Modeling of ANN to determine optimum adsorption capacity for removal of pollutants in wastewater
The rise in wastewater as a source of pollution rank equal to climate change as the most urgent environmental issues currently. The rapidly growing industrialization and urbanization, improper sanitary disposal and household wastewater has contributed to this pollution. Such concerns pose potential...
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
10.1109/ICSIMA.2017.8311989
2018
|
| Subjects: | |
| Online Access: | http://irep.iium.edu.my/61395/ http://irep.iium.edu.my/61395/ http://irep.iium.edu.my/61395/ http://irep.iium.edu.my/61395/1/61395_Modeling%20of%20ANN%20to%20determine%20optimum%20adsorption_complete.pdf http://irep.iium.edu.my/61395/2/61395_Modeling%20of%20ANN%20to%20determine%20optimum%20adsorption_scopus.pdf |
| Summary: | The rise in wastewater as a source of pollution rank equal to climate change as the most urgent environmental issues currently. The rapidly growing industrialization and urbanization, improper sanitary disposal and household wastewater has contributed to this pollution. Such concerns pose potential risks and hazards towards the public health and animal ecosystems. Meanwhile, the wastewater treatment process involves both physical and chemical process chain which is susceptible to error due to the human factor, variation in the quality of raw water as well as chemical/physical characteristics of such raw materials used. An intelligent method for predicting the optimal adsorption capacity for removal of pollutants in wastewater based on ANN is proposed to reduce the percentage error and obtain optimal treatment efficiency. The primary focus is to identify the operating parameters which affect adsorption capacity. Using the parameters as input factors, the proposed system is trained and tested to obtain an optimal adsorption capacity to remove pollutants. Evaluation and validation of the proposed method on real data depend on the mean absolute error (MAE), mean square error (MSE), root mean square error(RMSE), normalized mean square error(NMSE) and correlation of efficiency. The correlation between the experimental and ANN result is 0.99881 of 1.00000 which indicates that ANN is a perfect match. |
|---|