PEMFC performance improvement through oxygen starvation prevention, modeling, and diagnosis of hydrogen leakage
Bibliographic record
Abstract
Catalyst degradation results in emerging pinholes in Proton Exchange Membrane Fuel Cells (PEMFCs) and subsequently hydrogen leakage. Oxygen starvation resulting from hydrogen leaks is one of the primary life-limiting factors in PEMFCs. Voltage reduces as a result of oxygen starvation, and the cell performance deteriorates. Starved PEMFCs also work as a hydrogen pump, increasing the amount of hydrogen on the cathode side, resulting in hydrogen emissions. Therefore, it is important to delay the occurrence of oxygen starvation within the Membrane Electrode Assembly (MEA) while simultaneously be able to diagnose the hydrogen crossover through the pinholes. In this work, first, we focus on catalyst configuration as a novel method to prevent oxygen starvation and catalyst degradation. It is hypothesized that the redistribution of the platinum catalyst can increase the maximum current density and prevent oxygen starvation and catalyst degradation. Therefore, a multi-objective optimization problem is defined to maximize fuel cell efficiency and to prevent oxygen starvation in the PEMFC. Results indicate that the maximum current density rises about eight percent, while the maximum PEMFC power density increases by twelve percent. In the next step, a previously developed pseudo two-dimensional model is used to simulate fuel cell behavior in the normal and the starvation mode. This model is developed further to capture the effect of the hydrogen pumping phenomenon and to measure the amount of hydrogen in the outlet of the cathode channel. The results obtained from the model are compared with the experimental data, and validation shows that the proposed model is fast and precise. Next, Machine Learning (ML) estimators are used to first detect whether there is a hydrogen crossover in the fuel cell and second to capture the amount of hydrogen cross over. K Nearest Neighbour (KNN) and Artificial Neural Network (ANN) estimators are chosen for leakage detection and classification. Eventually, a pair of ANN classifier-regressor is chosen to first isolate leaky PEMFCs and then quantify the amount of leakage. The classifier and regressor are both trained on the datasets that are generated by the pseudo two-dimensional model. Different performance indexes are evaluated to assure that the model is not underfitting/overfitting. This ML diagnosis algorithm can be employed as an onboard diagnosis system that can be used to detect and possibly prevent cell reversal failures.
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 | 0.000 |
| 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 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.000 |
| 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".