| Author | Huang, Wen-cheng |
| Call Number | AIT Diss. no. WA-89-01 |
| Subject(s) | Reservoirs--Data processing
|
| 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. WA-89-01 |
| Abstract | An Operational Mode Switch (OMS) system is developed for the
on-line operation of a multipurpose reservoir situated in a
typhoon-prone area. The decision whether or not and when to shift
the operation back and forth between the long-term "normal mode"
and short-term "emergency mode " is determined by the OMS model.
The decision criterion to be maximized is the expected
multiattribute utility trading off flood alleviation, water
supply and hydropower production. This study demonstrates the
practical applicability of Bayes decision theory to this type of
problem when the posterior probabilities are calculated on the
basis of routinely available meteorological information. As
compared with the existing operating rules, the OMS model appears
to provide a more flexible and efficient way to operate reservoir
systems.
The OMS system relies basically on the interlinkage of a
simulation and a decision model. The use of this model is
activated upon a typhoon announcement by the Central Weather
Bureau. At this stage the reservoir operation is in normal mode
following a Stochastic Dynamic Programming- based (SDP) release
policy adopted for long-term operation, which accounts for
tradeoffs between water supply, power generation and low flow
augmentation.
In this study, four types of SDP are presented for the online
reservoir operation, in which observed or forecasted inflows
are employed. Among the proposed four types of SDP, the one with
observed inflows performed better than the others with forecasted
inflows. Nevertheless, under different hydrological regimes the
"truth" might not be universal but depend upon the
characteristics of the particular water resources system.
Concerning the forecasted inflows incorporated in SDP, Box Jenkins models are proposed to tackle the streamflow forecasts.
The results showed appropriateness of Box- Jenkins models to fit
the stationary streamflow series with the exception of one
affected by floods. A rainfall-runoff model with multiple linear
regression type is proved appropriate and used during the floods. |
| Year | 1989 |
| Corresponding Series Added Entry | Asian Institute of Technology. Dissertation ; no. WA-89-01 |
| Type | Dissertation |
| School | School of Engineering and Technology |
| Department | Department of Civil and Infrastucture Engineering (DCIE) |
| Academic Program/FoS | Water Resources Research Engineering (WA) |
| Chairperson(s) | Bogardi, Janos J.;Harboe, Ricardo; |
| Examination Committee(s) | Gupta, Ashim Das;Huynh, Ngoc Phien;Paudyal, Guna N.;Plate, Erich J.; |
| Scholarship Donor(s) | Republic of China; |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 1989 |