Dynamic Control Domain Of Limited Water Level And Risk Analysis Based On Flood Classification

Flood disaster is happening with increasing frequency, which causes further losses, and the issues of water resource deficiency and water environment deterioration are becoming prominent increasingly. With scientific advance, the improvement of flood resource utilization under the safe flood control by taking full advantage of hydraulic engineering is an important approach of mitigating these water issues. Reservoir is important structural measure of flood and benefit control, and reservoir operation is nonstructural measure of flood and benefit control implementation. Lots of flood resource could be lost with static control of limited water level in traditional operation. To take full advantage of the storage capability of hydraulic engineering and utilize flood resource safely and reasonably, the dynamic control domain of limited water level is being set up to implement dynamic control of limited water level with forecasting information of meteorology and hydrology in our country. It is proved that dynamic control of limited water level is one of the key techniques of solving or mitigating water source deficiency.Most studies of the dynamic control of limited level are focused on how to confirm a domain of the limited water level, seldom considering of making the best of the storm and flood characteristics. In addition, scheme selection and risk analysis methods are relatively weak even in such study. Arming at the existing problems, a flood classification method was proposed in this study and different domains of limited water level for different kinds of floods were calculated based on the previous works. Secondly, the application method of dynamic control domains of limited water level based on flood classification and its risk and benefits were studied. At last, a multiple objective decision-making model was used to comparison of the schemes of the dynamic control of limited water level. The study contents and results are as follows:(1) Based on the storm flood similarity of the classified floods, according to the features in different stages of the storm flood evolutionary process, the variable fuzzy pattern recognition method was used to select classes of typical floods in this study, taking the Dahuofang and Danjiangkou reservoirs as examples to study the classing feasibility and selecting the corresponding typical flood, the result shows that using the method to classify the floods is feasible, besides the selected typical floods are the same with that using the experience rules, so it proved the reasonable and dependability of the selecting principles from the quantitative analysis. The flood classifying results can provide the foundation for designing different dynamic control domain of limited water level in flood seasons.(2) Based on the classification results of storm floods and features of reservoirs, the proper method of determining dynamic control domain of limited water level is proposed to set the range of limited water level in real-time control. According to the current design conception, the application methods of dynamic control domains of limited water level based on flood classification are stutied, the model and its process are given respectively. Results show that the upper limits of dynamic control domain of limited water level are different based on different kinds of floods, which can improve the utilization rate of flood resources.(3) The risks of missing report of rainfall forecasts are calculated based on bayas theorem, and the risk ratio of dynamic control of limited water level is calculated based on the results of risk analysis of dynamic control of limited water level. Take Dahuofang and Danjiangkou reservoirs as exemplas, the results show that the risk ratios are increased than which doesn’t classified, but they are lower than original designs.(4) Based on the calculation method of risk ratio of dynamic control of limited water level, analyzed the distribution pattern of the flood forecast errors; calculated the flood forecast error range; and calculated the risks of dynamic control of flood limited water level within different flood forecast error ranges using flood control forecast operation mode. Take Dahuofang reservoir as an example. The results show that the risk ratio is increased than which doesn’t classified, but they are lower than original designs. So, it’s safety.(5) Against the multi-objective decision contradiction caused by the dynamic control of reservoir limited water level, utilizes consultation countermeasures multi-objective decision-making model, confirms the weight of each indicator by binary contrast method, entropy method and combination weighting method based on Game theory, to establish the plan optimization evaluation model for the dynamic control of reservoir limited water level. Take Dahuofang Reservoir as an exemple, and tried to seek the optimal decisions and results. The results show that the decision values determined by the model were accordant with the practical situation.At last, the conclusions are draw and the problems need to be further studied are discussed.