Learning hydrocracking reaction dynamics via neural <scp>ODEs</scp> : A data‐driven, gradient‐interpretable lumped modelling framework
Bibliographic record
Abstract
Abstract This study applies neural ordinary differential equations (neural ODEs) to model hydrocracking kinetics, a key process for converting heavy hydrocarbons into lighter products like gasoline and diesel. Neural ODEs provide a data‐driven approach, learning reaction dynamics directly from data without requiring explicit assumptions on kinetics, addressing limitations in traditional methods. Two neural ODE models were trained on synthetic hydrocracking data representing different kinetic assumptions: one based on a 2.5‐order reaction scheme (Model A) and the other on a first‐order scheme (Model B), across varying temperatures and feedstocks. The models demonstrated high predictive accuracy when predicting within the range of training data, with RMSE values remaining below 0.5 wt.% under most conditions. However, performance declined during high‐temperature extrapolation scenarios, particularly for the higher‐order model, revealing challenges in capturing nonlinear dynamics at extreme conditions. This work also enhanced the interpretability of neural ODEs by analyzing gradients within the model, which validated alignment with known kinetic principles, uncovering critical information about reaction pathways and temperature sensitivities. This analysis demonstrated the models' ability to capture temperature‐dependent behaviour and rate stabilization, as illustrated through heat maps, which further emphasized the potential of neural ODEs for both predictive accuracy and interpretative insights in hydrocracking modelling. Additionally, the extracted gradients present an exciting avenue for future advancements, such as leveraging symbolic regression techniques to uncover governing equations.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".