Summary: | 碩士 === 逢甲大學 === 水利工程與資源保育學系 === 102 === Taiwan due to the steep terrain, rivers short, resulting in uneven distribution of rainfall, while the construction of reservoir storage is one of the ways to ensure a stable water supply. Taiwan&;#39;s forty several reservoirs, where around seventy percent of the reservoir system as the general public drinking water sources by EPA data show that currently there are more than 40 percent of the reservoir throughout the year, half the time in eutrophic state, or are in accelerating eutrophication, therefore, an urgent need for a more active measures against eutrophication of reservoirs and to strengthen prevention, and improve water quality have been eutrophication of the reservoir. Currently the majority adopts Carlson eutrophication indicators, but according to a lot of literature that, Taiwan due to uneven distribution of rainfall in time and space, heavy rainfall, could easily lead to excessive water turbidity thus affecting Carlson eutrophication indicators accuracy. Therefore, review of the Carlson eutrophication index, and amendments to eutrophication indicators suitable for reservoirs in Taiwan is more important. In this study, the use of Shimen Reservoir water quality data to analyze and compare. To investigate whether environmental factors may interfere with Carlson eutrophication index, And the use of back propagation neural network and determine the scope of amendments to eutrophication, made applicable to the reservoir eutrophication index in Taiwan. According to the results show that the determine the scope of amendments to eutrophication through back-propagation neural network and corrected, will enhance subject to interference reservoir eutrophication indicators recognition ability.
|