Studying the Risk Based Design Criteria of Reservoir Capacity

• Main Authors: Chung-Nan Wang, 王宗南
• Other Authors: Frederick N-F Chou
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/35263050487728601599

碩士 === 國立成功大學 === 水利及海洋工程學系碩博士班 === 94 === The shortage index, which proposed by US Army Corps of Engineers, was adopted as a major criteria for designing the capacity of a reservoir in the last two decades in Taiwan. Shortage index was designed to demonstrate the severity of deficit by averaging the squares of annual shortage rates of all years. If the annual shortage rate is 10% in every year, the shortage index will equal to 1. However, the amount and duration of shortages are not uniform in different years due to stochastic hydrologic process. Therefore, by assigning a specific value of shortage index as design criteria for all reservoirs in different water resources systems would result different shortages. This study investigated the characteristics of shortages, which included rates, durations and occurrence probabilities, from different design scenarios of required reservoir capacities and project demands. Three streams located in the northern, central and southern Taiwan respectively were chosen as case study to demonstrate the influences of different hydrologic processes. The historical flow series of the three streams were proportionally standardized first to make the annual average flow to be the same. This study also suggested the concept of “tolerable shortages” as design criteria of reservoir capacity. Required reservoir capacity of each stream was computed to supply the project demand by satisfying the tolerable shortages. Appropriate tolerable shortage for designing the capacity of a reservoir can be determined based on the shortage impact to all purposes. Each stream should adopt its own shortage index to reach the corresponded tolerable shortage. This study proved that the shortage index, expected annual shortage rate and expected deficit percent-day had the same tendency to increase. The relationship is nearly one to one. This study developed the relationships between the shortage index, expected annual shortage rate and expected deficit percent-day with regression analysis. Therefore, one may choose either the expected deficit percent-day or expected annual shortage rate to demonstrate the shortage index. Further research may be focused on developing the relationship between tolerable shortage and expected annual shortage rate or expected deficit percent-day. An appropriate shortage index can be determined for designing reservoir capacity from the expected annual shortage rate or expected deficit percent-day.