Application of Deep Learning Convolutional Neural Network for Spray Characterization
Bibliographic record
Abstract
The application of a machine learning artificial intelligence (AI) for spray characterization is investigated.Images of sprays in regions where droplet formation has not taken place, therefore not allowing any insight into the spray droplets themselves.The objective is to bridge the gap from the process of droplet formation to the characteristics of the spray that is produced at the end of this process.To achieve this, convolutional neural networks (CNN) are trained to classify images of sprays that were captured at different operating fluid pressures.Even though this is not directly characterizing the spray, it provides evidence for the potential of machine learning methods in spray characterization as distinctions are made prior to spray formation meaning CNNs are able to distinguish patterns in sprays prior to the droplet formation process, hence proving the possibility of bridging the aforementioned gap.Our models were able to accurately identify images of sprays taken at different operating pressures.Moreover, the convolutional neural networks were further analysed to understand how they were able to make these distinctions, that are not easily visible to the human eye.For this gradient class activation maps were determined to understand the inner workings of the convolutional neural networks.These gradient class activation mappings could prove useful in determining new physical patterns that were previously unknown, which could contribute to a better understanding of sprays and the droplet formation process.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".