Development of a deep neural network and empirical model for predicting local gas holdup profiles in bubble columns
Bibliographic record
Abstract
Abstract Estimating local gas holdup profiles in bubble columns is key for their performance evaluation and optimization, as well as for design and scale‐up tasks. Up to the current day, there are important limitations in the accuracy and range of applicability of the available models in literature. Two alternatives for the prediction of such local fields can be found in the application of empirical models and the development of deep neural networks (DNN). The main drawback preventing the application of these techniques in previous years was the availability of a large enough databank of local gas holdup experimental measurements. Advances over the last decades in measurement techniques have resulted enough data reported in literature to gather a significative databank for these models' development. A databank containing 1252 experimental points was gathered and used for the development of a quadratic model and a DNN with the rectified linear unit (ReLU) algorithm as the activation function and the adaptive moment estimation (ADAM) algorithm as the optimizer function. The quadratic model and the DNN allowed a highly accurate prediction of the local gas holdup profiles, exhibiting a MSE of 0.0013 and 0.0010, respectively, and an and for the quadratic model and the DNN, respectively. Furthermore, these developed models allowed for the estimation of the single and multi‐feature effects of the operation conditions, geometrical characteristics, and physical properties of the fluids, over the local gas holdup profiles. The two developed models exhibited an enhanced predictive quality when compared with other models available in literature.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".