PEMFC Operation Parameter Prediction Method Development

Wahrach, Alvin (2026) PEMFC Operation Parameter Prediction Method Development. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6007221026-Master_Thesis.pdf Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

This study examines the performance of Proton Exchange Membrane Fuel Cells (PEMFCs) under dynamic vehicular loads using real-world data from the Shell Eco-marathon 2024. Vehicle power output was calculated from GPS data, and PEMFC operating temperature was simulated. The AutoGluon machine learning framework was employed to predict PEMFC operating pressure based on dynamic conditions accurately. Results confirmed the model’s high predictive accuracy (R² ≈ 0.977) and revealed correlations between operating pressure, temperature, and power demand/acceleration. This research demonstrates the effective use of machine learning with real world data for analyzing and modelling PEMFC behaviour in dynamic vehicular applications, offering insights for system optimization.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	AutoGluon, Prediction Method, Machine Learning AutoGluon, Metode Prediksi, Machine Learning
Subjects:	Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21101-(S2) Master Thesis
Depositing User:	Alvin Wahracha
Date Deposited:	09 Feb 2026 00:42
Last Modified:	09 Feb 2026 00:42
URI:	http://repository.its.ac.id/id/eprint/132197

Actions (login required)

View Item