Artificial Intelligence Techniques for Predicting Water Quality Parameters and Management in a Complex River System: A Review

• Main Authors: Aljumaily H., Hayder G., Yussof S., Ali R.H.
• Other Authors: 57202391436
• Format: Conference Paper
• Published: Springer Nature 2024
• Subjects: ANN approaches; Parameters; Ringlet river and Ringlet reservoir; Sedimentation; Water quality
• Institution: Universiti Tenaga Nasional

Evaluation of the water quality in the rivers is necessary to enhance the health of humans and ecosystems, but it quite difficult and requires more time and effort due to many parameters affected. Although, the Ringlet River considered the more important rivers in the Cameron Highland, probably it is the most badly affected river with records of incidences where permissible levels of certain parameters such as suspended solids, agrochemicals, and sewage have been contravened and drinking water. Recently, the tourism, agriculture, and industry in that region dramatically increased. This parameter leads to increase the pollution rate in the water of the four rivers that flow into the lake, which changes the quality of the water to the worse level. The nature of the muddy area is a parameter that allows soil erosion on the banks of the rivers especially at the time of floods. Due to these parameters, the dam loses the goal of hydroelectric construction benefit, in addition to protect catchment areas from floods, recently this dam become less useful due to the accumulated amount of mud increasing steadily, therefore, the annual removing mud sediments from the Ringlet reservoir is difficult and expensive. The aim of the study is to find logical, feasible, and economic solutions to mitigate the damage and treat it in the future. The study is concerned to estimate, evaluate, predict, and solve the problems of the Ringlet River and Ringlet reservoir which are located in the north of Malaysia. Numerous intelligent models have been developed to accurately predict and evaluate the quality of the surface water by using conventional methods that may be ineffective, inaccurate, and relatively slow in predicting results. Therefore, the use of artificial intelligence (AI) technology was resorted to predict water quality and sediment volume in the Ringlet Basin to assess its applicability of water quality. ANN method is selected as a statistical method applied to develop an interpretability model for decision-maker. A methodology has been scheduled to include the Ringlet river and Ringlet catchment reservoir. There are many parameters affected the pollution of the Ringlet reservoir such as urban activities, seasonal effects (flood and climate), animal�s wastes, agricultural residues, and `soil properties that may cause the erosion of the specified area. � 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.