CSTNet: A Dual-Branch Convolutional Neural Network for Imaging of Reactive Flows Using Chemical Species Tomography
Source check: Source checked. Not independent scientific confirmation.
Finding: CSTNet reconstructs species concentration and temperature from chemical-species tomography. The paper reports improved accuracy and noise robustness on simulated data and measured real-flame experiments using a low-complexity optical sensor; broader transfer to other processes is not established.
Technology: CSTNet, a dual-branch convolutional neural network using tomography measurements and sensor layout
Publication: Peer-reviewed paper · Test context: direct test. Catalog categories, not an evidence grade.
- AI contribution
- A dual-branch CNN jointly reconstructs concentration and temperature fields from chemical-species tomography measurements and sensor layout.
- Test and result
- CSTNet reconstructed concentration and temperature fields more accurately and robustly than comparison methods, including measured real-flame data.
- Limitation / still unknown
- Simulated and experimental flame evaluation is reported; transfer to other flow settings is not shown.
Evidence sources and annotation context
Evidence context (descriptions follow the cited sources; test type is a catalog classification):
Evidence sources: https://researchonline.gcu.ac.uk/files/60481121/TNNLS_2020_P_CSTNet_A_Dual_Branch_CNN.pdf
Source checked · editorial details
The source page was checked against this entry. Independent confirmation is not assessed by this label.
Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.
Review date: 22 Nov 2026. Review scope: Primary author-manuscript full text, Methods and Results; first-version chronology checked against the indexed journal record..
No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.
Editorial assessment: Impact 3/5 · Validation 3/5 · Novelty 4/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.