Skip navigation
Please use this identifier to cite or link to this item: http://repository.iitr.ac.in/handle/123456789/20004
Title: Modelling and Simulation of Reverse Osmosis System Using PSO-ANN Prediction Technique
Authors: Mahadeva R.
Manik, Gaurav
Verma O.P.
Goel A.
Kumar S.
Pant M.
Sharma T.K.
Verma O.P.
Singla R.
Sikander A.
Published in: Advances in Intelligent Systems and Computing
Proceedings of 3rd International Conference on Soft Computing: Theories and Applications, SoCTA 2020
Abstract: Nowadays, among various water treatment and desalination technologies such as reverse osmosis (RO), multi-effect distillation (MED), and multi-stage flash (MSF), RO is an appropriate and suitable technology in the world. It is extremely used technology (>60%) around the globe. It is quite popular in separation and filtering process, especially for drinking water services as well as industrial applications. Modelling and simulation of such plants are necessary for better analysis and understanding with minimum effort, energy, and time. It involves various machine learning techniques such as an artificial neural network (ANN), support vector machine (SVM). Among these techniques, ANN is one of the best and reliable techniques, which provides good results. ANN may be learned through numerous training algorithms such as back-propagation (BP), particle swarm optimization (PSO); PSO-ANN learning algorithm generated the optimal values of initial weights and biases and to train the network. In this article, experimental datasets of RO plants have been collected from the literature and the regression coefficient (R) along with minimum mean square error (MSE) are evaluated. Four input variables (temperature T (°C), pressure P (MPa), feed concentration C (Mg/L), and pH) and three output variables (water recovery (%), total dissolved solids (TDS) rejection (%), and specific energy consumption (SEC) (kWh/m )) are considered for analysis. The simulated results observed better regression coefficients (R) (0.98557, 0.96016, and 0.97118) with minimum MSE (0.5502%, 0.9389%, and 1.5755%), respectively, corresponding to output variables of the RO plant. 
Citation: Advances in Intelligent Systems and Computing, 2020, 1209- 1219
URI: https://doi.org/10.1007/978-981-15-0751-9_111
http://repository.iitr.ac.in/handle/123456789/20004
Issue Date: 2020
Publisher: Springer
Keywords: Artificial neural network (ANN)
Desalination
Modelling and simulation
Particle swarm optimization (PSO)
Reverse osmosis
Backpropagation
Computation theory
Desalination
Distillation
Energy utilization
Learning systems
Mean square error
Neural networks
Potable water
Reverse osmosis
Soft computing
Support vector machines
Water treatment
Desalination technologies
ISBN: 9.78981E+12
ISSN: 21945357
Author Scopus IDs: 57204158068
56595314900
56594677400
57210524796
57211991468
Author Affiliations: Mahadeva, R., Department of Polymer and Process Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhand 247667, India
Manik, G., Department of Polymer and Process Engineering, Indian Institute of Technology Roorkee, Roorkee, Uttarakhan
Corresponding Author: Manik, G.; Department of Polymer and Process Engineering, India; email: manikfpt@iitr.ac.in
Appears in Collections:Conference Publications [PE]

Files in This Item:
There are no files associated with this item.
Show full item record


Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.