Please use this identifier to cite or link to this item:
https://ptsldigital.ukm.my/jspui/handle/123456789/487034Full metadata record
| DC Field | Value | Language |
|---|---|---|
| dc.contributor.advisor | Othman Jaafar, Prof. Dr. | - |
| dc.contributor.author | Mohammed Falah Allawi (P88695) | - |
| dc.date.accessioned | 2023-10-11T02:27:51Z | - |
| dc.date.available | 2023-10-11T02:27:51Z | - |
| dc.date.issued | 2019-03-05 | - |
| dc.identifier.other | ukmvital:121004 | - |
| dc.identifier.uri | https://ptsldigital.ukm.my/jspui/handle/123456789/487034 | - |
| dc.description | Reservoirs help to regulate and increase the reliability of water supplies in promoting livelihoods, raising agricultural productivity, and reducing farmers vulnerability to droughts. Successful operation of dam and reservoir systems to ensure optimal use of water resources is not attainable without accurate and reliable forecasting models. Due to the highly stochastic nature of hydrologic parameters, developing an accurate forecasting model that efficiently mimics such a complex pattern is an increasing domain of research. During the last two decades, Artificial Intelligence (AI) methods have been significantly utilized for developing an accurate model to handle different hydrological parameters. Existing forecasting models encounter several drawbacks, due to the difficulty of the underlying mathematical procedures to cope with these parameters. In this study, three AI models have been proposed, namely Radial Basis Function (RBF-NN), Adaptive Neuro-fuzzy Inference System (ANFIS) and CoActive Neuro-Fuzzy Inference System (CANFIS). These models have been utilized to forecast two hydrological parameters which are inflow and evaporation, being the major components of the reservoir simulation. In addition, appropriate adjustment to the CANFIS method is proposed to improve the mathematical procedure, thus enabling better detection of the high nonlinearity and stochastic patterns found in the reservoir inflow and evaporation training data. To demonstrate the models efficacy, the proposed models have been applied to forecast monthly inflow and evaporation for two different case studies with two different environmental conditions, namely; Aswan High Dam (AHD), located in southern Egypt and Timah Tasoh Dam (TTD), located in north Malaysia representing semi-arid and tropical environmental conditions, respectively. Comparative analyses of the forecasting ability of the models used are carried out with several statistical metrics to assess the reliability of the models. The statistical metrics support the better performance of the modified CANFIS model, in forecasting reservoir inflow, which significantly outperforms the other models to attain a low (RMSE (0.71 BCM month-1 at AHD, 2.4 MCM month-1 at TTD), MAE (0.42 BCM month-1, 1.24 MCM month-1), RE (28%, -62%)). In addition, the results revealed that the modified CANFIS model is better than the other proposed models in forecasting evaporation, RMSE (14.73 mm month-1 at AHD, 5.18 mm month-1 at TTD), MAE (10.9 mm month-1, 2.18 mm month-1), RE (25%, 17%). The present study ascertains the betterment of the CANFIS model with the proposed modifications with respect to other models. Furthermore, a proposal for a new simulation procedure to operate a reservoir system under realistic conditions has been carried out by integrating the results from the forecasting model with the generated operation rules from the previous studies. A critical assessment for the situation of a reservoir system under two different operating procedures has been reported using several statistical indicators. It is observed that all statistical indicator values have changed while operating the reservoir using forecasted inflow and evaporation data rather than using the deterministic value. The study concluded that the CANFIS model is better than ANFIS and RBF-NN to forecast reservoir inflow and evaporation. The simulation procedure based on forecasted data should also be adopted while evaluating the operation rules.,Ph.D. | - |
| dc.language.iso | eng | - |
| dc.publisher | UKM, Bangi | - |
| dc.relation | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina | - |
| dc.rights | UKM | - |
| dc.subject | UniversitiKebangsaanMalaysia -- Dissertations | - |
| dc.subject | Dissertations, Academic -- Malaysia | - |
| dc.subject | Reservoirs | - |
| dc.subject | Hydraulic structures | - |
| dc.title | Reservoir inflow and evaporation forecasting utilizing canfis model for improving operation evaluation | - |
| dc.type | Theses | - |
| dc.format.pages | 180 | - |
| dc.identifier.callno | TD395.L339 2019 3 tesis | - |
| dc.identifier.barcode | 005436(2021)(PL2) | - |
| Appears in Collections: | Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina | |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| ukmvital_121004+SOURCE1+SOURCE1.1.PDF Restricted Access | 4.73 MB | Adobe PDF |  View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.