International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 81 - Number 17 |
Year of Publication: 2013 |
Authors: Shekhar Pandharipande, Rachana S. Ranshoor |
10.5120/14216-2417 |
Shekhar Pandharipande, Rachana S. Ranshoor . Combined Artificial Neural Network Model for Estimation of Pressure Drop for Flow of CMC and Soil in Aqueous Solution. International Journal of Computer Applications. 81, 17 ( November 2013), 20-26. DOI=10.5120/14216-2417
Estimation of pressure drop for flow of Non -Newtonian fluid is a common situation & conventional models fail to address it with high accuracy & are to be system specific. Present work is aimed to explore the possible use of the Artificial Neural Network in developing combined models for the estimation of pressure drop as a function of flowrate, density, & concentration of CMC & soil in water mixture in a pipeline. Experimental runs are conducted & the 81 data points generated are divided into 64 & 17 as training & test data points respectively. The RMSE values for S1 & C1 models are 0. 023 & 0. 016 respectively. Further evaluation done by calculating & comparing the percentage relative error shows that, most of the predicted values have accuracy level of around 90% & is acceptable. The present work has successfully highlighted the potential of Artificial Neural Network in modeling complex processes.