Hydrogen Recovery Analyses Based on LOGSIG Activation Functions in Feedforward Neural Networks

DOI

10.2991/978-94-6463-589-8_36How to use a DOI?

Keywords

Hydrogen; Palm Kernel Shell; Activated Carbon; Feedforward; Artificial Neural Networks; PSA

Abstract

Hydrogen is a crucial intermediate product that is widely used in the petrochemical and oil sectors. Efficiently improving hydrogen recovery technology is crucial for achieving sustainable success and fulfilling environmental obligations, given the increasing demand for hydrogen and the growing urgency of environmental concerns. To solve this issue, a Feedforward Artificial Neural Network (FANN) model was developed to predict and improve the amount of hydrogen gas that can be made from palm kernel shell activated carbon (PKS-AC) syngas production. The study used 60 experimental data points to examine the impact of adsorption processes, such as adsorption, pressure equalisation, desorption, and re-pressurisation modes, on hydrogen recovery through pressure swing adsorption procedures. The optimisation process was performed using the MATLAB (R2017b) software, specifically the Neural Network (NN) tool, with a dataset containing adsorption pressure, duration of adsorption and blowdown time data. The study demonstrated that using the LOGSIG activation function achieved the smallest mean square error (MSE) of 0.00010 when 19 hidden neurons were utilised. The regression coefficients (R) for training, validation, and testing were 0.91598, 0.99042, and 0.91718, respectively. The utilisation of this model has the potential to facilitate the development of cost-effective and efficient designs for on separation of pressure swing adsorption processes.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Maimunatun Nawar Mohd Yazan
AU  - Ashraf Azmi
AU  - Muhammad Azan Tamar Jaya
AU  - Mohd Roslee Othman
AU  - Illi Khairunnisa binti Shamsudin
AU  - Iylia Idris
PY  - 2024
DA  - 2024/12/01
TI  - Hydrogen Recovery Analyses Based on LOGSIG Activation Functions in Feedforward Neural Networks
BT  - Proceedings of the International Conference on Innovation & Entrepreneurship in Computing, Engineering & Science Education (InvENT 2024)
PB  - Atlantis Press
SP  - 388
EP  - 409
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-589-8_36
DO  - 10.2991/978-94-6463-589-8_36
ID  - Yazan2024
ER  -