdroodbary et al. (2026) studied this question.

synapsesocial.com/papers/6a99352e636c6408cfa7d2fa https://doi.org/10.1016/j.actaastro.2026.08.076

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark