NARX-based water quality index model of Air Busuk River using chemical parameter measurements

• Published in: Indonesian Journal of Electrical Engineering and Computer Science
• Main Author: Hasnan M.I.; Jaffar A.; Thamrin N.M.; Misnan M.F.; Yassin A.I.M.; Ali M.S.A.M.
• Format: Article
• Language: English
• Published: Institute of Advanced Engineering and Science 2021
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85114291129&doi=10.11591%2fijeecs.v23.i3.pp1663-1673&partnerID=40&md5=79a6c7e0ef9b6bbf559b75cd9a27c276

Water quality plays a major role in issues related to public health and marine life. Hence, monitoring river for contaminations is vital for ensuring safe and sustainable water resources. Conventional method for assessing water quality index is costly as it requires considerable amount of time and laboratory resources. Therefore, this study proposes a water quality index model based on artificial neural network. A six-year data for Air Busuk River is obtained from the Department of Environment. Dissolved oxygen, biological oxygen demand, and ammoniacal nitrogen has shown high correlation with water quality index. The water quality index model is then developed based on these parameters, employing the non-linear autoregressive with exogeneous input structure. Generally, the model which is based on three chemical parameters has shown satisfactory performance with overall regression of 0.8767 and passed the correlation function tests. The model offers a potentially efficient method for assessing water quality with cost-saving benefits for government agencies and monitoring authorities. © 2021 Institute of Advanced Engineering and Science. All rights reserved.