| Abstract Title: | Artificial neural network-based modeling of H2 recovery through liquid membranes |
| Session Choice: | Gaseous CEMS: technology suitability, operation and maintenance including NOx, SO2, CO and CO2 |
| Presenter Name: | Ms Shruti Samantarai |
| Co-authors: | Mr Riddhim Sehgal Ms Aadhya Roy |
| Company/Organisation: | Delhi Technological University (formerly Delhi College of Engineering) |
| Country: | India |
Abstract Information :
The world's primary energy source is currently derived from fossil fuels, such as oil, coal, and natural gas. Unfortunately, the reliance on these resources has led to environmental issues, notably climate change, due to the substantial emission of greenhouse gases, particularly carbon dioxide. The lack of an efficient technology for capturing and separating these gases has been a significant challenge. To address this, research has focused on developing new membrane materials, both inorganic and polymer-based, to purify hydrogen from carbon dioxide. Mixed matrix membranes (MMMs) offer unique characteristics that enhance their suitability for various separation processes. They serve as a versatile platform for improving the efficiency of gas and liquid separations in diverse industries. This study utilizes an artificial neural network (ANN) approach to model the liquid membrane process for extracting hydrogen from a mixture of H2/CO2. Experimental data from a previous study was employed to train the ANN. The optimization process involved determining the ideal number of neurons in the hidden layer to minimize the mean squared error between experimental results and model predictions. The trained ANN was then validated using experimental data, and the model was employed to analyze the impact of different operating parameters on the amount of hydrogen gas adsorbed as an effect of the membrane process.
