Saisunee Budhakooncharoen creator Bogardi, Janos J. Chairpeson Gupta, Ashim Das Examination Committee Prida Thimakorn Examintion Committee Government of Australia Scholarship Donor text th Bangkok Asian Institute of Technology 1986 monographic eng

print

83 p. Dynamic Programming (DP) is widely regarded as one of the most effective method to derive optimal release policy f or reservoir operation. In the present study, two types of DP , Incremental Dynamic Programming (IDP) and Stochastic Dynamic Programming (SDP) have been applied for a single purpose (hydropower generation) reservoir. The results from bot h techniques were compared with the historical operational records of the Kariba Reservoir on the Zambezi River, Central Africa. A simulation model has been developed to assess system operation according to the operation rules based on the SOP. Beyond the comparison of the historical records with potential release strategies, the operation release policies derived by the SDP have been coupled with a simple flow forecasting based on linear regression analysis. This model can serve as nucleus to develop a real time operational procedure. of particular interest was to find out, whether the excessive drawdown of the reservoir in years 1981 - 1984 could have been avoided, if the inflows were known in advance (IDP) or by using the simple forecasting and SDP based operational policies. Finally, sensitivity analyses, by varying installed capacity of the power plants, size of the reservoir and minimum drawdown levels have been carried out to help identifying potential increase of installed power generation and/or reservoir capacity. A thesis study submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology Thesis (M.Eng.) - Asian Institute of Technology, 1986 Reservoirs Asian Institute of Technology. http://203.159.5.9/ait-thesis/detail.php?q=B19779 http://203.159.5.9/ait-thesis/detail.php?q=B19779 210598 20260818084320.0