| id |
my.uniten.dspace-20678 |
|---|---|
| record_format |
dspace |
| spelling |
my.uniten.dspace-206782023-05-05T12:25:23Z A Novel Hybrid Time Series Prediction Model for Dissolved Oxygen In Reservoir Balahaha Fadi Ziyad Sami Water Oxygen Interim Semester 2020/2021 In the previous decades, eutrophication has been one of the most well-known issues prompting the blunder of water characteristics as much as its reality is of extraordinary advantage to both human and oceanic lives particularly Fei-Tsui repository. This task is to research the creation of time series in Fei Tsui reservoir for prediction of dissolved oxygen (DO). The project is designed to produce a predictive model which can be used for monitoring the water quality. The test was performed mainly at reservoir Fei Tsui, Taipei city, Taiwan. The objective of this project is to monitor the concentration of DO in the reservoir and to develop a model based on the previous records of events and without the need to sit visit to test the concentration of DO. It has been an important concern in the Fei Tsui reservoir for decades. Due to eutrophication problems in the reservoir, this matter has affected the water sources in Taipei City, as Fei Tsui Reservoir was considered to be the primary water source for Taipei City. Twenty-seven elements of ten years average monthly records and thirteen years average annual records have been selected and correlated to see which one has a strong relationship with DO. The parameters which have an active relationship with DO are water treatment, Biochemical oxygen demand, chloride, iron, and total organic carbon. A model was developed based on linear regression to determine the DO concertation in the water body. MATLAB software is used to construct the model of prediction because it considers one of the software for machine learning. The result revealed that the model of an average annual data records of 13 years shows better accuracy than the average monthly records of 10 years. Eutrophication can affect living aquatic life and drinking water. As a result, the model produced by the average annual reading of dissolved oxygen for 14 years has a more accurate result than the monthly reading due to the lack of data in some months. The performance data for this analysis is keyed by average annual readings and monthly readings. WI was given an optimal model validation result. The optimal model was found with a regression value of 0.999. This research can be extended to other reservoirs to analyze the hydrological situation. 2023-05-03T15:12:48Z 2023-05-03T15:12:48Z 2020-09 Resource Types::text::Thesis https://irepository.uniten.edu.my/handle/123456789/20678 application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| institution |
Universiti Tenaga Nasional |
| building |
UNITEN Library |
| collection |
Institutional Repository |
| continent |
Asia |
| country |
Malaysia |
| content_provider |
Universiti Tenaga Nasional |
| content_source |
UNITEN Institutional Repository |
| url_provider |
http://dspace.uniten.edu.my/ |
| topic |
Water Oxygen |
| spellingShingle |
Water Oxygen Balahaha Fadi Ziyad Sami A Novel Hybrid Time Series Prediction Model for Dissolved Oxygen In Reservoir |
| description |
Interim Semester 2020/2021 |
| format |
Resource Types::text::Thesis |
| author |
Balahaha Fadi Ziyad Sami |
| author_facet |
Balahaha Fadi Ziyad Sami |
| author_sort |
Balahaha Fadi Ziyad Sami |
| title |
A Novel Hybrid Time Series Prediction Model for Dissolved Oxygen In Reservoir |
| title_short |
A Novel Hybrid Time Series Prediction Model for Dissolved Oxygen In Reservoir |
| title_full |
A Novel Hybrid Time Series Prediction Model for Dissolved Oxygen In Reservoir |
| title_fullStr |
A Novel Hybrid Time Series Prediction Model for Dissolved Oxygen In Reservoir |
| title_full_unstemmed |
A Novel Hybrid Time Series Prediction Model for Dissolved Oxygen In Reservoir |
| title_sort |
novel hybrid time series prediction model for dissolved oxygen in reservoir |
| publishDate |
2023 |
| _version_ |
1806428163719299072 |
| score |
13.214268 |