A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis Functions
The goal of increasing efficiency and durability of fuel cells can be achieved through optimal control of their operating conditions. In order to implement such controllers, accurate and computationally efficient fuel cell models must be developed. This work presents a hybrid (physics-based and data...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/17/2/508 |
| _version_ | 1827372376176721920 |
|---|---|
| author | José Agustín Aguilar Damien Chanal Didier Chamagne Nadia Yousfi Steiner Marie-Cécile Péra Attila Husar Juan Andrade-Cetto |
| author_facet | José Agustín Aguilar Damien Chanal Didier Chamagne Nadia Yousfi Steiner Marie-Cécile Péra Attila Husar Juan Andrade-Cetto |
| author_sort | José Agustín Aguilar |
| collection | DOAJ |
| description | The goal of increasing efficiency and durability of fuel cells can be achieved through optimal control of their operating conditions. In order to implement such controllers, accurate and computationally efficient fuel cell models must be developed. This work presents a hybrid (physics-based and data-driven), control-oriented model for approximating the output voltage of proton exchange membrane fuel cells (PEMFCs) while operating under dynamical conditions. First, a physics-based model, built from simplified electrochemical, membrane dynamics and mass conservation equations, is developed and validated through experimental data. Second, a data-driven, neural network (echo state network) is trained, fitted and tested with the same dataset. Then, the hybrid model is formed as a parallel structure, where the simplified physics-based model and the trained data-driven model are merged through an algorithm based on Gaussian radial basis functions. The merging algorithm compares the output of both single models and assigns weights for computing the prediction of the hybrid result. The proposed hybrid model structure is successfully trained, validated and tested with an experimental dataset originating from fuel cells within an automotive PEMFC stack. The hybrid model is assessed through the mean square error index, with the result of a low tracking error. |
| first_indexed | 2024-03-08T10:58:42Z |
| format | Article |
| id | doaj.art-ec6a8659e80b4751a4458574b25a66de |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-08T10:58:42Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-ec6a8659e80b4751a4458574b25a66de2024-01-26T16:21:59ZengMDPI AGEnergies1996-10732024-01-0117250810.3390/en17020508A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis FunctionsJosé Agustín Aguilar0Damien Chanal1Didier Chamagne2Nadia Yousfi Steiner3Marie-Cécile Péra4Attila Husar5Juan Andrade-Cetto6Institut de Robòtica i Informàtica Industrial, Consejo Superior de Investigaciones Científicas-Universitat Politèctnica de Catalunya, Llorens Artigas 4-6, 08028 Barcelona, SpainInstitut FEMTO-ST, Université de Franche-Comté, UTBM CNRS, F-90000 Belfort, FranceInstitut FEMTO-ST, Université de Franche-Comté, UTBM CNRS, F-90000 Belfort, FranceInstitut FEMTO-ST, Université de Franche-Comté, UTBM CNRS, F-90000 Belfort, FranceInstitut FEMTO-ST, Université de Franche-Comté, UTBM CNRS, F-90000 Belfort, FranceInstitut de Robòtica i Informàtica Industrial, Consejo Superior de Investigaciones Científicas-Universitat Politèctnica de Catalunya, Llorens Artigas 4-6, 08028 Barcelona, SpainInstitut de Robòtica i Informàtica Industrial, Consejo Superior de Investigaciones Científicas-Universitat Politèctnica de Catalunya, Llorens Artigas 4-6, 08028 Barcelona, SpainThe goal of increasing efficiency and durability of fuel cells can be achieved through optimal control of their operating conditions. In order to implement such controllers, accurate and computationally efficient fuel cell models must be developed. This work presents a hybrid (physics-based and data-driven), control-oriented model for approximating the output voltage of proton exchange membrane fuel cells (PEMFCs) while operating under dynamical conditions. First, a physics-based model, built from simplified electrochemical, membrane dynamics and mass conservation equations, is developed and validated through experimental data. Second, a data-driven, neural network (echo state network) is trained, fitted and tested with the same dataset. Then, the hybrid model is formed as a parallel structure, where the simplified physics-based model and the trained data-driven model are merged through an algorithm based on Gaussian radial basis functions. The merging algorithm compares the output of both single models and assigns weights for computing the prediction of the hybrid result. The proposed hybrid model structure is successfully trained, validated and tested with an experimental dataset originating from fuel cells within an automotive PEMFC stack. The hybrid model is assessed through the mean square error index, with the result of a low tracking error.https://www.mdpi.com/1996-1073/17/2/508PEMFChybrid modelESNradial basis functions |
| spellingShingle | José Agustín Aguilar Damien Chanal Didier Chamagne Nadia Yousfi Steiner Marie-Cécile Péra Attila Husar Juan Andrade-Cetto A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis Functions Energies PEMFC hybrid model ESN radial basis functions |
| title | A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis Functions |
| title_full | A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis Functions |
| title_fullStr | A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis Functions |
| title_full_unstemmed | A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis Functions |
| title_short | A Hybrid Control-Oriented PEMFC Model Based on Echo State Networks and Gaussian Radial Basis Functions |
| title_sort | hybrid control oriented pemfc model based on echo state networks and gaussian radial basis functions |
| topic | PEMFC hybrid model ESN radial basis functions |
| url | https://www.mdpi.com/1996-1073/17/2/508 |
| work_keys_str_mv | AT joseagustinaguilar ahybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT damienchanal ahybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT didierchamagne ahybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT nadiayousfisteiner ahybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT mariececilepera ahybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT attilahusar ahybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT juanandradecetto ahybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT joseagustinaguilar hybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT damienchanal hybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT didierchamagne hybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT nadiayousfisteiner hybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT mariececilepera hybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT attilahusar hybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions AT juanandradecetto hybridcontrolorientedpemfcmodelbasedonechostatenetworksandgaussianradialbasisfunctions |