News Updates Tuesday 19th Jun 2018 :
  • Welcome to INPRESSCO, world's leading publishers, We have served more than 10000+ authors
  • Articles are invited in engineering, science, technology, management, industrial engg, biotechnology etc.
  • Paper submission last date of May/June 2018 extended to 20 June 2018, Submit online or at editor.ijcet@inpressco.com
  • Our journals are indexed in NAAS, University of Regensburg Germany, Google Scholar, Cross Ref etc.
  • DOI is given to all articles

Prediction of Water Treatment Plant Outlet Turbidity using Artificial Neural Network


Author : Jabbar H. Al-Baidhani and Mohammed Adnan Alameedee

Pages : 1559-1565
Download PDF
Abstract

The present study was conducted for the purpose of simulation of some units that was used intensively in a conventional water treatment plant. A pilot plant used to test the simulated raw water by mixing kaolinite clay with tap water and treated with difference alum dosage. The simulation tool that was used recently based on artificial intelligence and experimental tests, which is the neural network (ANN). The ANN was used to simulate and predict the required alum dosage for treated the turbidity in raw water. The experimental work was studied in a pilot plant scale, where a rig was designed and manufactured to simulated these treatment units. The synthetic turbidity was used to create the turbidity water that was used in this work. The experimental study was consisted of four runs with different inlet flowrate; each run contained five different influent turbidity sets (25, 50, 75, 100, 150) NTU and ten series of alum dosage varies form (5 – 50) mg/l for each turbidity set. The collected data for the neural network was about 200 set of data. The process model had eight input nodes and two output nodes, which is the effluent turbidity and pH. The correlation coefficient of this model was 0.98 and 0.86 for turbidity and pH respectively. The error indices of this model where 2.03 NTU, 1.38 NTU, 13% for RMSE, MAE, MAPE respectively for turbidity, while the pH errors were 0.19, 0.14, 2% for RMSE, MAE, MAPE respectively. These errors indices showed a good performance for the model in the prediction process of these parameters within the ranges of the study.

Keywords: Artificial Intelligence; Artificial Neural Network (ANN); Process Model; Correlation Coefficient (R); Root Mean Square Error; Mean Absolute Error; and Mean Absolute Percentage Error

Article published in International Journal of Current Engineering and Technology, Vol.7, No.4 (Aug-2017)

Call for Papers
  1. IJCET- May/June 2018 Issue

    Submission Last Date extended to
    20 June
  2. DOI is given to all articles
  3. Current Issue
  4. IJTT-June-2018
  5. IJAIE-June-2018
  6. IJCSB-June-2018
  • Inpressco Google Scholar
  • Inpressco Science Central
  • Inpressco Global impact factor
  • Inpressco aap

International Press corporation is licensed under a Creative Commons Attribution-Non Commercial NoDerivs 3.0 Unported License
©2010-2018 INPRESSCO® All Rights Reserved