Prediction of Adsorption Isotherm of Methane Gas onto Activated Coconut Shell Charcoal by Neuro-Fuzzy Logic Model

ANG (Adsorption Natural Gas) is a technology in which natural gas is adsorbed
bya porous adsorbent material (i.e. activated carbon) at relatively low pressure (3.5 MPa
up to 5.0 Mpa) that is applied onANG vehicle. This technology, which the storage vessel
is filled with a suitable adsorbent material, will have greater energy density compared to
the same storage vessel without the adsorbent when filled to the same pressure. Besides
that, natural gas is found as a potentially attractive fuel for vehicle use. It is because it is
normally cheaper than diesel and gasoline and the vehicle has a less adverse effect
compared to liquid-fuel vehicles which emitting more C02 as well as several other air
pollutants. However, natural gas has a problem to be utilized at ambient temperature
since it is difficult to liquefy because it comprises mostly of methane gas, which having a
very low critical temperature (191 K). Therefore, this project is mainly to predict the
adsorption isotherm of methane gas onto Activated Coconut Shell Charcoal using Neuro-
Fuzzy Logic design system.
From the modeling done, the prediction of adsorption isotherm at three different
temperatures (273 K, 303 K and 323 K) for methane gas onto activated coconut shell
charcoal was successfully implemented by using Neuro-Fuzzy module system. The
software used was Fuzzytech 5.52 Professional Demo software. The reference data used
was taken from research done by E.Loren Fuller,Jr entitled Characterization of Porous
Carbon Fibre and Related Materials. From the simulation done, the objective of this
project has been achieved due to average error between experimental/reference data and
Neuro-Fuzzy Logic data gained is less than 3%. The isotherm plotted between Neuro-
Fuzzy Logic data and experimental data shown almost similar to each other. It can be
concluded that the equilibrium model of adsorption isotherm of methane gas onto
activated coconut shell charcoal was fitted with the experimental data at different
temperature, which in this case were 273 K, 303 K and 323 K. Therefore, from the
model developed, it is easier to predict the adsorption isotherm at other temperature but
limited within the temperature range (273 K- 323 K) andpressure up to 760Torr.