7 / 44
Volume 4
Nano Research & Applications
ISSN: 2471-9838
JOINT EVENT
October 04-05, 2018 Moscow, Russia
&
2
nd
Edition of International Conference on
26
th
International Conference on
Advanced Nanotechnology
Materials Technology and Manufacturing Innovations
Advanced Nanotechnology 2018
& Materials-Manufacturing 2018
October 04-05, 2018
Page 27
Hemant Parmar
Ujjain Engineering College, India
Hemant Parmar, Nano Res Appl 2018, Volume 4
DOI: 10.21767/2471-9838-C5-020
Application of artificial neural networks for the prediction of performance of desiccant material
for desiccant cooling system
D
esiccant cooling system (DCS) is an alternate suitable option against conventional cooling system in humid climates.
A typical system combines a dehumidifier that uses dry desiccant wheel with desiccant material, with direct or
indirect evaporative systems and a sensible cooling system. DCS is the environmental protection technique for cooling
purpose of the building. Desiccant wheel is main key component of the DCS and this wheel has desiccant material. The
desiccant materials may be a suitable option for better use of evaporative cooling techniques in warm and humid climate.
The dehumidification of air by adsorption is a physical process by desiccant materials that attracts the molecules of water
present in air on the adsorbent surface. Desiccant materials attract moisture from the air by creating an area of low
vapour pressure at the surface of desiccant. Some common adsorbents used are SiO
2
(Silica gel), LiCl, Al
2
O
3
(Activated
alumina), LiBr and Zeolite. This system reduces the level of chlorofluorocarbons (CFC) in the environment because it
restricts the use of conventional refrigerant. An Artificial Neural Network (ANN) was constructed to predict symptoms
of desiccant wheel for any climatic conditions. The symptoms of desiccant wheel were outlet specific humidity and
outlet dry bulb temperatures. Feed-forward network was employed with resilient back-propagation (
trainrp
) algorithm
used as the training function. The outputs from the network were obtained and validated with the experimental results.
Among the constructed networks, the best prediction performance was observed in two-hidden-layered network with
minimum error. The modeling of the desiccant wheel was carried out with neural network toolbox of MATLAB® with
two inputs (specific humidity and temperature at inlet) and two outputs (specific humidity and temperature at outlet)
values. The performance index (Mean square error) for specific humidity and temperature were calculated as 0.00209
and 0.00579 respectively. The proposed model may be used for any climatic conditions to predict the output from the
desiccant wheel for the design of the solid desiccant based cooling system in humid climate.
Recent Publications
1. Parmar H., Hindoliya D.A., 2013, “Performance of solid desiccant based evaporating cooling system under
the climatic zones of India”. International Journal of Low Carbon Technologies, vol.8, pp 52-57.
2. Parmar H., Hindoliya D.A., 2011, “Artificial Neural Network based modeling of Desiccant Wheel”. Energy
and Buildings, vol.43, pp 3505-3513.
3. Parmar H., Hindoliya D.A., 2011, “Solid Desiccant Cooling System Employed with Ventilation Cycle”.
Journal of Institution of Engineers (Springer), DOI 10.1007/s40032-012-0038-9.