Determination and prediction of blue water footprint at Sungai Lembing, Bukit Sagu and Bukit Ubi water treatment plant
The majority of the earth is covered by water, but only a small percentage of that amount is available for use as clean water. Currently, one-third of the world populations are facing the water shortages. Therefore, accounting blue water footprint (WFb) will help in assessed overall water consumptio...
| Main Author: | |
|---|---|
| Format: | Undergraduates Project Papers |
| Language: | English |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/29633/ http://umpir.ump.edu.my/id/eprint/29633/1/Determination%20and%20prediction%20of%20blue%20water%20footprint%20at%20Sungai%20Lembing%2C%20Bukit%20Sagu%20and%20Bukit%20Ubi%20water%20treatment%20plant.pdf |
| Summary: | The majority of the earth is covered by water, but only a small percentage of that amount is available for use as clean water. Currently, one-third of the world populations are facing the water shortages. Therefore, accounting blue water footprint (WFb) will help in assessed overall water consumption for three different water treatment plant in Kuantan river basin. This paper illustrates the prediction of blue water footprint of Sungai Lembing, Bukit Sagu and Bukit Ubi WTPs throughout year 2015 to 2017. The parameters considered in the study were water intake, rainfall intensity and evaporation. In this study, water footprint manual was used to account blue water footprint throughout all water treatment plants. In order to make a prediction, Bayesian Networks (BN) and Artificial Neural Network (ANN) were used as an algorithm to train the result. Thus, prediction trend for three different water treatments has been able to be produced by using WEKA software. As a result, total blue water footprints for Sungai Lembing WTP, Bukit Sagu WTP and Bukit Ubi WTP for 2015 to 2017 were 4,905,076 mᶾ, 5,924,203 mᶾ and 26,400,519 mᶾ respectively. Results proved that ANN is the best algorithm for all WTPs as it produced lower value of root mean square error (RMSE) compared to Bayesian Network. |
|---|