Optimizing dam and reservoir operations using machine learning
2024
Saved in:
| Main Author: | |
|---|---|
| Format: | text::Thesis |
| Language: | English |
| Published: |
2025
|
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| id |
my.uniten.dspace-37394 |
|---|---|
| record_format |
dspace |
| spelling |
my.uniten.dspace-373942025-03-16T02:01:56Z Optimizing dam and reservoir operations using machine learning Mohammad Abdullah Adib Almubaidin 2024 Optimizing reservoir operation is a complex problem with non-linearities, numerous decision variables, and challenging constraints to simulate and solve. Optimization methods vary depending on objectives, reservoir type, and algorithms used. The challenge associated with a specific reservoir type lies in its full reliance on rainfall for water inflows. The monthly fluctuations in water inflow are challenging to forecast, posing a hurdle for decision-makers in managing the reservoir. Additionally, the sediment problem further reduces the reservoir's efficiency. The thesis aims to study optimizing the operation of the Mujib reservoir in Jordan using Metaheuristics Algorithms (MHAs) to reduce the water deficit through two objectives using the historical data from the years 2004 to 2019. The first objective, finding the optimal rule curves for the reservoir using different algorithms (artificial bee colony (ABC), particle swarm optimization (PSO), genetic algorithm (GA), firefly algorithm (FA), invasive weed optimization (IWO), teaching learning-based optimization (TLBO), and harmony search (HS) to reduce water deficit and flooding cases. And evaluate the impact of water shortages and excess releases when reducing water demand by 10%, 20%, and 30% in the reservoir. The findings showed that the algorithms effectively reduced cases of water shortage and excess release compared to the current operation. The best solutions using the TLBO algorithm reduced the frequency and average of the water shortage to 55.09 % and 56.26 %, respectively, and reduced the frequency and the average of the excess release to 63.16 % and 73.31 %, respectively. The second objective is finding the optimal operation of the reservoir through different scenarios and studying the impact of these scenarios on the operation of the reservoir by using the CSS algorithm to reduce the water deficit of the reservoir. These scenarios are sediment effect, water demand management, increasing the volume of the reservoir, and finding the optimal operation for the reservoir. And compare the results with the current operation of the reservoir. Risk analysis (volumetric reliability, shortage index (SI), resilience, vulnerability) and error indexes (correlation coefficient 2, the root mean square error (RMSE), and the mean absolute error (MAE)) were used to compare results between scenarios, in addition to the annual water deficit values from the CSS algorithm for each scenario. The simulation of monthly sediment values in the Mujib reservoir showed that sediment accumulation accounted for 14.6% of the reservoir's volume at the end of 2019. Removing sediments retained by the dam can reduce the water deficit by 19.42% when using the CSS algorithm. Additionally, reducing agricultural water demand by 11% and removing sediment reduced the water deficit by 42.40%. The study also examined the impact of increasing the storage capacity of the reservoir by 10%, 20%, and 30%, revealing a decrease in water deficit by 35.44% when sediment removal was included in the analysis. The study examined the scenario of increasing the storage capacity of the Mujib reservoir by 30%, reducing water demand by 11%, and removing sediment. This scenario resulted in a 53.59% decrease in water deficit, providing decision makers with viable solutions to address the water deficit problem in the reservoir. The thesis recommends the necessity of Investigating sediment levels, optimizing water conservation, and adopting hybrid techniques are crucial for sustainable reservoir management, impacting economic and social aspects 2025-03-10T05:01:53Z 2025-03-10T05:01:53Z 2025-03-05 Resource Types::text::Thesis https://irepository.uniten.edu.my/handle/123456789/37394 en application/pdf |
| 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/ |
| language |
English |
| description |
2024 |
| format |
Resource Types::text::Thesis |
| author |
Mohammad Abdullah Adib Almubaidin |
| spellingShingle |
Mohammad Abdullah Adib Almubaidin Optimizing dam and reservoir operations using machine learning |
| author_facet |
Mohammad Abdullah Adib Almubaidin |
| author_sort |
Mohammad Abdullah Adib Almubaidin |
| title |
Optimizing dam and reservoir operations using machine learning |
| title_short |
Optimizing dam and reservoir operations using machine learning |
| title_full |
Optimizing dam and reservoir operations using machine learning |
| title_fullStr |
Optimizing dam and reservoir operations using machine learning |
| title_full_unstemmed |
Optimizing dam and reservoir operations using machine learning |
| title_sort |
optimizing dam and reservoir operations using machine learning |
| publishDate |
2025 |
| _version_ |
1827448133009801216 |
| score |
13.251813 |