Formulation Of Fitness Function To Predict Ph Value Of Adjacent Block Via Ph Value, Water Flow Speed And Direction
This project is based on determining the water quality as poor water quality can also pose a health risk for ecosystems. Water quality is measured by several factors, such as potential of hydrogen (pH), the concentration of dissolved oxygen (DO), bacteria levels, the amount of salt (or salinity), or...
Saved in:
| Main Author: | |
|---|---|
| Format: | Undergraduates Project Papers |
| Language: | English |
| Published: |
2022
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/39902/1/EA18091_NURUL%20NAJIHAH%20MOHD%20RADZI_Thesis%20-%20Nurul%20Najihah.pdf http://umpir.ump.edu.my/id/eprint/39902/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This project is based on determining the water quality as poor water quality can also pose a health risk for ecosystems. Water quality is measured by several factors, such as potential of hydrogen (pH), the concentration of dissolved oxygen (DO), bacteria levels, the amount of salt (or salinity), or the amount of material suspended in the water (turbidity). In this project, we more focus on the pH of water as it given more impact on determine the quality of water. In this research, we focus on measuring water quality parameters of University Malaysia of Pahang main lake (3.5431680, 103.4355621). Water quality criteria are being measured and analyses to determine the state of the lake. As we monitor the water quality measurement especially the pH value, we notice that the speed of water flow also manipulates the value of pH water. Collect large set of data which comprises of five location, four of the locations pH are used to determine the fifth location pH. To predict the lake water quality, we are using fitness function that has been formulate using Multi-Layer Neural Network by Genetic Algorithm (MLNN-GA) and compare the results in terms of accuracy of prediction. The collected data is split into two which are training and testing data. The training data will keep being inserted until the best weightage for each data is obtained. The best weightage will be used to test the testing data and acquired the prediction value. The results of prediction will be compared with the actual pH value. The results of prediction obtained by predicted data pH with speed with 94.27% compared with predicted data without speed with only 93.83%. The result shown that speed is one of the factor that contribute to pH prediction. |
|---|