Rosmin, Arby Pratama Putra (2026) Pengembangan dan Optimisasi Desain 3D Cathode Flow Field untuk Meningkatkan Kinerja Proton Exchange Membrane Fuel Cell Menggunakan Computational Fluid Dynamics, Deep Neural Network, dan Pelican Optimization Algorithm. Masters thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 6011232006-Master_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
6011232006-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (18MB) | Request a copy |
Abstract
Proton Exchange Membrane Fuel Cell (PEMFC) merupakan perangkat konversi energi yang menghasilkan listrik beremisi rendah melalui reaksi elektrokimia antara hidrogen dan oksigen. Salah satu komponen penting yang menentukan kinerja PEMFC adalah flow field pada bipolar plate (BP), karena berperan dalam meningkatkan perpindahan massa, menyeragamkan distribusi konsentrasi reaktan, serta menjaga keseimbangan manajemen air. Pada penelitian ini, desain flow field katoda tiga dimensi (3D) dikembangkan dan dioptimisasi menggunakan pendekatan Computational Fluid Dynamics (CFD), Deep Neural Network (DNN) sebagai surrogate model, serta Pelican Optimization Algorithm (POA) untuk optimasi parameter desain. Pada tegangan 0,6 V, desain pembanding yaitu stepped flow field menghasilkan kerapatan arus sebesar 1,45707 A/cm², sedangkan desain yang dikembangkan mencapai 1,63283 A/cm² atau meningkat sekitar 12%. Selain peningkatan kinerja elektrokimia, desain ini juga memperbaiki manajemen air dengan menurunkan nilai maksimum liquid saturation serta memperluas area liquid saturation rendah pada zona reaksi, sehingga risiko flooding berkurang secara signifikan dan distribusi konsentrasi reaktan pada antarmuka GDL/CL meningkat. Di sisi lain, desain yang dioptimisasi menunjukkan penurunan pressure drop, yang berdampak positif terhadap peningkatan kepadatan daya bersih PEMFC akibat berkurangnya kebutuhan daya pompa. Dengan demikian, berdasarkan hasil simulasi CFD dan prediksi surrogate model, pendekatan CFD–DNN–POA menunjukkan potensi dalam menghasilkan desain flow field katoda 3D dengan kinerja yang lebih baik pada aspek kerapatan arus, manajemen air, dan efisiensi energi sistem PEMFC.
==================================================================================================================================
Proton exchange membrane fuel cell (PEMFC) is a low-emission energy conversion device that generate electricity through electrochemical reactions between hydrogen and oxygen. One of the key components governing PEMFC performance is the flow field design on the bipolar plate (BP), as it enhances mass transport, promotes reactant concentration uniformity, and maintains proper water management. In this study, a three-dimensional (3D) cathode flow field was developed and optimized using an integrated framework comprising Computational Fluid Dynamics (CFD), a Deep Neural Network (DNN) as a surrogate model, and the Pelican Optimization Algorithm (POA) for design-parameter optimization. At an operating voltage of 0.6 V, the baseline design which is stepped flow field produced a current density of 1.45707 A/cm², whereas the proposed design achieved 1.63283 A/cm², corresponding to an improvement of approximately 12%. In addition to enhanced electrochemical performance, the proposed flow field improved water management by reducing the maximum liquid saturation and expanding the low liquid-saturation region within the reaction zone, thereby significantly mitigating flooding risk and increasing reactant concentration distribution at the GDL/CL interface. Furthermore, the optimized design exhibited a reduced pressure drop, which is expected to improve the net power density of the PEMFC by lowering the pumping power requirement. Therefore, based on CFD simulations and surrogate-model predictions, the proposed CFD–DNN–POA approach demonstrates potential for producing an optimized 3D cathode flow field with improved current density, water management, and energy efficiency in PEMFC systems.
| Item Type: | Thesis (Masters) |
|---|---|
| Uncontrolled Keywords: | PEMFC, optimisasi flow field, DNN, POA, flow field optimization |
| Subjects: | Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. Q Science > Q Science (General) > Q337.3 Swarm intelligence Q Science > QA Mathematics > QA336 Artificial Intelligence T Technology > TA Engineering (General). Civil engineering (General) > TA357 Computational fluid dynamics. Fluid Mechanics T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2931 Fuel cells |
| Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Material & Metallurgical Engineering > 27101-(S2) Master Thesis |
| Depositing User: | Arby Pratama Putra Rosmin |
| Date Deposited: | 27 Jan 2026 01:43 |
| Last Modified: | 27 Jan 2026 01:43 |
| URI: | http://repository.its.ac.id/id/eprint/130644 |
Actions (login required)
 |
View Item |