preprints_ui: ycmn4_v1
Data license: ODbL (database) & original licenses (content) · Data source: Open Science Framework
| id | title | description | date_created | date_modified | date_published | original_publication_date | publication_doi | provider | is_published | reviews_state | version | is_latest_version | preprint_doi | license | tags_list | tags_data | contributors_list | contributors_data | first_author | subjects_list | subjects_data | download_url | has_coi | conflict_of_interest_statement | has_data_links | has_prereg_links | prereg_links | prereg_link_info | last_updated |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ycmn4_v1 | Age Ain’t Just a Number: Leveraging Recurrent Neural Networks to Predict Fuel Cell Degradation | The long-term durability is beside costs currently the biggest challenge for the commercialization of Proton Exchange Membrane Fuel Cells (PEMFC). Precise monitoring mechanisms that can identify anomalies at an early stage are essential to support the long-term operation. Cell voltage is a suitable monitoring parameter, but its prediction is challenging due to its non-linear dependence on numerous different operating conditions as well as on the history of the fuel cell. An existing machine learning approach divides the prediction task into two components: a stationary model and a conceptual degradation model. While the stationary model has been already successfully developed, the degradation model only has been proposed in theory. This work addresses this gap by developing and implementing a degradation model that complements the existing stationary model. A methodology is devised to calculate the targets for the degradation model based on the stationary model. Subsequently, a foundation model is trained using real data, testing different neural network structures. To ensure practical applicability, transfer learning is employed to evaluate the generalizability of the foundation model to three different data sets. The results demonstrate that the newly developed degradation model successfully integrates with the existing fuel cell monitoring approach, solving issues in prior work and leading to significant improvements in long-term fuel cell monitoring. | 2025-04-03T15:35:25.665904 | 2025-04-03T15:48:22.793738 | 2025-04-03T15:47:31.404737 | ecsarxiv | 1 | accepted | 1 | 1 | https://doi.org/10.1149/osf.io/ycmn4_v1 | CC-By Attribution-NonCommercial-NoDerivatives 4.0 International | Anomaly Detection; Digital Twin; Fuel Cells; Machine Learning; Monitoring | ["Anomaly Detection", "Digital Twin", "Fuel Cells", "Machine Learning", "Monitoring"] | Laurin Holz; Lukas Klass; Alexander Kabza; Frank Sehnke; Markus Hölzle | [{"id": "xrm83", "name": "Laurin Holz", "index": 0, "orcid": null, "bibliographic": true}, {"id": "f5sru", "name": "Lukas Klass", "index": 1, "orcid": "0000-0002-1379-3607", "bibliographic": true}, {"id": "b5xru", "name": "Alexander Kabza", "index": 2, "orcid": null, "bibliographic": true}, {"id": "sghwf", "name": "Frank Sehnke", "index": 3, "orcid": null, "bibliographic": true}, {"id": "t6cga", "name": "Markus H\u00f6lzle", "index": 4, "orcid": null, "bibliographic": true}] | Laurin Holz | Engineering; Physical Sciences and Mathematics; Computational Engineering; Energy; Computer Sciences; Fuel Cells; Artificial Intelligence and Robotics | [{"id": "5ae728ad4667e6000f98dd92", "text": "Engineering"}, {"id": "5ae728ae4667e6000f98dd9d", "text": "Physical Sciences and Mathematics"}, {"id": "5ae728af4667e6000f98dda5", "text": "Computational Engineering"}, {"id": "5ae728b24667e6000f98dde3", "text": "Energy"}, {"id": "5ae728b44667e6000f98de41", "text": "Computer Sciences"}, {"id": "5ae728b54667e6000f98de47", "text": "Fuel Cells"}, {"id": "5ae728b74667e6000f98de87", "text": "Artificial Intelligence and Robotics"}] | https://osf.io/download/67eeaae83ee99dbee16dde9e | 0 | not_applicable | not_applicable | [] | 2025-04-09T21:06:18.944586 |