[1] This paper introduces an early warning model for floods for real-time reservoir operation during typhoons. The model contains three main parts: (a) "flood watch" to monitor the current flood situation, (b) "flood release" to estimate probable reservoir releases in the near future, and (c) "decision analysis" to determine an appropriate release policy. A new flood alert index is defined to account for risk and is used to augment reservoir flood operation. In this study, a genetic algorithm-based rule is presented to specify suitable releases in response to the nature of flood inflows and reservoir water level. To demonstrate the procedure, the proposed method was successfully used in a real reservoir in northern Taiwan. The implementation of such a model proves that it can assist reservoir operators in taking timely action during typhoon events. Statement of the Problem[2] Typhoons greatly influence people's lives in Taiwan and generally occur from July through October. The occurrence of major typhoon-borne floods and consequent damage is highly erratic. Implementing effective measures to prevent damage caused by typhoons represents an important task for the government. In addition, the rainfall brought by typhoons is substantial. A typical typhoon is associated with 200-600 mm of rainfall on the island, which is roughly equal to 8%-24% of the annual average rainfall. Accordingly, when a drought lasts for a prolonged period, typhoons with accompanying rainfall are expected to ease the situation.[3] Flooding is a challenge to reservoir operators. When a storm alert is announced, operators need to make a serious decision regarding how much flood water should be released from the reservoir. The operators have to protect both the reservoir itself and the downstream reach and must preserve sufficient water for future use. Obviously, a reservoir built in a typhoon-prone area should be considered for its potential to alleviate some of the typhoon-related problems. The HEC-5 model is considered to be the best documented tool for reservoir simulation of flood and conservation systems [Yeh, 1985]. A significant quantity of work has been published regarding flood operation on reservoirs [Hydrologic Engineering Center, 1976;Sigvaldason, 1976;Yazicigil et al., 1983;Unver and Mays, 1990;Marien et al., 1994;Chang and Chen, 1998;Karbowski et al., 2005;Chang and Chang, 2006;Hsu and Wei, 2007]. However, little literature is available on an early warning model for floods during typhoon attacks, especially considering decision making under uncertainty.[4] Making decisions is not an easy task when there is uncertainty. The timing for appropriate releases of flood waters is imperative. To take into account the variability of incoming inflows, reservoir operators have difficulty operating reservoir releases properly, particularly at the onset of the typhoon season. Decisions regarding when and how to release water are crucial for real-time flood operation because a prolonged mismatch between the required and the actual reservoir s...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.