| Autor: |
Rose Ellen N. Macabiog, Jennifer C. Dela Cruz, Timothy M. Amado |
| Rok vydání: |
2018 |
| Předmět: |
|
| Zdroj: |
2018 IEEE 10th International Conference on Humanoid, Nanotechnology, Information Technology,Communication and Control, Environment and Management (HNICEM). |
| DOI: |
10.1109/hnicem.2018.8666363 |
| Popis: |
Hydroelectric power is a significant source of renewable energy generated by run-of-river hydropower plants. However, operation and maintenance of these plants pose a threat to the water quality of rivers. Diversion scheme adopted in these plants can substantially modify river ecosystems instream resulting to changes in the water quality parameters. These changes degrade the river ecosystem, thereby, compromising the health and growth of aquatic species. This study aimed to analyze water quality parameters used to evaluate the compliance to water quality standards. Based on DENR allowable values, change in temperature should not exceed 30 C and dissolved oxygen should not be lower than 5 mg/l. Regression analysis was used to establish relationships in analyzing water quality parameters. With the use of various regression machine learning models, the water quality dataset was modelled to predict the change in temperature and dissolved oxygen downstream using water level downstream and water temperature downstream as predictors. Based on the Stochastic Gradient Boosting Model, while the water level downstream decreases and the water temperature downstream increases, the change in temperature increases. Based on the Linear Model or the Ridge Model, while the water level downstream decreases and the water temperature downstream increases, the dissolved oxygen downstream relatively decreases. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
