An integrated approach to real-time optimal operation of an irrigation water system : a case study for Nam Oon Irrigation Project, Thailand | |
| Author | Watchara Suiadee |
| Call Number | AIT Diss. no.WM-05-02 |
| Subject(s) | Nam Oon Irrigation Project Irrigation water--Thailand |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering, School of Engineering and Technology |
| Publisher | Asian Institute of Technology |
| Series Statement | Dissertation ; no. WM-05-02 |
| Abstract | Increasing water demand on the existing supply systems causes numerous problems in operation of irrigation system. Water conservation can be a better alternative considering the difficulty in developing new supply systems. Even if, in the existing supply system, the shortcomings are often merely failures in the method of operation rather than the inadequacy caused by the physical limitations of the system. Therefore. optimal operation of reservoir system and efficiency operation and management of irrigation systems are important for effective management of available water resources. The use of mathematical models with the ever-growing computer technology in management process has been done by both simulation and optimization models. Therefore, the study concentrates on real-time optimal water operation of a selected irrigation project, Nam Oon Irrigation Project, Thailand. The Nam Oon Irrigation Project is a large scale irrigation projects -in the Northeast Thailand, which was constructed in 1967. The project stores and distributes water to its cultivated area to reduce flood damages and to supply sufficient irrigation water to the crops with the aim to increase agricultural production. The present operational rule curves for the Nam Oon reservoir have been developed by the Royal Irrigation Department in 2001 based on a simulation study by the HEC-3 model. The operation rule for Nam Oon is a combination of both management and technical issues. Though the reservoir operation rules are developed with technical considerations, at times, it is necessary to deviate from these rules while operating the reservoir. In practice, during flood periods, when the reservoir storage level exceeds a specific amount, an overflow discharge occurs. This causes flooding and damages in the downstream cultivated area along the river. To improve the reservoir operation performance and water use efficiency, it is necessary to develop rainfall, inflow forecasting methods and to develop the optimal rule curves considering maximum net benefit. These forecasted data and optimal rule curves arc used as guide for real-time reservoir operation based on downstream area water requirement. By comparing various models according to their various characteristics, it is found that WASAM model (Water Allocation Scheduling And Monitoring Model) developed by Irrigation Development Institute, Royal Irrigation Department of Thailand is the most suitable model for applying to the project to improve the system operation effectively. The original WASAM has been used successfully in the Greater Mae Klong Irrigation Project since 1987. The program also has been applied for water allocation scheduling and monitoring in many irrigation projects in Thailand. However, the original WASAM model does not have reservoir operation model operated in parallel over time with interactions through their solution procedure. Therefore, in this study, a reservoir operation module using Genetic Algorithm (GA) approach is developed and combined to the original WASAM to increase model capability. The new developed model namely WASAM-RGA (Water Allocation Scheduling And Monitoring - Reservoir Operation Using Genetic Algorithm) can be used for short and long terms planning based on the developed optimal rule curves and adaptive forecasted rainfall and inflow data. In real-time, the developed WASAM-RGA model solves from the current week to the last week of a year to obtain the optimal operating policy. The observed week end storage volume is used as the input to adjust the optimal forecasted water releases in realtime for operating horizon of 1 year. To continue for the subsequent periods the state of the system is updated with the storage, forecasted rainfall and inflow data obtained to that period. A new optimal operating policy is derived for the subsequent periods. This procedure is continued until the end of last period in the year. When operating in realtime, only the current period's decision is implemented. For implementation, the application of WASAM-RGA is done on weekly basis and tested for a 4-year period. 1998 to 2001. For upcoming week operation (1 week ahead) the adaptive forecasted rainfall and forecasted inflow data using Time Delay Neural Network (TDNN) is used. For long term planning, the forecasted inflow using Multiplicative ARIMA model and long term average forecasted rainfall are used as the input into the model. To determine the optimal releases, genetic algorithm (GA) is used to find the optimal releases for the periods starting from current period to the last period in the year. The objective function of the optimization model for reservoir operation is to minimize the difference of the optimal rule curve and calculated storage volumes. The results of WASAM-RGA model to Nam Oon Irrigation project show that the developed model provides robust and acceptable solutions in determining the optimal releases in short and long terms planning based on changing downstream water requirement, rainfall and inflow data. The water use efficiency are higher than the existing condition. The results of optimal reservoir operation are compared to the results obtained by using Discrete Differential Dynamic Programming approach, Standard Operating Policy and others. A number of practical applications are presented that demonstrate the effectiveness of the GA approach of WASAM-RGA model. The numerical results show that WASAM-RGA provides realistic optimal releases in term of the water quantity, flood duration, etc. The WASAM-RGA model developed in this study is user's friendly with window interface for data input and output presentation. The model can be simply modified to incorporate other conditions for reservoir operation and water allocation in irrigation projects |
| Year | 2006 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertations ; no. WM-05-02 |
| Type | Dissertation |
| School | School of Engineering and Technology |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Water Engineering and Management (WM) |
| Chairperson(s) | Tawatchai Tingsanchali; |
| Examination Committee(s) | Clemente, Roberto S.;Athapol Noomhorm;Van-Thanh-Van Nguyen; |
| Scholarship Donor(s) | Royal Thai Government (RT-HRD Phase 2); |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2005 |