The intelligent operation of pumping station for flood mitigation
Journal
Taiwan Water Conservancy
Journal Volume
59
Journal Issue
2
Start Page
1
End Page
9
ISSN
0492-1550
Date Issued
2011-01
Author(s)
Abstract
The pumping stations are the major hydraulic facilities for elimination of flood in highly developed cities and therefore play an important role in flood mitigation in Taipei and other metropolitan area. Nevertheless, the time of concentration in such watersheds is quite short and usually resulted in great damage due to the nonfunctioning pumping stations caused by flash flood. During typhoon periods, current operation of pumping stations adopts the designed operating rules and it is necessary to further investigate the operating rules to decrease the operating risk of pumping stations. In view of the characteristics of artificial neural networks, an ANN model was constructed herein by integrating the operational strategies and historical records. In this research we analyzed the correlations of the number of functioning pumps with water level, precipitation, and gravitational gates, respectively, in order to find outappropriate operating mechanisms for pumping stations during torrential storm events. The counterpropagatiom fuzzy neural network (CFNN) was then adopted to construct intelligent flood control models for pumping stations by using predictive water level information. Besides, the accuracy and stability of the models was further evaluated through selected criteria. The model was expected to reduce the risk of flooding and increase the operating efficiency of pumping stations via different combinations of pumps.
Type
journal article